id
int64 393k
2.82B
| repo
stringclasses 68
values | title
stringlengths 1
936
| body
stringlengths 0
256k
⌀ | labels
stringlengths 2
508
| priority
stringclasses 3
values | severity
stringclasses 3
values |
|---|---|---|---|---|---|---|
592,976,950 |
godot
|
Path2D shouldn't be editable with tool other than select
|
**Godot version:**
3.2.1
**Issue description:**
I was really surprised when I tried to measure something with Ruler Tool and it added a point to my Path2D. I was slightly less surprised (but still) when I tried to move my Path2D with Move Tool and it also added a point.
Polygon2D doesn't do this. When you use Move Tool, it actually moves instead of adding unexpected points. You can only place points with Select Tool. Path2D should work like that too.
|
enhancement,topic:editor,usability
|
low
|
Minor
|
593,003,290 |
flutter
|
Nested packages won't work
|
## Steps to Reproduce
The idea here is to create a package that will evolve along with one or more applications that use it. The logical course of action is to create a nested package inside an application so we can modify and debug the package while the application evolves, also, keeping the dependencies of this package apart from the application's dependencies, even share the package along different teams as a git submodule.
1. Create a new Flutter project with flutter create...
2. Inside the application's lib folder, create a new Flutter package with flutter create --template=package
3. Add the package as path dependency in the application's pubspe.yaml
4. Change the application sample code to use the Calculator sample code created in the package to add +1 to the counter state of the application sample.
## Expected behavior:
✅ Be able to set a breakpoint inside the package's Calculator class and make VSCode pause on it when the (+) button is hit. Works fine.
🟥 Add a `print("Hello, console");` to the Calculator.addOne method (i.e.: change the code of the package) and see that change in hot reload. Doesn't work.
IMPORTANT:
I know it is possible to use the package if we import as a folder of the root project (i.e.: instead of `import 'package:pkgname/pkgfile.dart'`, use `import 'package:appname/pkgfolder/lib/pkgfile.dart'`. But this solution will use the root pubspec.yaml, not the package's one (so we don't have a separated list of dependencies for the library we're building).
NOTICE: saving code inside the package folder indeed triggers a hot reload, but it says "Reloaded 0 of 481 libraries" (i.e.: it doesn't reload the code, just knows it changed).
Related: https://github.com/flutter/flutter/issues/49028
## Logs
```
(no related log generated)
```
```
[√] Flutter (Channel stable, v1.12.13+hotfix.9, on Microsoft Windows [Version 10.0.18362.720], locale pt-BR)
• Flutter version 1.12.13+hotfix.9 at C:\Flutter
• Framework revision f139b11009 (3 days ago), 2020-03-30 13:57:30 -0700
• Engine revision af51afceb8
• Dart version 2.7.2
(non relevant parts ommited)
```
|
tool,c: proposal,P3,team-tool,triaged-tool
|
low
|
Critical
|
593,007,920 |
godot
|
Custom compile of godot engine doesn't work with GDNative C++
|
**Godot version:**
3.2.1-stable.custom
**OS/device including version:**
Windows 10.0.18362
Visual Studio 2019
**Issue description:**
If I compile 3.2.1 godot engine, then run a game that uses a C++ GDNativeScript dll, the dll is sometimes (not always) receiving garbage parameters. But the same dll works fine with 3.2.1-stable-official, downloaded from godotengine.org. Both the custom build and the official build produce identical api.json
**Steps to reproduce:**
Compile Godot Editor:
download the 3.2.1-stable source from https://github.com/godotengine/godot/releases
extract, cd in to the directory
scons -j6 target=release_debug platform=windows
cd bin
godot.windows.opt.tools.64.exe --gdnative-generate-json-api api_custom.json
Compile a C++ DLL:
git clone --recursive https://github.com/GodotNativeTools/godot-cpp
cd godot-cpp
scons -j6 platform=windows generate_bindings=yes bits=64 use_custom_api_file=yes custom_api_file=api_custom.json target=release
cd .. (where SConstruct for the dll project resides)
scons target=release
Set it up so that gdscript is calling the dll with some functions and parameters. Press play in the custom compiled godot editor. The dll runs, is able to print things, but receives garbage parameters.
I want to stress that I very carefully made sure the api.json I gave to my custom compiled project matched the api.json produced by https://downloads.tuxfamily.org/godotengine/3.2.1/Godot_v3.2.1-stable_win64.exe.zip and that api.json matched the one output by my compiled godot editor.
**Minimal reproduction project:**
After two days on this morning to night, I've given up. I'm sorry, but I'm just not up to spending another minute on this. But I am curious if there are any easy answers. I can't for the life of me figure out why the stable-official version works and my custom compiled version does not. If anyone could tell me what was magically done different in compiling the stable-official binary, I'd greatly appreciate it. The stable-official binary is 64089328 bytes. The one I compiled, that is supposed to be identical, is 56765440 bytes.
|
discussion,topic:gdextension
|
low
|
Critical
|
593,070,904 |
rust
|
Infinite recursion is not catched by the compiler.
|
Hi, consider this code:
```rust
#![deny(unconditional_recursion)]
struct Object {
}
impl Object {
// RENAME HERE:
fn foo(&self) -> f32 {
7.0
}
}
impl ToObject for Object {
fn to_object(&self) -> &Object {
self
}
}
trait ToObject {
fn to_object(&self) -> &Object;
}
trait ObjectOps : ToObject {
fn foo(&self) -> f32 {
self.to_object().foo()
}
}
impl<T:ToObject> ObjectOps for T {}
fn main() {
let x = Object{};
println!("{}",x.foo());
}
```
It works fine. Let's rename `foo` to `foo2` next to the comment "RENAME HERE". What happens is that this code has infinite recursion now and `rustc` doesn't report any error nor warning despite the usage of `#![deny(unconditional_recursion)]`.
This is a serious problem, as a simple refactoring can cause something like that and it is very, very hard to debug, especially on WASM platform, where infinite recursion can just return a random number (it is UB). This problem could be solvable if I was able to write somehow `MAGIC::Object::foo(self.to_object())` instead of `self.to_object().foo()` but in such a way, that the methods from `ObjectOps` would not be visible in `MAGIC::Object`. Currently, when refactoring such code we can get infinite recursion and spend hours debugging it :(
|
C-enhancement,A-lints,T-compiler
|
low
|
Critical
|
593,111,910 |
flutter
|
[Web] open link in a new tab on mouse wheel click
|
On a flutter web page, center clicking on links does not open it in a new tab - instead it wants to scroll up and down.
Is this standard behavior? Will this change to feel like a normal web page?
|
c: new feature,framework,a: fidelity,f: gestures,platform-web,P3,team-web,triaged-web
|
low
|
Minor
|
593,118,302 |
pytorch
|
MKLDNN_conv2d 2X slower than the native TH implementation
|
## 🐛 Bug
It seems the mkldnn_convolution runs 2X slower than native thnn_conv2d_forward(use MKL for gemm).
<!-- A clear and concise description of what the bug is. -->
## To Reproduce
```
import torch
import torch.nn.functional as F
import numpy as np
from timeit import Timer
num = 50
N = 1
C = 512
H = 4
W = 4
M = 512
kernel_h = 3
kernel_w = 3
stride_h = 1
stride_w = 1
padding_h = 1
padding_w = 1
X_np = np.random.randn(N, C, H, W).astype(np.float32)
W_np = np.random.randn(M, C, kernel_h, kernel_w).astype(np.float32)
X = torch.from_numpy(X_np)
conv2d_pt = torch.nn.Conv2d(
C, M, (kernel_h, kernel_w), stride=(stride_h, stride_w),
padding=(padding_h, padding_w), groups=1, bias=False)
class ConvNet(torch.nn.Module):
def __init__(self):
super(ConvNet, self).__init__()
self.conv2d = conv2d_pt
def forward(self, x):
return self.conv2d(x)
model = ConvNet()
def pt_forward():
with torch.autograd.profiler.profile(record_shapes=True) as prof:
model(X)
print(prof.key_averages().table(sort_by="self_cpu_time_total"))
# torch._C._set_mkldnn_enabled(False)
t = Timer("pt_forward()", "from __main__ import pt_forward, X")
print("pt time = {}".format(t.timeit(num) / num * 1000.0))
```
Steps to reproduce the behavior:
1. You need to build Pytorch with USE_BLAS=ON, USE_MKL=ON, USE_MKLDNN=ON
2. Then you can torch._C._set_mkldnn_enabled(False) to switch between MKLDNN convolution and native convolution.
<!-- If you have a code sample, error messages, stack traces, please provide it here as well -->
## Expected behavior
```
---------------------- --------------- --------------- --------------- --------------- --------------- ---------------
Name Self CPU total % Self CPU total CPU total % CPU total CPU time avg Number of Calls
---------------------- --------------- --------------- --------------- --------------- --------------- ---------------
mkldnn_convolution 99.26% 3.037ms 99.26% 3.037ms 3.037ms 1
_convolution 0.46% 14.153us 99.77% 3.053ms 3.053ms 1
conv2d 0.13% 3.880us 100.00% 3.060ms 3.060ms 1
convolution 0.10% 3.083us 99.87% 3.056ms 3.056ms 1
contiguous 0.05% 1.403us 0.05% 1.403us 0.701us 2
---------------------- --------------- --------------- --------------- --------------- --------------- ---------------
Self CPU time total: 3.060ms
```
```
------------------------ --------------- --------------- --------------- --------------- --------------- ---------------
Name Self CPU total % Self CPU total CPU total % CPU total CPU time avg Number of Calls
------------------------ --------------- --------------- --------------- --------------- --------------- ---------------
thnn_conv2d_forward 98.52% 1.619ms 98.52% 1.619ms 1.619ms 1
_convolution 0.57% 9.306us 99.63% 1.638ms 1.638ms 1
_convolution_nogroup 0.27% 4.363us 99.00% 1.627ms 1.627ms 1
conv2d 0.22% 3.551us 100.00% 1.644ms 1.644ms 1
thnn_conv2d 0.19% 3.154us 98.71% 1.622ms 1.622ms 1
convolution 0.16% 2.559us 99.78% 1.640ms 1.640ms 1
contiguous 0.07% 1.087us 0.07% 1.087us 1.087us 1
_nnpack_available 0.02% 0.366us 0.02% 0.366us 0.366us 1
------------------------ --------------- --------------- --------------- --------------- --------------- ---------------
Self CPU time total: 1.644ms
```
<!-- A clear and concise description of what you expected to happen. -->
## Environment
Please copy and paste the output from our
[environment collection script](https://raw.githubusercontent.com/pytorch/pytorch/master/torch/utils/collect_env.py)
(or fill out the checklist below manually).
You can get the script and run it with:
```
wget https://raw.githubusercontent.com/pytorch/pytorch/master/torch/utils/collect_env.py
# For security purposes, please check the contents of collect_env.py before running it.
python collect_env.py
```
- PyTorch Version (e.g., 1.0):
- OS (e.g., Linux):
- How you installed PyTorch (`conda`, `pip`, source):
- Build command you used (if compiling from source):
- Python version:
- CUDA/cuDNN version:
- GPU models and configuration:
- Any other relevant information:
## Additional context
<!-- Add any other context about the problem here. -->
cc @VitalyFedyunin @gujinghui @PenghuiCheng @XiaobingSuper @jianyuh @ngimel
|
module: performance,module: cpu,triaged,module: mkldnn
|
low
|
Critical
|
593,145,689 |
react-native
|
xcode 11.4 build fatal error: module map file xxx/Build/Products/Debug-iphoneos/YogaKit/YogaKit.modulemap' not found
|
Please provide all the information requested. Issues that do not follow this format are likely to stall.
## Description
fatal error: module map file '/Users/miaohao/Library/Developer/Xcode/DerivedData/apex_baojia2-cdczyhhwbgshmtbdymitajpzamao/Build/Products/Debug-iphoneos/YogaKit/YogaKit.modulemap' not found
## React Native version:
Run `react-native info` in your terminal and copy the results here.
System:
OS: macOS 10.15.3
CPU: (4) x64 Intel(R) Core(TM) i5-7360U CPU @ 2.30GHz
Memory: 219.52 MB / 8.00 GB
Shell: 5.7.1 - /bin/zsh
Binaries:
Node: 12.16.1 - /usr/local/bin/node
Yarn: 1.22.0 - /usr/local/bin/yarn
npm: 6.13.4 - /usr/local/bin/npm
Watchman: 4.9.0 - /usr/local/bin/watchman
Managers:
CocoaPods: 1.8.4 - /usr/local/bin/pod
SDKs:
iOS SDK:
Platforms: iOS 13.4, DriverKit 19.0, macOS 10.15, tvOS 13.4, watchOS 6.2
Android SDK:
API Levels: 23, 24, 25, 26, 27, 28, 29
Build Tools: 26.0.2, 28.0.3
System Images: android-23 | Android TV Intel x86 Atom, android-23 | Google APIs Intel x86 Atom, android-28 | Google APIs Intel x86 Atom, android-29 | Google Play Intel x86 Atom
Android NDK: Not Found
IDEs:
Android Studio: 3.6 AI-192.7142.36.36.6241897
Xcode: 11.4/11E146 - /usr/bin/xcodebuild
Languages:
Python: 2.7.16 - /usr/local/bin/python
npmPackages:
@react-native-community/cli: Not Found
react: 16.11.0 => 16.11.0
react-native: 0.62.0 => 0.62.0
npmGlobalPackages:
*react-native*: Not Found
## Steps To Reproduce
Provide a detailed list of steps that reproduce the issue.
1.react-native init XXX
2.cd ios && pod install
3. xcode Product/build
## Expected Results
Describe what you expected to happen.
## Snack, code example, screenshot, or link to a repository:
Please provide a Snack (https://snack.expo.io/), a link to a repository on GitHub, or provide a minimal code example that reproduces the problem.
You may provide a screenshot of the application if you think it is relevant to your bug report.
Here are some tips for providing a minimal example: https://stackoverflow.com/help/mcve
|
Platform: iOS,Impact: Crash,Tool: Xcode
|
high
|
Critical
|
593,245,624 |
pytorch
|
Build PyTorch-1.4.0 from source failed
|
## 🐛 Bug
I am trying to build pytorch 1.4.0 from source and i am seeing this error.
```
gmake: *** [all] Error 2
Traceback (most recent call last):
File "setup.py", line 755, in <module>
build_deps()
File "setup.py", line 316, in build_deps
cmake=cmake)
File "/local/home/yingleiz/workspace/PyTorchWS/build/PyTorch/PyTorch-1.4.x/AL2012/DEV.STD.PTHREAD/build/private/src/tools/build_pytorch_libs.py", line 62, in build_caffe2
cmake.build(my_env)
File "/local/home/yingleiz/workspace/PyTorchWS/build/PyTorch/PyTorch-1.4.x/AL2012/DEV.STD.PTHREAD/build/private/src/tools/setup_helpers/cmake.py", line 337, in build
self.run(build_args, my_env)
File "/local/home/yingleiz/workspace/PyTorchWS/build/PyTorch/PyTorch-1.4.x/AL2012/DEV.STD.PTHREAD/build/private/src/tools/setup_helpers/cmake.py", line 141, in run
check_call(command, cwd=self.build_dir, env=env)
File "/home/yingleiz/brazil-pkg-cache/packages/CPython36Runtime/CPython36Runtime-release.316387.0/AL2012/DEV.STD.PTHREAD/build/python3.6/lib/python3.6/subprocess.py", line 311, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['cmake', '--build', '.', '--target', 'install', '--config', 'Release', '--', '-j', '36']' returned non-zero exit status 2.
```
Here is the full log of this build: https://gist.github.com/YingleiZhang/fc98c94d493ac39506daacefa0298b38
Can someone provide some guidance on this issue? I don't have problem build Pytorch 1.2.0 with the same env.
cc @ezyang @yf225 @peterjc123
|
module: build,triaged
|
low
|
Critical
|
593,260,376 |
pytorch
|
Copy activations from one parts to another part in tensor, but report error
|
## 🐛 Bug
I want to copy the activations from one region in image (tensor, shape=(3, 800, 800)) to another region with same shape, and after dealing with parts of dataset, a bug will be reported.
ValueError: could not broadcast input array from shape (3,97,39) into shape (3,97,40)
Parts of my code is :
tmp = image[:, old_y0:old_y1+1, old_x0:old_x1+1]
image[:, old_y0+offset_y:old_y1+offset_y+1, old_x0+offset_x:old_x1+offset_x+1] = tmp
What I confused is that the location interval is exactly the same between two parts, but a bug was reported.
Version:
Pytorch 1.4
|
triaged
|
low
|
Critical
|
593,278,424 |
PowerToys
|
[Run][WindowWalker plugin] Allow spaces to be treated as implicit AND operators
|
# Summary of the new feature/enhancement
I usually know the application I'm searching for. If I have 20 or 30 open powershell consoles it would be convenient if I could type "power foo" and have the search results be the intersection of searching for "power" AND "foo" rather than the current mechanism of searching for the literal string "power foo".
# Proposed technical implementation details (optional)
It seems likely this is a straightforward change though the default of treating spaces as AND (vs OR) operators may be controversial. To me this seems like a sensible policy short of implementing a full logical expression parser though perhaps it could be a user-option?
|
Idea-Enhancement,Product-PowerToys Run,Priority-2,Run-Plugin
|
low
|
Major
|
593,297,583 |
react-native
|
ScrollView doesn't scroll to focused element
|
## Environment
Android API: 28
Device: AndroidTV emulator and Xiaomi MI BOX
React Native: 0.61.5
## Description
Sometimes ScrollView doesn't scroll to currently focused element on AndroidTV devices. Looks like issue is triggered because of two scroll events, one from _ReactScrollView.java_ _arrowScroll_ method and one from _requestChildFocus_. Issue seems to occur only when scrolling on the bottom of _ScrollView_.
## React Native version:
"0.61.5"
## Steps To Reproduce
1. Navigate with keyboard/remote controller up and down on between last rows of ScrollView.
## Expected Results
Scroll follows focused element
## Snack, code example, screenshot, or link to a repository:
App.ts example:
```
import React from 'react';
import {View, StyleProp, TextStyle, ViewStyle, TouchableOpacity, ScrollView, Text} from 'react-native';
const styles: {[name: string]: StyleProp<ViewStyle> | StyleProp<TextStyle>} = {
container: {
position: 'absolute',
width: '100%',
height: '100%',
backgroundColor: '#000',
justifyContent: 'flex-end',
alignContent: 'center'
}
};
const rows = [1,2,3,4,5,6,7,8,9];
const columns = ['a', 'b', 'c', 'd', 'e', 'f'];
type Props = {
text: string;
number: number;
}
const MyButton = (props: Props) => {
return (
<TouchableOpacity
style={{width: 200, height: 200, margin: 50, backgroundColor: `rgba(100,255,0,0.3)`}}
key={props.text + props.number}
onPress={() => {}}
>
<Text>{props.text + props.number}</Text>
</TouchableOpacity>
);
};
const App = () => {
return (
<View style={styles.container}>
<ScrollView>
{rows.map((number) => (
<ScrollView
horizontal={true}
key={'scrollview' + number}
>
{columns.map((text) => <MyButton text={text} number={number} key={text + number} />)}
</ScrollView>
))
}
</ScrollView>
</View>
);
};
export default App;
```
In gif below i navigated up and down only:
|
Help Wanted :octocat:,Platform: Android,Component: ScrollView,Newer Patch Available,Needs: Attention
|
low
|
Critical
|
593,312,022 |
youtube-dl
|
please add for https://aikidojournal.vhx.tv/
|
<!--
######################################################################
WARNING!
IGNORING THE FOLLOWING TEMPLATE WILL RESULT IN ISSUE CLOSED AS INCOMPLETE
######################################################################
-->
## Checklist
<!--
Carefully read and work through this check list in order to prevent the most common mistakes and misuse of youtube-dl:
- First of, make sure you are using the latest version of youtube-dl. Run `youtube-dl --version` and ensure your version is 2020.03.24. If it's not, see https://yt-dl.org/update on how to update. Issues with outdated version will be REJECTED.
- Make sure that all provided video/audio/playlist URLs (if any) are alive and playable in a browser.
- Make sure that site you are requesting is not dedicated to copyright infringement, see https://yt-dl.org/copyright-infringement. youtube-dl does not support such sites. In order for site support request to be accepted all provided example URLs should not violate any copyrights.
- Search the bugtracker for similar site support requests: http://yt-dl.org/search-issues. DO NOT post duplicates.
- Finally, put x into all relevant boxes (like this [x])
-->
- [ x] I'm reporting a new site support request
- [x ] I've verified that I'm running youtube-dl version **2020.03.24**
- [x ] I've checked that all provided URLs are alive and playable in a browser
- [ x] I've checked that none of provided URLs violate any copyrights
- [ x] I've searched the bugtracker for similar site support requests including closed ones
## Example URLs
<!--
Provide all kinds of example URLs support for which should be included. Replace following example URLs by yours.
-->
https://aikidojournal.vhx.tv/videos/morihei-ueshiba-divine-techniques-os05
## Description
<!--
Provide any additional information.
If work on your issue requires account credentials please provide them or explain how one can obtain them.
-->
WRITE DESCRIPTION HERE
This video is available only for one weeks it is for educational purpose
|
site-support-request
|
low
|
Critical
|
593,325,440 |
TypeScript
|
Exclude<keyof T, 'x'> should be assignable to Exclude<Exclude<keyof T, 'x'>, 'x'>
|
<!--
Please try to reproduce the issue with the latest published version. It may have already been fixed.
For npm: `typescript@next`
This is also the 'Nightly' version in the playground: http://www.typescriptlang.org/play/?ts=Nightly
-->
**TypeScript Version:** 3.8.3
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:**
* type Exclude keyof is not assignable to type Exclude Exclude keyof
* exclude idempotent
**Code**
```ts
const f = <T>(arg: Omit<T, 'field'>) => {
const a = { ...arg, field: 'value' };
const { field, ...b } = a;
const c: Omit<T, 'field'> = a;
const d: Omit<T, 'field'> = b;
return { c, d };
};
```
**Expected behavior:**
Code compiles.
**Actual behavior:**
Compilation error on `d` initialization (but not on `c` initialization):
```
tmp.ts:5:11 - error TS2322: Type 'Pick<Pick<T, Exclude<keyof T, "field">> & { field: string; }, Exclude<Exclude<keyof T, "field">, "field">>' is not assignable to type 'Pick<T, Exclude<keyof T, "field">>'.
Type 'Exclude<keyof T, "field">' is not assignable to type 'Exclude<Exclude<keyof T, "field">, "field">'.
Type 'keyof T' is not assignable to type 'Exclude<keyof T, "field">'.
5 const d: Omit<T, 'field'> = b;
~
Found 1 error.
```
**Playground Link:** [Playground Link](https://www.typescriptlang.org/play/?ssl=8&ssc=1&pln=1&pc=1#code/MYewdgzgLgBAZjAvDAPAFQHwAoCGAnAcwC4YB5AWwEsp0AaGAcjkoFMAbAEwYwEokMYAbwBQMMTFCRYOJEJgA6RfgL1m7DiQYA3HGwCuLBjAC+AblHjJ0OWs71F8gEYnZOc+InhrwEhWp1GWy4BZDcLMStYDTIqGjR6JlZObllHd3E8Fig9PDA5YHoOE3MzYSA)
**Related Issues:** I did not find other related issues.
|
Bug
|
low
|
Critical
|
593,345,042 |
terminal
|
Feature request: Parenthesis matching in text
|
Wonderful work. This might finally persuade some of us Mac diehards to move to Windows!
**Request**: Create a new feature to allow the user to invoke parenthesis matching in text by double clicking on a parenthesis, anywhere in the terminal. If a match exists, the text from the clicked parenthesis until the matching parenthesis gets selected. If no match exists, the parenthesis that the user clicked on, is shown in red, or the terminal blinks. This is similar to xemacs' behavior, when initialized with the correct options.
|
Area-TerminalControl,Area-Extensibility,Product-Terminal,Issue-Task
|
low
|
Minor
|
593,372,195 |
vscode
|
vscode://file/ URI not working from chromium
|
<!-- ⚠️⚠️ Do Not Delete This! bug_report_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: https://code.visualstudio.com/insiders/ -->
<!-- Use Help > Report Issue to prefill these. -->
- VSCode Version: 1.43.2
- OS Version: Ubuntu 18.04
- Chromium version: 80.0.3987.163 (installed from snap in Ubuntu Software)
Steps to Reproduce:
1. Click on vscode://file/xxx link in chrome
<!-- Launch with `code --disable-extensions` to check. -->
Does this issue occur when all extensions are disabled?: Yes
|
bug,upstream,linux,snap
|
low
|
Critical
|
593,405,287 |
flutter
|
EventChannel.setStreamHandler(null) should cancel previous Stream Handler
|
## Use case
Plugin may publish continuous values using an EventChannel.
As plugins go through lifecycle, they should clean up event channels in the same way that method call handlers are cleaned up: By setting the handler on the channel to null.
EventChannel's do pose a little more complexity as they have a `onCancel` method which should help to clear up any used memory.
Given a V2 plugin, a setup & clean up method currently could look like this:
``` java
@Override
public void onAttachedToEngine(@NonNull FlutterPluginBinding binding) {
aEventChannel = new EventChannel(messenger, CHANNEL_NAME);
aEventChannelHandler = new StreamHandler() { ... };
aEventChannel.setStreamHandler(aEventChannelHandler);
}
@Override
public void onDetachedFromEngine(@NonNull FlutterPluginBinding binding) {
aEventChannelHandler.onCancel(null);
aEventChannel.setStreamHandler(null);
aEventChannelHandler = null;
}
```
## Proposal
When setting the StreamHandler to `null`, the framework should internally call `onCancel`, so that above method becomes:
``` java
@Override
public void onAttachedToEngine(@NonNull FlutterPluginBinding binding) {
aEventChannel = new EventChannel(messenger, CHANNEL_NAME);
aEventChannel.setStreamHandler(new StreamHandler() { ... });
}
@Override
public void onDetachedFromEngine(@NonNull FlutterPluginBinding binding) {
aEventChannel.setStreamHandler(null);
}
```
From a glimpse in the code, `DartMessenger.setMessageHandler()` should maybe be adjusted to read like this:
```
@Override
public void setMessageHandler(
@NonNull String channel, @Nullable BinaryMessenger.BinaryMessageHandler handler) {
if (handler == null) {
Log.v(TAG, "Removing handler for channel '" + channel + "'");
BinaryMessenger.BinaryMessageHandler previous = messageHandlers.remove(channel);
if (previous != null) {
previous.onCancel();
}
} else {
Log.v(TAG, "Setting handler for channel '" + channel + "'");
messageHandlers.put(channel, handler);
}
}
```
|
platform-android,engine,c: proposal,P3,team-android,triaged-android
|
low
|
Minor
|
593,413,018 |
flutter
|
[animations] [proposal] OpenContainer not inherit closed/openColor from app theme
|
<!-- Thank you for using Flutter!
If you are looking for support, please check out our documentation
or consider asking a question on Stack Overflow:
* https://flutter.dev/
* https://api.flutter.dev/
* https://stackoverflow.com/questions/tagged/flutter?sort=frequent
If you have found a bug or if our documentation doesn't have an answer
to what you're looking for, then fill our the template below. Please read
our guide to filing a bug first: https://flutter.dev/docs/resources/bug-reports
-->
## Use case
I'm using the amazing `animations` Flutter package and it works as expected.
The only one thing that I cannot understand is why we need to specified properties like `closedColor` or `openColor` that has a default value (`Colors.white` in this case).
This is my code snippet that use `OpenContainer` widget:
```dart
return OpenContainer(
closedColor: Theme.of(context).scaffoldBackgroundColor,
closedShape: const RoundedRectangleBorder(),
tappable: true,
transitionType: ContainerTransitionType.fade,
closedBuilder:
(BuildContext _, VoidCallback openContainer) {
return MyHomeCard(
title: _getCardTitle(tour));
},
openBuilder: (BuildContext context, VoidCallback _) {
final String _currentLocale =
AppLocalizations.of(context).locale.languageCode;
return TourPage(
tour: tour,
langCode: _currentLocale,
);
},
);
```
## Proposal
In my opinion in most of cases these two colors are comparable to `backgroundColor` widgets property.
It will be beautiful if, as other widgets (like ListView, Center, etc..), by default animations widgets automatically inherit the closed/openColor from the app theme.
Using this strategy we can avoid unnecessary code as `Theme.of(context).scaffoldBackgroundColor` due to te fact that colors are inherited by the default app theme.
I think that this behaviour can improve developers themes usage and it's more Flutter oriented.
Thank you
|
c: new feature,package,c: proposal,p: animations,team-ecosystem,P3,triaged-ecosystem
|
low
|
Critical
|
593,428,772 |
flutter
|
Talkback does not honor scopesRoute unless namesRoute and label are also set
|
When creating a Semantics widget with `scopesRoute` set to true, iOS VO will honor that and focus the top element when a new widget is inserted into the tree with that property set, but Android's TalkBack will not, unless namesRoute and label are also set.
The following code shows the problem:
```
import 'package:flutter/material.dart';
import 'package:flutter/rendering.dart';
void main() {
runApp(MyApp());
}
class MyApp extends StatelessWidget {
// This widget is the root of your application.
@override
Widget build(BuildContext context) {
return MaterialApp(
title: 'Flutter Demo',
theme: ThemeData(
primarySwatch: Colors.blue,
),
home: SemanticsHomePage(),
);
}
}
class SemanticsHomePage extends StatefulWidget {
const SemanticsHomePage();
_SemanticsHomePageState createState() => _SemanticsHomePageState();
}
class _SemanticsHomePageState extends State<SemanticsHomePage> {
Widget _tree1;
Widget _tree2;
Key _tree1Key = UniqueKey();
Key _tree2Key = UniqueKey();
int _currentTree = 0;
void _buildTrees() {
_tree1 = Semantics(
key: _tree1Key,
scopesRoute: true,
explicitChildNodes: true,
// uncomment these lines for Android!!
//namesRoute: true,
//label: "Tree 1",
child: Column(
mainAxisSize: MainAxisSize.min,
children: <Widget>[
Text("Test 1", style: TextStyle(fontSize: 50)),
Text("Test 2", style: TextStyle(fontSize: 50)),
Text("Test 3", style: TextStyle(fontSize: 50)),
FlatButton(
child: Text("Change", style: TextStyle(fontSize: 20)), onPressed: () {
setState(() {
_currentTree = (_currentTree + 1) % 2;
});
})
],
),
);
_tree2 = Semantics(
key: _tree2Key,
scopesRoute: true,
explicitChildNodes: true,
// uncomment these lines for Android!!
//namesRoute: true,
//label: "Tree 2",
child: Column(
mainAxisSize: MainAxisSize.min,
children: <Widget>[
FlatButton(
child: Text("Change", style: TextStyle(fontSize: 20)), onPressed: () {
setState(() {
_currentTree = (_currentTree + 1) % 2;
});
}),
Text("Test 4", style: TextStyle(fontSize: 50)),
Text("Test 5", style: TextStyle(fontSize: 50)),
Text("Test 6", style: TextStyle(fontSize: 50)),
Text("Test 7", style: TextStyle(fontSize: 50)),
Text("Test 8", style: TextStyle(fontSize: 50)),
],
),
);
}
@override
void initState() {
_buildTrees();
super.initState();
}
@override
void didUpdateWidget(SemanticsHomePage oldWidget) {
_buildTrees();
super.didUpdateWidget(oldWidget);
}
@override
Widget build(BuildContext context) {
return Scaffold(
body: Container(
child: Center(
child: _currentTree == 0 ? _tree1 : _tree2
),
)
);
}
}
```
Running this code on iOS works as expected, which is that the first element is focused and read after interacting with the "Change button", but on Android it does not. Uncommenting the lines in the code for Android makes it work, however it would be best to not have to set namesRoute and label, as that's not always something that makes sense in certain scenarios.
|
platform-android,framework,engine,a: accessibility,f: routes,customer: amplify,has reproducible steps,P2,found in release: 3.3,found in release: 3.6,team-android,triaged-android
|
low
|
Major
|
593,437,200 |
TypeScript
|
Add feature to re-declare variable
|
## Search Terms
redeclare variable, re-declare variable, augment variable, merge variable declaration, declaration override
## Suggestion
define a `redeclare` keyword which allows re-declaration of a typed variable.
e.g.
```typescript
redeclare var foo: {
prop: string
};
```
## Use Cases
<!--
What do you want to use this for?
What shortcomings exist with current approaches?
-->
When an external library declares a variable, e.g.
```typescript
declare var x: {
prop: string;
};
```
The `redeclare` keyword can be used somewhere else to override the declaration:
```typescript
redeclare var x: number;
```
or augment it:
```typescript
redeclare var x: typeof x & {
augmentedProp: number;
}
```
This would allow to type the pure JavaScript case of:
```javascript
// lib.js
var x = {
prop: "foo"
};
// app.js
x.augmentedProp = 10;
console.log(x.prop, x.augmentedProp);
```
## Examples
<!-- Show how this would be used and what the behavior would be -->
> ```typescript
> declare var URLSearchParams: {
> prototype: URLSearchParams;
> new(init?: string[][] | Record<string, string> | string | URLSearchParams): URLSearchParams;
> toString(): string;
> };
> ```
> defined in [`lib.dom.d.ts`](https://github.com/microsoft/TypeScript/blob/712967b2780e8ecd28f8f1e2e89c1ebd2592bb4c/lib/lib.dom.d.ts#L16156)
to extend the variable with additional static methods to `URLSearchParams` is not possible currently (see [here](https://stackoverflow.com/questions/60856144/how-to-augment-a-global-variable-type-with-static-methods)).
The feature can be used in the following way:
```typescript
redeclare var URLSearchParams: typeof URLSearchParams & {
fromObject(obj: object): URLSearchParams;
};
if (!URLSearchParams.fromObject) {
URLSearchParams.fromObject = function(obj: object): URLSearchParams { /* implementation */ }
}
```
## Checklist
My suggestion meets these guidelines:
* [x] This wouldn't be a breaking change in existing TypeScript/JavaScript code
* [x] This wouldn't change the runtime behavior of existing JavaScript code
* [x] This could be implemented without emitting different JS based on the types of the expressions
* [x] This isn't a runtime feature (e.g. library functionality, non-ECMAScript syntax with JavaScript output, etc.)
* [x] This feature would agree with the rest of [TypeScript's Design Goals](https://github.com/Microsoft/TypeScript/wiki/TypeScript-Design-Goals).
|
Suggestion,Awaiting More Feedback
|
low
|
Major
|
593,456,958 |
TypeScript
|
Exposing canonical filename logic
|
## Search Terms
canonical filename
## Suggestion
I'd like to re-use the get-canonical-filename logic:
https://github.com/microsoft/TypeScript/blob/d68295e74e4897f588e73edf72089eb238904f02/src/compiler/core.ts#L1906-L1909
## Use Cases
I want to use `ts.resolveModuleName` and the accompanying `ts.createModuleResolutionCache` to resolve `ImportDeclaration`s (as per [this comment's suggestion](https://github.com/microsoft/TypeScript/issues/28276#issuecomment-435016356)).
`createModuleResolutionCache` requires a parameter `getCanonicalFileName`. I think it makes the most sense to use the same logic as the compiler itself here, which is the function I linked above.
I *could* just copy-and-paste this function, but that *feels* wrong and might lead to discrepancies in the future if [`toFileNameLowerCase`](https://github.com/microsoft/TypeScript/blob/d68295e74e4897f588e73edf72089eb238904f02/src/compiler/core.ts#L1407-L1445) (which is not exposed either) is ever updated.
## Examples
```ts
const cache = ts.createModuleResolutionCache(currentDirectory, ts.createGetCanonicalFileName(true/false));
const resolved = ts.resolveModuleName(name, file, compilerOptions, host, cache);
```
## Checklist
My suggestion meets these guidelines:
* [x] This wouldn't be a breaking change in existing TypeScript/JavaScript code
* [x] This wouldn't change the runtime behavior of existing JavaScript code
* [x] This could be implemented without emitting different JS based on the types of the expressions
* [x] This isn't a runtime feature (e.g. library functionality, non-ECMAScript syntax with JavaScript output, etc.)
* [x] This feature would agree with the rest of [TypeScript's Design Goals](https://github.com/Microsoft/TypeScript/wiki/TypeScript-Design-Goals).
|
API
|
low
|
Minor
|
593,464,679 |
PowerToys
|
Queue files automatically on copy/move to same target
|
Maybe I'm asking too much of PowerToys but I would really like to see a feature like this:
Make it possible to automatically queue files when copying different files to the same harddisk.
I just hate that Windows itself is just starting to copy everything simultaneously. This is ok if the targets are different devices. But if everything copied goes to the same target this causes performance issues.
I know there is tooling like "TeraCopy" out there which does exactly that. But I rather trust an open source tool managed by Microsoft than some tool from some guy.
In a perfect world this functionality would already be part of Windows 10, but my Feedback request didn't get any response yet.
|
Idea-New PowerToy
|
medium
|
Major
|
593,469,075 |
go
|
x/tools/gopls: show progress report when gopls is downloading a module
|
@ardan-bkennedy suggested that we should try to use the progress reporting functionality of LSP to indicate when a module is being downloaded by go/packages. Otherwise, we just have a slow save that don't give the user any indication that `gopls` is communicating with the network.
This will be tricky because only go/packages knows that it's downloading a module. My current best guess at how to do this is to put something in the config's Logf function...or maybe use a trick like packagesinternal?
|
FeatureRequest,gopls,Tools
|
low
|
Major
|
593,524,168 |
node
|
doc: Should `runMain()` and `globalPaths` of module add in doc?
|
By chance, I find `runMain()` method was used in [babel-node](https://github.com/babel/babel/blob/master/packages/babel-node/src/_babel-node.js#L207) but it can't be found in API.
I also checked issue and code, but nothing.
```bash
$ git grep runMain $(git rev-list --all doc/api/modules.md) -- doc/api/modules.md
# nothing
$ git grep globalPaths $(git rev-list --all doc/api/modules.md) -- doc/api/modules.md
# nothing
$ git grep builtinModules $(git rev-list --all doc/api/modules.md) -- doc/api/modules.md
534c204e223d85c44cd6b1b642f29143095077f6:doc/api/modules.md:### `module.builtinModules`
534c204e223d85c44cd6b1b642f29143095077f6:doc/api/modules.md:const builtin = require('module').builtinModules;
a220202a47ee87d1ef91fa8f65a3e048f298b7c1:doc/api/modules.md:### `module.builtinModules`
```
However [the `builtinModules` property near `globalPaths`](https://github.com/nodejs/node/blob/master/lib/internal/modules/cjs/loader.js#L176-L182) can be found. Should `runMain()` and `globalPaths` of module add to doc? If needed, I would be pleased to write them.
|
doc,module
|
low
|
Minor
|
593,531,757 |
TypeScript
|
isolatedModules doesn't respect enabled preserveConstEnum option what the project might be build with
|
## Search Terms
isolatedModules, preserveConstEnum, const enum
## Suggestion
The compiler should respect/understand that the package/module was compiled with enabled `preserveConstEnums` compiler option.
## Use Cases
Let's say we have a project written in TypeScript, which uses const enums.
The project uses them because the main `const enum`'s advantage is inlining values instead of runtime accessing to the value.
In other hand the maintainers understand that const enums compiles in nothing in JS code, and that means that consumers can't use a const enums in their JS code (in some cases in TS code either, see above). Thus, they decided to enable `preserveConstEnums` compiler option to leave all const enums in JS code and allow consumers use them as well.
Up to this moment everything is good. But the next who appears on the stage is `isolatedModules` (say hello to `create-react-app`, webpack's transpile-only mode, `transpileModule` compiler's API and so on). This compiler option is used to detect that imported module could be transpiler without compilation into JS code and everything will work well. One of checks for this option is that you don't import const enums (error `TS2748: Cannot access ambient const enums when the '--isolatedModules' flag is provided`).
To summarize, if the project/package is compiled with enabled `preserveConstEnums`, in `isolatedModules` mode the compiler shouldn't fail with error about unavailability of using const enums, because that's not actually true.
## Examples
Quite possible we need to provide that information to the compiler somehow (by providing tsconfig.export.json or in package.json for instance), but I don't even know how it would be done.
## Checklist
My suggestion meets these guidelines:
* [x] This wouldn't be a breaking change in existing TypeScript/JavaScript code
* [x] This wouldn't change the runtime behavior of existing JavaScript code
* [x] This could be implemented without emitting different JS based on the types of the expressions
* [x] This isn't a runtime feature (e.g. library functionality, non-ECMAScript syntax with JavaScript output, etc.)
* [x] This feature would agree with the rest of [TypeScript's Design Goals](https://github.com/Microsoft/TypeScript/wiki/TypeScript-Design-Goals).
|
Suggestion,Awaiting More Feedback,Fix Available
|
medium
|
Critical
|
593,555,230 |
TypeScript
|
Lib dependencies are transitive
|
<!-- 🚨 STOP 🚨 𝗦𝗧𝗢𝗣 🚨 𝑺𝑻𝑶𝑷 🚨
Half of all issues filed here are duplicates, answered in the FAQ, or not appropriate for the bug tracker. Even if you think you've found a *bug*, please read the FAQ first, especially the Common "Bugs" That Aren't Bugs section!
Please help us by doing the following steps before logging an issue:
* Search: https://github.com/Microsoft/TypeScript/search?type=Issues
* Read the FAQ: https://github.com/Microsoft/TypeScript/wiki/FAQ
Please fill in the *entire* template below.
-->
<!--
Please try to reproduce the issue with the latest published version. It may have already been fixed.
For npm: `typescript@next`
This is also the 'Nightly' version in the playground: http://www.typescriptlang.org/play/?ts=Nightly
-->
**TypeScript Version:** 3.8.3
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:** lib target dependencies triple-slash
**Code**
I have a reproduction repository at https://github.com/astorije/repro-tsc-lib-issue.
Essentially, consider the following code:
```ts
console.log(Object.values({ foo: BigInt(42) }));
```
`Object.values` is a ES2017 feature, and `BigInt` is a ES2020.
Also consider the `target` is `tsconfig.json` is `es2015`, and `lib` is either not specified or simply contains `es2015` and `dom`.
**Expected behavior:**
Given the code and the TS configuration above, the type checker should _always_ fails with:
```ts
index.ts:2:20 - error TS2339: Property 'values' does not exist on type 'ObjectConstructor'.
2 console.log(Object.values({ foo: BigInt(42) }));
~~~~~~
index.ts:2:34 - error TS2304: Cannot find name 'BigInt'.
2 console.log(Object.values({ foo: BigInt(42) }));
~~~~~~
```
And if you only have `typescript` as a dependency in `package.json`, that is indeed what you'll get. So far, so good. **See the `expected` folder of my repro repo.**
**Actual behavior:**
If you have a (direct or indirect) dependency to `@types/node`, or if _any_ dependency contains a declaration file that has `/// <reference lib="es2018" />`, `/// <reference lib="es2020.bigint" />`, etc., the type checker will now stop reporting these errors. **See the `actual` folder of my repro repo.**
This seems significant to me because it means that dependencies can silence potential browser incompatibilities. If I specified a `target` of `es2015`, with no extra `lib`, Babel transpiling, or polyfills, I should not be able to compile the code above.
Is there something obvious I'm missing? I realize it's the very design of `lib`but as explained in #15732, that assumes the project uses polyfills, which is not something that can be enforced/checked by a third-party library at compile time.
**Related Issues:** <!-- Did you find other bugs that looked similar? -->
Apologies for the vagueness of the search terms. It's significant enough that it must already exist in an issue or a StackOverflow question, but after spending some time looking, I did not find anything similar.
The only issues that seem somewhat related, but different:
- https://github.com/microsoft/TypeScript/issues/33111
- https://github.com/microsoft/TypeScript/issues/35656
- https://github.com/microsoft/TypeScript/issues/15732
|
Suggestion,In Discussion
|
low
|
Critical
|
593,566,749 |
flutter
|
ScrollController jumpTo fails assertion 'activity.isScrolling': is not true
|
## Steps to Reproduce
Code in words:
I start of with an CustomScrollView with an initial offset. If the user scrolls to the top from this initial offset, I will direct him to another screen. But if the user scrolls to the top from another offset, i want the ScrollController to jump back to the initial offset.
Code to reproduce:
```dart
import 'package:flutter/material.dart';
void main() {
runApp(MyApp());
}
class MyApp extends StatelessWidget {
// This widget is the root of your application.
@override
Widget build(BuildContext context) {
return MaterialApp(
title: 'Flutter Demo',
theme: ThemeData(
primarySwatch: Colors.blue,
visualDensity: VisualDensity.adaptivePlatformDensity,
),
home: TestScreen(),
);
}
}
class TestScreen extends StatefulWidget {
static double initialScrollOffset = 50;
_TestScreenState createState() => _TestScreenState();
}
class _TestScreenState extends State<TestScreen> {
ScrollController _controller = ScrollController(
initialScrollOffset: TestScreen.initialScrollOffset,
);
double scrollStartOffset;
@override
void initState() {
super.initState();
_controller.addListener(_scrollListener);
}
@override
void dispose() {
super.dispose();
_controller.dispose();
}
void _scrollListener() {
if (_controller.offset == 0) {
if (scrollStartOffset == TestScreen.initialScrollOffset) {
// move to the next screen
}
_controller.jumpTo(TestScreen.initialScrollOffset);
}
}
@override
Widget build(BuildContext context) {
return Scaffold(
appBar: AppBar(
centerTitle: true,
title: Text(
'TestScreen',
),
),
body: NotificationListener<ScrollNotification>(
onNotification: (ScrollNotification scrollNotification) {
if (scrollNotification is ScrollStartNotification) {
scrollStartOffset = scrollNotification.metrics.pixels;
}
return;
},
child: CustomScrollView(
controller: _controller,
slivers: <Widget>[
SliverToBoxAdapter(
child: Container(
height: TestScreen.initialScrollOffset,
color: Colors.yellow,
),
),
SliverToBoxAdapter(
child: Container(
height: 1000,
color: Colors.red,
),
)
],
),
),
);
}
}
```
**Expected results:** <!-- what did you want to see? -->
If the user scrolls to the top from a different position than the initialScreenOffset, the controller should jump back to this position.
**Actual results:** <!-- what did you see? -->
```
════════ Exception caught by gesture ═══════════════════════════════════════════
The following assertion was thrown while handling a gesture:
'package:flutter/src/widgets/scroll_position.dart': Failed assertion: line 737 pos 12: 'activity.isScrolling': is not true.
Either the assertion indicates an error in the framework itself, or we should provide substantially more information in this error message to help you determine and fix the underlying cause.
In either case, please report this assertion by filing a bug on GitHub:
https://github.com/flutter/flutter/issues/new?template=BUG.md
When the exception was thrown, this was the stack
#2 ScrollPosition.didOverscrollBy
package:flutter/…/widgets/scroll_position.dart:737
#3 ScrollPosition.setPixels
package:flutter/…/widgets/scroll_position.dart:248
#4 ScrollPositionWithSingleContext.setPixels
package:flutter/…/widgets/scroll_position_with_single_context.dart:83
#5 ScrollPositionWithSingleContext.applyUserOffset
package:flutter/…/widgets/scroll_position_with_single_context.dart:126
#6 ScrollDragController.update
package:flutter/…/widgets/scroll_activity.dart:372
...
Handler: "onUpdate"
Recognizer: VerticalDragGestureRecognizer#7b432
start behavior: start
════════════════════════════════════════════════════════════════════════════════
```
<details>
<summary>Logs</summary>
<!--
Run `flutter analyze` and attach any output of that command below.
If there are any analysis errors, try resolving them before filing this issue.
-->
```
[flo@localhost bugtest]$ flutter analyze
Analyzing bugtest...
No issues found! (ran in 13.0s)
```
<!-- Finally, paste the output of running `flutter doctor -v` here. -->
```
[flutter] flutter doctor -v
[✓] Flutter (Channel master, v1.17.1-pre.10, on Linux, locale en_US.UTF-8)
• Flutter version 1.17.1-pre.10 at /home/flo/Documents/flutter
• Framework revision 63f8b9a4d5 (29 minutes ago), 2020-04-03 10:34:14 -0700
• Engine revision f5127cc07a
• Dart version 2.8.0 (build 2.8.0-dev.19.0 fae35fca47)
[✓] Android toolchain - develop for Android devices (Android SDK version 28.0.3)
• Android SDK at /home/flo/Documents/android/sdk
• Platform android-28, build-tools 28.0.3
• ANDROID_SDK_ROOT = /home/flo/Documents/android/
• Java binary at: /home/flo/Documents/android-studio/jre/bin/java
• Java version OpenJDK Runtime Environment (build 1.8.0_212-release-1586-b4-5784211)
• All Android licenses accepted.
[✓] Android Studio (version 3.6)
• Android Studio at /home/flo/Documents/android-studio
• Flutter plugin version 44.0.2
• Dart plugin version 192.7761
• Java version OpenJDK Runtime Environment (build 1.8.0_212-release-1586-b4-5784211)
[✓] VS Code (version 1.43.2)
• VS Code at /usr/share/code
• Flutter extension version 3.8.1
[✓] Connected device (1 available)
• ONEPLUS A5010 • b6f09ad2 • android-arm64 • Android 9 (API 28)
• No issues found!
```
</details>
|
c: crash,framework,f: scrolling,has reproducible steps,P2,workaround available,found in release: 3.7,found in release: 3.9,team-framework,triaged-framework
|
low
|
Critical
|
593,577,304 |
pytorch
|
TensorOptions shouldn't provide default values
|
TensorOptions shouldn't be in the business of providing default values for missing properties.
TensorOptions provides a simple kwargs-like structure for ferrying around the bundle of Tensor-relevant properties in C++, in a multitude of contexts. Semantically it's a struct holding a bunch of `optional`s, nothing more.
Currently the API provides the following query methods for most properties:
* `optional<T> foo_opt()` returns the value of `foo`, as held by the TO object.
* `T foo()` returns the value of `foo` if it's set. **Otherwise, it returns a default value.**
* note that all properties are initialized as unset, but can also be unset explicitly, by passing `nullopt` to their setters.
* `bool has_foo()` is true iff the property value is set. `foo_opt().has_value() == has_foo()`
The problem is that baking defaults into TO's API canonicalizes particular default values that may be irrelevant or inappropriate for some contexts. Also, it's easier to lose or overlook information about which properties are actually set.
Ideally the API should be trimmed back to `foo_opt()` getters only. If this would bloat client code disastrously, we *could* retain `has_foo()` and the nonoptional `foo()` getters and make the latter throw. But it's a choice we would want to make carefully.
|
triaged,better-engineering
|
low
|
Minor
|
593,586,059 |
create-react-app
|
Pin babel-preset-react-app to a minor core-js version as per core-js readme
|
### Is your proposal related to a problem?
I've been trying to polyfill ```Promise.allSettled``` for Edge by using react-app-polyfill/stable but failing. My expectation was that it should work since [email protected] depends on [email protected] and ```Promise.allSettled``` was moved into core-js/stable on v3.2.0.
I've dug a bit deeper to find that babel-preset-react-app has corejs set to major version 3 (https://github.com/facebook/create-react-app/blob/v3.4.1/packages/babel-preset-react-app/dependencies.js#L91)
Looking at the core-js readme I see this:
> Warning! Recommended to specify used minor core-js version, like corejs: '3.6', instead of corejs: 3, since with corejs: 3 will not be injected modules which were added in minor core-js releases.
Locally I've set corejs in babel-preset-react-app to 3.2 (just a proof this is the issue) and I can see ```Promise.allSettled``` is now polyfilled correctly.
It's not clear to me why babel-preset-react-app does not follow the advice of core-js and pin to a minor version?
### Describe the solution you'd like
Update babel-preset-react-app to pin to a minor version of corejs in it's configuration.
### Describe alternatives you've considered
Eject and manage babel-preset-env myself but I'm very reluctant to go down this route.
I look forward to being told why things aren't as simple as changing babel-preset-react-app configuration :)
|
issue: proposal,needs triage
|
low
|
Minor
|
593,635,593 |
flutter
|
[Web] Early errors are not reported to the flutter tools
|
In Flutter for Web, errors are only sent after the main.dart.js script loads, this means that, unfortunately, errors from things like _loading_ that main.dart.js, or from script errors in external scripts, are not always reported to the flutter tools.
This leads to frustrated users that don't know/forget about the browser's builtin debug tools thinking that "Flutter Web is broken" and "Their app only shows white and no errors in the console".
It could be important to either find a way to report early errors (via, maybe, some JS element that only does error catching/reporting), or a big prompt/automatic opening of the browser's developer tools in order to bring attention to them.
CC. @jonahwilliams
|
tool,platform-web,P3,team-web,triaged-web
|
low
|
Critical
|
593,643,376 |
PowerToys
|
[Image Resizer] Show "Resize pictures" on all supported formats
|
**Resize pictures** should appear when right-clicking on any image format with a corresponding WIC codec. This would mean that installing things like following would make it appear.
- [HEIF Image Extensions](https://www.microsoft.com/p/heif-image-extensions/9pmmsr1cgpwg)
- [Raw Image Extension](https://www.microsoft.com/p/raw-image-extension/9nctdw2w1bh8)
- [Webp Image Extensions](https://www.microsoft.com/p/webp-image-extensions/9pg2dk419drg)
- [Microsoft Camera Codec Pack](https://www.microsoft.com/download/details.aspx?id=26829)
Ideally the formats would also appear in the fallback encoder list.
|
Idea-Enhancement,Product-Image Resizer
|
low
|
Major
|
593,644,515 |
PowerToys
|
[Settings][Image Resizer] Allow reordering the default sizes
|
You should be able to drag-and-drop reorder the default sizes in the **Settings** dialog.
|
Product-Settings,Area-User Interface,Product-Image Resizer,Priority-3
|
low
|
Major
|
593,650,284 |
rust
|
deeply-nested chain hangs with Item = u32
|
<!--
Thank you for filing a bug report! 🐛 Please provide a short summary of the bug,
along with any information you feel relevant to replicating the bug.
-->
I tried this code:
```rust
pub fn foo() -> Box<Iterator<Item = u32>> {
use std::iter::empty;
Box::new(empty()
.chain(empty())
.chain(empty())
.chain(empty())
.chain(empty())
.chain(empty())
.chain(empty())
.chain(empty())
.chain(empty())
.chain(empty())
.chain(empty()) // 10th .chain(empty())
.chain(empty())
.chain(empty())
.chain(empty())
.chain(empty())
.chain(empty())
.chain(empty()) // 16th .chain(empty())
)
}
```
This is adapted from the [deeply-nested](https://github.com/rust-lang/rustc-perf/blob/master/collector/benchmarks/deeply-nested/src/lib.rs) test, just changing `Item = ()` to `Item = u32`.
Building with `rustc +nightly --crate-type lib -Copt-level=2 src/lib.rs` seems to hang indefinitely.
I expected to see this happen: completion in a few seconds at most.
Instead, this happened: I'm still waiting...
### Meta
<!--
If you're using the stable version of the compiler, you should also check if the
bug also exists in the beta or nightly versions.
-->
`rustc --version --verbose`:
```
rustc 1.44.0-nightly (537ccdf3a 2020-04-02)
binary: rustc
commit-hash: 537ccdf3ac44c8c7a8d36cbdbe6fb224afabb7ae
commit-date: 2020-04-02
host: x86_64-unknown-linux-gnu
release: 1.44.0-nightly
LLVM version: 9.0
```
Also happens on 1.42.0 and 1.43.0-beta.3.
Attaching rust-gdb, I see two busy LLVM threads:
<!--
Include a backtrace in the code block by setting `RUST_BACKTRACE=1` in your
environment. E.g. `RUST_BACKTRACE=1 cargo build`.
-->
<details><summary>Backtrace</summary>
<p>
```
Thread 6 (Thread 0x7f70bd5ff700 (LWP 207946)):
#0 0x00007f70c72133db in llvm::DominatorTreeBase<llvm::BasicBlock, false>::dominates(llvm::BasicBlock const*, llvm::BasicBlock const*) const () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#1 0x00007f70c7e2298a in llvm::GVN::findLeader(llvm::BasicBlock const*, unsigned int) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#2 0x00007f70c7e23279 in llvm::GVN::processInstruction(llvm::Instruction*) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#3 0x00007f70c7e24977 in llvm::GVN::processBlock(llvm::BasicBlock*) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#4 0x00007f70c7e1aff0 in llvm::GVN::runImpl(llvm::Function&, llvm::AssumptionCache&, llvm::DominatorTree&, llvm::TargetLibraryInfo const&, llvm::AAResults&, llvm::MemoryDependenceResults*, llvm::LoopInfo*, llvm::OptimizationRemarkEmitter*) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#5 0x00007f70c7e27098 in llvm::gvn::GVNLegacyPass::runOnFunction(llvm::Function&) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#6 0x00007f70c727842f in llvm::FPPassManager::runOnFunction(llvm::Function&) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#7 0x00007f70c8233a78 in (anonymous namespace)::CGPassManager::runOnModule(llvm::Module&) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#8 0x00007f70c7279030 in llvm::legacy::PassManagerImpl::run(llvm::Module&) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#9 0x00007f70c71e210a in LLVMRunPassManager () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#10 0x00007f70caa79b43 in rustc_codegen_llvm::back::write::optimize () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/librustc_driver-ebfe476c9299964b.so
#11 0x00007f70caa3ef17 in rustc_codegen_ssa::back::write::execute_work_item () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/librustc_driver-ebfe476c9299964b.so
#12 0x00007f70cab2680e in std::sys_common::backtrace::__rust_begin_short_backtrace () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/librustc_driver-ebfe476c9299964b.so
#13 0x00007f70caa08eb5 in core::ops::function::FnOnce::call_once{{vtable-shim}} () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/librustc_driver-ebfe476c9299964b.so
#14 0x00007f70c9fe92ff in <alloc::boxed::Box<F> as core::ops::function::FnOnce<A>>::call_once () at /rustc/537ccdf3ac44c8c7a8d36cbdbe6fb224afabb7ae/src/liballoc/boxed.rs:1008
#15 0x00007f70ca01c813 in <alloc::boxed::Box<F> as core::ops::function::FnOnce<A>>::call_once () at /rustc/537ccdf3ac44c8c7a8d36cbdbe6fb224afabb7ae/src/liballoc/boxed.rs:1008
#16 std::sys_common::thread::start_thread () at src/libstd/sys_common/thread.rs:13
#17 std::sys::unix::thread::Thread::new::thread_start () at src/libstd/sys/unix/thread.rs:80
#18 0x00007f70c9f534e2 in start_thread () from /lib64/libpthread.so.0
#19 0x00007f70c9e706d3 in clone () from /lib64/libc.so.6
Thread 5 (Thread 0x7f70be1ff700 (LWP 207945)):
#0 0x00007f70c7213354 in llvm::DominatorTreeBase<llvm::BasicBlock, false>::dominates(llvm::BasicBlock const*, llvm::BasicBlock const*) const () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#1 0x00007f70c7e2298a in llvm::GVN::findLeader(llvm::BasicBlock const*, unsigned int) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#2 0x00007f70c7e23279 in llvm::GVN::processInstruction(llvm::Instruction*) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#3 0x00007f70c7e24977 in llvm::GVN::processBlock(llvm::BasicBlock*) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#4 0x00007f70c7e1aff0 in llvm::GVN::runImpl(llvm::Function&, llvm::AssumptionCache&, llvm::DominatorTree&, llvm::TargetLibraryInfo const&, llvm::AAResults&, llvm::MemoryDependenceResults*, llvm::LoopInfo*, llvm::OptimizationRemarkEmitter*) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#5 0x00007f70c7e27098 in llvm::gvn::GVNLegacyPass::runOnFunction(llvm::Function&) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#6 0x00007f70c727842f in llvm::FPPassManager::runOnFunction(llvm::Function&) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#7 0x00007f70c8233a78 in (anonymous namespace)::CGPassManager::runOnModule(llvm::Module&) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#8 0x00007f70c7279030 in llvm::legacy::PassManagerImpl::run(llvm::Module&) () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#9 0x00007f70c71e210a in LLVMRunPassManager () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/../lib/libLLVM-9-rust-1.44.0-nightly.so
#10 0x00007f70caa79b43 in rustc_codegen_llvm::back::write::optimize () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/librustc_driver-ebfe476c9299964b.so
#11 0x00007f70caa3ef17 in rustc_codegen_ssa::back::write::execute_work_item () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/librustc_driver-ebfe476c9299964b.so
#12 0x00007f70cab2680e in std::sys_common::backtrace::__rust_begin_short_backtrace () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/librustc_driver-ebfe476c9299964b.so
#13 0x00007f70caa08eb5 in core::ops::function::FnOnce::call_once{{vtable-shim}} () from /home/jistone/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/../lib/librustc_driver-ebfe476c9299964b.so
#14 0x00007f70c9fe92ff in <alloc::boxed::Box<F> as core::ops::function::FnOnce<A>>::call_once () at /rustc/537ccdf3ac44c8c7a8d36cbdbe6fb224afabb7ae/src/liballoc/boxed.rs:1008
#15 0x00007f70ca01c813 in <alloc::boxed::Box<F> as core::ops::function::FnOnce<A>>::call_once () at /rustc/537ccdf3ac44c8c7a8d36cbdbe6fb224afabb7ae/src/liballoc/boxed.rs:1008
#16 std::sys_common::thread::start_thread () at src/libstd/sys_common/thread.rs:13
#17 std::sys::unix::thread::Thread::new::thread_start () at src/libstd/sys/unix/thread.rs:80
#18 0x00007f70c9f534e2 in start_thread () from /lib64/libpthread.so.0
#19 0x00007f70c9e706d3 in clone () from /lib64/libc.so.6
```
</p>
</details>
|
A-LLVM,T-compiler,C-bug,I-hang
|
low
|
Critical
|
593,656,143 |
TypeScript
|
'declare method' quick fix for adding a private method
|
<!-- 🚨 STOP 🚨 𝗦𝗧𝗢𝗣 🚨 𝑺𝑻𝑶𝑷 🚨
Half of all issues filed here are duplicates, answered in the FAQ, or not appropriate for the bug tracker. Even if you think you've found a *bug*, please read the FAQ first, especially the Common "Bugs" That Aren't Bugs section!
Please help us by doing the following steps before logging an issue:
* Search: https://github.com/Microsoft/TypeScript/search?type=Issues
* Read the FAQ: https://github.com/Microsoft/TypeScript/wiki/FAQ
Please fill in the *entire* template below.
-->
From https://github.com/microsoft/vscode/issues/94118
<!--
Please try to reproduce the issue with the latest published version. It may have already been fixed.
For npm: `typescript@next`
This is also the 'Nightly' version in the playground: http://www.typescriptlang.org/play/?ts=Nightly
-->
**TypeScript Version:** 3.9.0-dev.20200330
**Search terms**
- quick fix
- declare method
- private
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Repro**
For ts file:
`index.ts`
```ts
class Bar {
bar() {
this._baz(123)
}
}
```
1. Trigger quick fixes on `_baz`
**Feature request:**
https://github.com/microsoft/TypeScript/issues/36249 added a quick fix for adding private properties. However there is no quick fix for adding a private method
**Playground Link:** <!-- A link to a TypeScript Playground "Share" link which demonstrates this behavior -->
**Related Issues:** <!-- Did you find other bugs that looked similar? -->
- https://github.com/microsoft/TypeScript/issues/36249
|
Suggestion,Help Wanted,Good First Issue
|
medium
|
Critical
|
593,657,611 |
opencv
|
CUDA Canny missing edges near image border
|
##### System information (version)
- OpenCV => 2.4.13.6 and 3.4 (example for 3.4 shown below)
- Operating System / Platform => Windows 64 Bit
- Compiler => Visual Studio 2017
##### Detailed description
Noticed that the CPU implementation of Canny will produce edges up to the image border, while the GPU/CUDA versions leave an empty border around the image (i.e., edges will not go to image border). This becomes an issue when breaking a larger image into sub parts and merging back together in the end as it creates gaps between edges for the GPU/CUDA version, but not the CPU version.
##### Steps to reproduce
#include <iostream>
#include <opencv2/core/core.hpp>
#include <opencv2/core/cuda.hpp>
#include <opencv2/cudaimgproc.hpp>
#include <opencv2/imgcodecs.hpp>
#include "opencv2/imgproc.hpp"
#include <opencv2/highgui.hpp>
int main()
{
cv::Mat img;
img = cv::imread("path_to_test_img");
cv::Mat gray;
cv::cvtColor(img, gray, cv::COLOR_BGR2GRAY);
cv::cuda::GpuMat gray_gpu;
gray_gpu.upload(gray);
cv::Ptr<cv::cuda::CannyEdgeDetector> edge = cv::cuda::createCannyEdgeDetector(12, 60);
cv::cuda::GpuMat edge_img;
edge->detect(gray_gpu, edge_img);
cv::Mat edge_out;
edge_img.download(edge_out);
cv::Mat edge_cpu;
cv::Canny(gray, edge_cpu,12,60);
std::cout << "done." << std::endl;
|
category: imgproc,category: gpu/cuda (contrib),incomplete
|
low
|
Minor
|
593,659,172 |
flutter
|
[Material] Add focus state, highlight state, and keyboard shortcuts to RangeSlider
|
Related Slider bug: https://github.com/flutter/flutter/issues/48905
The RangeSlider solution should be similar, but having 2 thumbs makes this more tricky.
|
c: new feature,framework,f: material design,c: proposal,P3,team-design,triaged-design
|
low
|
Critical
|
593,667,351 |
pytorch
|
[JIT] Tensor method API behavior discrepancy, Tensor.detach(..)
|
In Python, `Tensor.detach()` does not copy the tensor storage. In-place value changes update the original tensor.
See doc, https://pytorch.org/docs/stable/autograd.html#torch.Tensor.detach.
```py
b = a.detach()
b -= c
```
a is changed as well.
More examples in https://github.com/pytorch/pytorch/issues/6990
In JIT, `Tensor.detach()` always copies the tensor storage. As the implementation is,
https://github.com/pytorch/pytorch/blob/a72946dbab9c1b79a9c75df4f353a341e2ddb19c/aten/src/ATen/native/TensorProperties.cpp#L55-L80
cc @suo
|
needs reproduction,oncall: jit,triaged
|
low
|
Minor
|
593,677,131 |
TypeScript
|
Support bigint literals in const enums
|
<!-- 🚨 STOP 🚨 𝗦𝗧𝗢𝗣 🚨 𝑺𝑻𝑶𝑷 🚨
Half of all issues filed here are duplicates, answered in the FAQ, or not appropriate for the bug tracker.
Please help us by doing the following steps before logging an issue:
* Search: https://github.com/Microsoft/TypeScript/search?type=Issues
* Read the FAQ, especially the "Common Feature Requests" section: https://github.com/Microsoft/TypeScript/wiki/FAQ
-->
## Search Terms
- "const enum member initializers can only contain literal values"
- "bigint enum"
- "ts(2474)"
## Suggestion
A bigint literal is a literal, but it does not seem usable in const enums.
```ts
const enum Foo {
Bar = 123n
}
```
Currently produces, on 3.9.0-beta:
> const enum member initializers can only contain literal values and other computed enum values. ts(2474)
## Use Cases
Bigint compile-time constants 🙂
## Examples
```
const enum Foo {
Bar = 123n
}
console.log(Foo.Bar);
```
Should compile to
```
console.log(123n);
```
And log `123n` when run.
## Checklist
My suggestion meets these guidelines:
* [x] This wouldn't be a breaking change in existing TypeScript/JavaScript code
* [x] This wouldn't change the runtime behavior of existing JavaScript code
* [x] This could be implemented without emitting different JS based on the types of the expressions
* [x] This isn't a runtime feature (e.g. library functionality, non-ECMAScript syntax with JavaScript output, etc.)
* [x] This feature would agree with the rest of [TypeScript's Design Goals](https://github.com/Microsoft/TypeScript/wiki/TypeScript-Design-Goals).
|
Suggestion,Awaiting More Feedback
|
medium
|
Critical
|
593,681,613 |
pytorch
|
JITed GRU too slow
|
## 🐛 Bug
It is advertised, that forward pass of JITed RNNs (e.g. GRU) is as fast as cuDNN implementation.
But it is not the case.
## To Reproduce
Steps to reproduce the behavior:
See here: https://gist.github.com/usamec/af21be7b83e6b1a3f38c26136af811f3
## Expected behavior
Forward pass is as fast as cuDNN.
## Environment
```
Collecting environment information...
PyTorch version: 1.4.0
Is debug build: No
CUDA used to build PyTorch: 10.0
OS: Ubuntu 18.04.1 LTS
GCC version: (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
CMake version: version 3.10.2
Python version: 3.6
Is CUDA available: Yes
CUDA runtime version: 9.0.176
GPU models and configuration: GPU 0: TITAN Xp
Nvidia driver version: 430.50
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.6.0.21
/usr/lib/x86_64-linux-gnu/libcudnn.so.7.5.0
Versions of relevant libraries:
[pip3] numpy==1.15.0
[conda] _pytorch_select 0.2 gpu_0
[conda] blas 1.0 mkl
[conda] cudatoolkit 10.0.130 0
[conda] mkl 2019.4 243
[conda] mkl-service 2.3.0 py36he904b0f_0
[conda] mkl_fft 1.0.15 py36ha843d7b_0
[conda] mkl_random 1.1.0 py36hd6b4f25_0
[conda] numpy 1.18.1 py36h4f9e942_0
[conda] numpy-base 1.18.1 py36hde5b4d6_1
[conda] pytorch 1.4.0 py3.6_cuda10.0.130_cudnn7.6.3_0 pytorch
[conda] pytorch-qrnn 0.2.1 pypi_0 pypi
[conda] torch-scatter 1.4.0 pypi_0 pypi
[conda] torchvision 0.5.0 pypi_0 pypi
```
I am also getting same slowdown on GeForce RTX 2080 Ti
## Additional context
Posted here first.
https://discuss.pytorch.org/t/jited-gru-too-slow/68873
Warm starting does not change much at all.
cc @ezyang @gchanan @zou3519 @suo
|
high priority,triage review,oncall: jit,triaged
|
low
|
Critical
|
593,705,197 |
pytorch
|
C++ tensor print doesn't show requires_grad and grad_fn like Python tensor print
|
In Python, we have the following tensor print behavior:
```python
>>> x=torch.ones(2, 2, requires_grad=True)
>>> x
tensor([[1., 1.],
[1., 1.]], requires_grad=True)
>>> y=x*2
>>> y
tensor([[2., 2.],
[2., 2.]], grad_fn=<MulBackward0>)
```
However in C++, we have the following behavior:
```cpp
auto x = torch::ones({2, 2}, torch::requires_grad());
std::cout << x << std::endl;
auto y = x * 2;
std::cout << y << std::endl;
```
```cpp
1 1
1 1
[ CPUFloatType{2,2} ]
2 2
2 2
[ CPUFloatType{2,2} ]
```
It would be nice to show `requires_grad` and `grad_fn` in C++ tensor print as well.
cc @yf225
|
module: printing,module: cpp,triaged
|
low
|
Minor
|
593,737,212 |
rust
|
Multiple associated type-parametrized bounds fail to resolve
|
With this source
```rust
use core::marker::PhantomData;
pub trait Wrap<T> {
type Wrapper;
}
fn fail<
T,
C: Trait<<WrapMarker as Wrap<T>>::Wrapper>
+ Wrap<<C as Trait<<WrapMarker as Wrap<T>>::Wrapper>>::Assoc>,
>() {
}
fn succeed<
T,
C: Trait<PhantomData<T>> + Wrap<<C as Trait<<WrapMarker as Wrap<T>>::Wrapper>>::Assoc>,
>() {
}
pub trait Trait<P> {
type Assoc;
}
pub struct WrapMarker;
impl<T> Wrap<T> for WrapMarker {
type Wrapper = PhantomData<T>;
}
```
I expect both `fail` and `succeed` to compile. Instead, `fail` fails to compile with "E0277: the trait bound `C: Trait<std::marker::PhantomData<T>>` is not satisfied". `rustc` is clearly able to unambiguously resolve the type here, not only just in theory as associated types are singly defined for a given parametrization, but in practice as it is able to clearly specify the constraint to be added, indicating that `<C as Trait<<WrapMarker as Wrap<T>>::Wrapper` is already uniquely resolved.
[playground reproduction](https://play.rust-lang.org/?version=stable&mode=debug&edition=2018&gist=fe6ad4e097de20ae2887ee6caa4ca795)
|
A-trait-system,A-associated-items,T-compiler,C-bug,A-lazy-normalization
|
low
|
Critical
|
593,755,467 |
rust
|
rustdoc's treatment of cross-crate ("imported") files is suboptimal.
|
@bjorn3 noted in https://github.com/rust-lang/rust/issues/70025#issuecomment-606566181 that single files missing disables `[src]` links for *the entire crate*.
That's not ideal UX, we should render as many source files as we can possible load.
<hr/>
And there's also the paths `rustdoc` uses (see https://github.com/rust-lang/rust/issues/70025#issuecomment-607699347), e.g.:
* https://doc.rust-lang.org/nightly/nightly-rustc/rustc_span/def_id/struct.CrateId.html
* links to https://doc.rust-lang.org/nightly/nightly-rustc/src/rustc_span/checkout/src/librustc_index/vec.rs.html#110-112
* but it should use https://doc.rust-lang.org/nightly/nightly-rustc/src/rustc_index/vec.rs.html#110-112
If we use the `CrateNum` of the original source file and reuse the existing `rustdoc` logic to get the relative path of that file in that crate, *I believe* we can even support linking to the source correctly even when the *absolute* paths are not usable (e.g. `/rustc/$hash/src/libstd/thread/local.rs` when using `thread_local!` and the `rust-src` `rustup` component isn't installed)
Because the downstream crate uses an exported macro from the original crate, the rendered source file should always have been generated, so we should never really need to re-generate it.
cc @rust-lang/rustdoc @Aaron1011
|
T-rustdoc,C-enhancement
|
low
|
Minor
|
593,762,503 |
flutter
|
macOS project doesn't run pod install for me when building
|
flutter created a project with `--enable-macos-desktop` turned on.
added
```yaml
dependencies:
graphql: ^3.0.0
```
to dependencies.
flutter run
CLI just returned
```
Launching lib/main.dart on macOS in debug mode...
** BUILD FAILED **
Building macOS application...
Exception: Build process failed
```
which was rather cryptic.
Running in Xcode gave
```
Showing Recent Messages
<project>/macos/Runner/Configs/../../Flutter/Flutter-Debug.xcconfig:1: could not find included file 'Pods/Target Support Files/Pods-Runner/Pods-Runner.debug.xcconfig' in search paths
<project>/macos/Runner/Configs/../../Flutter/Flutter-Debug.xcconfig:1: could not find included file 'Pods/Target Support Files/Pods-Runner/Pods-Runner.debug.xcconfig' in search paths
<project>/macos/Runner/Configs/../../Flutter/Flutter-Debug.xcconfig:1: could not find included file 'Pods/Target Support Files/Pods-Runner/Pods-Runner.debug.xcconfig' in search paths
```
going into macos and `pod install`ing fixes it. But I would have assumed that flutter tool should run that for me.
|
platform-mac,a: desktop,a: build,P2,team-macos,triaged-macos
|
low
|
Critical
|
593,773,175 |
PowerToys
|
[Image Resizer] Better support for HEiC image formats
|
Can HEIC images be catered for in ImageResizer
|
Help Wanted,Product-Image Resizer
|
medium
|
Major
|
593,774,651 |
angular
|
TestBed: Error: Directive SomePipe has no selector, please add it!
|
<!--🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅
Oh hi there! 😄
To expedite issue processing please search open and closed issues before submitting a new one.
Existing issues often contain information about workarounds, resolution, or progress updates.
🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅🔅-->
# 🐞 bug report
### Affected Package
<!-- Can you pin-point one or more @angular/* packages as the source of the bug? -->
<!-- ✍️edit: --> The issue is caused by package @angular/core/testing
### Is this a regression?
Yes, this worked in 8.2.
### Description
When I have a Pipe in a library project that has a super class with a ngOnDestroy method, a TestBed that imports the library module fails with the error "Directive SomePipe has no selector, please add it!".
This doesn't happen when the super class does not have an ngOnDestroy method, and it doesn't happen if the Pipe is included directly in the TestBed, or if the Module is part of the default app instead of a library project.
```
export abstract class SomeBase implements OnDestroy {
/*
* My real base class had a transform method, ect, but this seems to be
* all that's needed to trigger it, and it wouldn't trigger without it
*/
ngOnDestroy() {
}
}
@Pipe({
name: 'some'
})
export class SomePipe extends SomeBase implements OnDestroy, PipeTransform {
transform(value: unknown, ...args: unknown[]): unknown {
return null;
}
// Is this needed, or will Ivy use the base class's ngOnDestroy without it??
// it's not needed to trigger the error
ngOnDestroy() {
super.ngOnDestroy();
}
}
```
```
beforeEach(async(() => {
TestBed.configureTestingModule({
imports: [SomeLibModule],
declarations: [
AppComponent
],
}).compileComponents();
}));
```
## 🔬 Minimal Reproduction
https://github.com/james-schwartzkopf/ng9-issues/tree/lib-pipe-issue
```
yarn
ng build some-lib
ng test
```
## 🔥 Exception or Error
<pre><code>
Error: Directive SomePipe has no selector, please add it!
at verifySemanticsOfNgModuleDef (http://localhost:9877/_karma_webpack_/node_modules/@angular/core/__ivy_ngcc__/fesm2015/core.js:39047:1)
at http://localhost:9877/_karma_webpack_/node_modules/@angular/core/__ivy_ngcc__/fesm2015/core.js:39010:1
at <Jasmine>
at verifySemanticsOfNgModuleDef (http://localhost:9877/_karma_webpack_/node_modules/@angular/core/__ivy_ngcc__/fesm2015/core.js:39004:1)
at Function.get (http://localhost:9877/_karma_webpack_/node_modules/@angular/core/__ivy_ngcc__/fesm2015/core.js:38956:1)
at R3TestBedCompiler.applyProviderOverridesToModule (http://localhost:9877/_karma_webpack_/node_modules/@angular/core/__ivy_ngcc__/fesm2015/testing.js:2087:44)
at R3TestBedCompiler.compileTestModule (http://localhost:9877/_karma_webpack_/node_modules/@angular/core/__ivy_ngcc__/fesm2015/testing.js:2386:1)
at R3TestBedCompiler.finalize (http://localhost:9877/_karma_webpack_/node_modules/@angular/core/__ivy_ngcc__/fesm2015/testing.js:1880:1)
at TestBedRender3.get testModuleRef [as testModuleRef] (http://localhost:9877/_karma_webpack_/node_modules/@angular/core/__ivy_ngcc__/fesm2015/testing.js:3253:1)
at TestBedRender3.inject (http://localhost:9877/_karma_webpack_/node_modules/@angular/core/__ivy_ngcc__/fesm2015/testing.js:3110:1)
</code></pre>
## 🌍 Your Environment
**Angular Version:**
<pre><code>
<!-- run `ng version` and paste output below -->
<!-- ✍️-->
$ ng version
_ _ ____ _ ___
/ \ _ __ __ _ _ _| | __ _ _ __ / ___| | |_ _|
/ △ \ | '_ \ / _` | | | | |/ _` | '__| | | | | | |
/ ___ \| | | | (_| | |_| | | (_| | | | |___| |___ | |
/_/ \_\_| |_|\__, |\__,_|_|\__,_|_| \____|_____|___|
|___/
Angular CLI: 9.1.0
Node: 12.16.1
OS: win32 x64
Angular: 9.1.0
... animations, cli, common, compiler, compiler-cli, core, forms
... language-service, platform-browser, platform-browser-dynamic
... router
Ivy Workspace: Yes
Package Version
------------------------------------------------------------
@angular-devkit/architect 0.901.0
@angular-devkit/build-angular 0.901.0
@angular-devkit/build-ng-packagr 0.901.0
@angular-devkit/build-optimizer 0.901.0
@angular-devkit/build-webpack 0.901.0
@angular-devkit/core 9.1.0
@angular-devkit/schematics 9.1.0
@ngtools/webpack 9.1.0
@schematics/angular 9.1.0
@schematics/update 0.901.0
ng-packagr 9.1.0
rxjs 6.5.5
typescript 3.8.3
webpack 4.42.0
</code></pre>
**Anything else relevant?**
<!-- ✍️Is this a browser specific issue? If so, please specify the browser and version. -->
<!-- ✍️Do any of these matter: operating system, IDE, package manager, HTTP server, ...? If so, please mention it below. -->
|
type: bug/fix,area: testing,freq2: medium,regression,area: compiler,state: confirmed,P3,compiler: jit
|
low
|
Critical
|
593,782,329 |
node
|
Exception while running 'es' benchmarks
|
* **Version**: master
* **Platform**: Linux x64
* **Subsystem**: console?
### What steps will reproduce the bug?
Running the 'es' benchmarks.
### How often does it reproduce? Is there a required condition?
Not often it seems, not sure if it's specific to any 'es' benchmark.
### What is the expected behavior?
To not crash.
### What do you see instead?
```
internal/console/global.js:45
globalConsole[kBindProperties](true, 'auto');
^
TypeError: globalConsole[kBindProperties] is not a function
at internal/console/global.js:45:31
at NativeModule.compileForInternalLoader (internal/bootstrap/loaders.js:276:7)
at nativeModuleRequire (internal/bootstrap/loaders.js:305:14)
at createGlobalConsole (internal/bootstrap/node.js:317:5)
at internal/bootstrap/node.js:120:38
```
|
benchmark
|
low
|
Critical
|
593,789,517 |
flutter
|
Ability to use ParagraphBuilder & Paragraph in other isolates
|
I need to calculate metrics for thousands of paragraphs in a way that doesn't lead to UI jank.
Therefore I'd like to Instantiate a ParagraphBuilder and build a Paragraph in a separate isolate to pass back metrics to the main isolate.
This is currently not possible.
Related issues:
#10647
#13343
_(I'm opening a separate issue because it was mentioned that it is theoretically possible to have parts of dart:ui accessible to other isolates. I don't need all of dart:ui, just the text related parts)_
Example:
```dart
import 'dart:ui';
import 'package:flutter/foundation.dart';
import 'package:flutter/material.dart';
void main() async {
final size = await compute(computeParagraph, "abc");
runApp(Text("$size"));
}
Size computeParagraph(String text) {
final builder = ParagraphBuilder(ParagraphStyle(textDirection: TextDirection.ltr));
builder.addText(text);
final paragraph = builder.build();
paragraph.layout(const ParagraphConstraints(width: double.infinity));
return Size(paragraph.width, paragraph.height);
}
```
throws
```
[ERROR:flutter/lib/ui/ui_dart_state.cc(157)] Unhandled Exception: Exception: error: native function 'ParagraphBuilder_constructor' (9 arguments) cannot be found
#0 new ParagraphBuilder (dart:ui/text.dart:2070:5)
#1 computeParagraph (package:flutter_companion_macos/main.dart:12:19)
#2 _IsolateConfiguration.apply (package:flutter/src/foundation/_isolates_io.dart:75:34)
#3 _spawn.<anonymous closure> (package:flutter/src/foundation/_isolates_io.dart:83:65)
#4 Timeline.timeSync (dart:developer/timeline.dart:163:22)
#5 _spawn (package:flutter/src/foundation/_isolates_io.dart:80:18)
#6 _startIsolate.<anonymous closure> (dart:isolate-patch/isolate_patch.dart:304:17)
#7 _RawReceivePortImpl._handleMessage (dart:isolate-patch/isolate_patch.dart:168:12)
```
|
engine,c: proposal,P3,team-engine,triaged-engine
|
low
|
Critical
|
593,796,472 |
go
|
cmd/go: -buildmode=c-shared without specifying output file fails does not add .dll extension on Windows and .so extension on linux
|
<!--
Please answer these questions before submitting your issue. Thanks!
For questions please use one of our forums: https://github.com/golang/go/wiki/Questions
-->
### What version of Go are you using (`go version`)?
<pre>
go version go1.14.1 windows/amd64
</pre>
### Does this issue reproduce with the latest release?
Yes
### What operating system and processor architecture are you using (`go env`)?
<details><summary><code>go env</code> Output</summary><br><pre>
set GO111MODULE=auto
set GOARCH=amd64
set GOBIN=
set GOCACHE=C:\Users\anon\AppData\Local\go-build
set GOENV=C:\Users\anon\AppData\Roaming\go\env
set GOEXE=.exe
set GOFLAGS=
set GOHOSTARCH=amd64
set GOHOSTOS=windows
set GOINSECURE=
set GONOPROXY=
set GONOSUMDB=
set GOOS=windows
set GOPATH=C:\Users\anon\go
set GOPRIVATE=
set GOPROXY=https://proxy.golang.org,direct
set GOROOT=c:\go
set GOSUMDB=sum.golang.org
set GOTMPDIR=
set GOTOOLDIR=c:\go\pkg\tool\windows_amd64
set GCCGO=gccgo
set AR=ar
set CC=gcc
set CXX=g++
set CGO_ENABLED=1
set GOMOD=
set CGO_CFLAGS=-g -O2
set CGO_CPPFLAGS=
set CGO_CXXFLAGS=-g -O2
set CGO_FFLAGS=-g -O2
set CGO_LDFLAGS=-g -O2
set PKG_CONFIG=pkg-config
set GOGCCFLAGS=-m64 -mthreads -fmessage-length=0 -fdebug-prefix-map=C:\Users\anon\AppData\Local\Temp\go-build481521658=/tmp/go-build -gno-record-gcc-switches
</pre></details>
### What did you do?
<!--
If possible, provide a recipe for reproducing the error.
A complete runnable program is good.
A link on play.golang.org is best.
-->
I tried to build a c-shared library
```golang
package main
import "fmt"
import "C"
//export test
func test() {
fmt.Println("Hello World!")
}
func main() {
}
```
The archive is linked against a C which just calls `test()` with:
```shell
gcc main.c -L<path to dll folder> -l<dll name without extension> -o test.exe
```
Executing `test.exe` does not produce any output.
### What did you expect to see?
A valid dll file (standard dynamic library format on windows)
### What did you see instead?
The call `go build -buildmode=c-shared`does not yield a valid dll (I think; it does not have any extension by default). It produces a `[folder name]` file without any extension.
I'm not sure what happens, but if you try to link it against a C `main`the `main` function does not get executed. The linker does not produce any warnings or errors about the file (after renaming [name] to [name].dll).
If you produce the archive with `go build -buildmode=c-shared -o <name>.dll` everything works as expected, but `go build -buildmode=c-shared` should be sufficient, but it isn't.
|
help wanted,OS-Windows,NeedsInvestigation,GoCommand
|
low
|
Critical
|
593,838,043 |
flutter
|
[macos] force touch trackpad support.
|
There's currently no way to react to force touch actions on MacOS systems with a force touch trackpad.
Related issues:
#23604
|
c: new feature,engine,platform-mac,a: desktop,P3,team-macos,triaged-macos
|
low
|
Major
|
593,907,087 |
pytorch
|
About tensorboard in pytorch record graph in different state
|
## ❓ Questions and Help
I want to save a model graph which changed along with training. For example, if the training epoch > 5, the model has a new structure. If I need to save both models' graph, how should I do? or can I achieve this? the graph will be overwritten if I save the model twice.
## Environment
- Pytorch: 1.3.1
- OS (e.g., Linux): ubuntu16.04
- Python version: 3.7
## Code
```
import torch
import torchvision
from torch.utils.tensorboard import SummaryWriter
from torchvision import datasets, transforms
import torch.nn as nn
import torch.nn.functional as F
writer = SummaryWriter('./log')
transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.5,), (0.5,))])
trainset = datasets.MNIST('./data/', train=True, download=True, transform=transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=64, shuffle=True)
class Net(nn.Module):
def __init__(self):
super(Net, self).__init__()
self.conv1 = nn.Conv2d(in_channels=1, out_channels=5, kernel_size=5)
self.conv2 = nn.Conv2d(5, 10, 5)
self.conv3 = nn.Conv2d(10, 20, 3)
self.fc = nn.Linear(6480, 10)
self.fc1 = nn.Linear(6480, 40)
self.fc2 = nn.Linear(40, 10)
self.epoch = 0
def forward(self, x):
self.epoch += 1
in_size = x.size(0)
x = F.relu(self.conv1(x))
x = F.relu(self.conv2(x))
x = F.relu(self.conv3(x))
x = x.view(in_size, -1)
if self.epoch < 5:
x = self.fc(x)
else:
x = self.fc1(x)
x = self.fc2(x)
return F.log_softmax(x)
model = Net()
images, labels = next(iter(trainloader))
writer.add_graph(model, images)
model.epoch = 5
writer.add_graph(model, images)
writer.close()
print("-----Finish!-----")
```
|
triaged,oncall: visualization
|
low
|
Minor
|
593,962,892 |
rust
|
Generate better memcpy code for types with alignment padding
|
Consider the type `S0` below, which has 9 bytes of payload data, but because of alignment requirements its size is 16 bytes. It implements `Copy`, so it can be cloned and copied by `memcpy`. rustc tends to emit 9-byte `memcpy` calls for it in several cases, even though it would be legal to emit `memcpy` calls of size anywhere between 9 bytes and 16 bytes. On x86-64, a 9-byte memcpy is 2x slower than a 16-byte memcpy: the former requires 2 loads and 2 stores, whereas the latter can use just 1 load and 1 store. It would be nice for Rust to emit the more efficient memcpy calls. Currently I'm working around this issue by manually padding my type up to 16 bytes of payload data, like in `S1`.
Compare the generated assembly of `copy_s0` and `copy_s1`:
```asm
playground::copy_s0:
movq (%rsi), %rax
movb 8(%rsi), %cl
movq %rax, (%rdi)
movb %cl, 8(%rdi)
retq
playground::copy_s1:
movups (%rsi), %xmm0
movups %xmm0, (%rdi)
retq
```
Similar issues show up for `clone_from_slice`.
```rust
#[derive(Clone, Copy)]
pub struct S0(u64, u8);
pub fn clone_s0(dst: &mut S0, src: &S0) {
*dst = src.clone();
}
pub fn copy_s0(dst: &mut S0, src: &S0) {
*dst = *src;
}
pub fn clone_s0_array(dst: &mut [S0; 8], src: & [S0; 8]) {
*dst = src.clone();
}
pub fn copy_s0_array(dst: &mut [S0; 8], src: & [S0; 8]) {
*dst = *src;
}
pub fn clone_s0_slice(dst: &mut [S0; 8], src: & [S0; 8]) {
dst.clone_from_slice(src);
}
pub fn copy_s0_slice(dst: &mut [S0; 8], src: & [S0; 8]) {
dst.copy_from_slice(src);
}
#[derive(Clone, Copy)]
pub struct S1(u64, u8, [u8; 7]);
pub fn clone_s1(dst: &mut S1, src: &S1) {
*dst = src.clone();
}
pub fn copy_s1(dst: &mut S1, src: &S1) {
*dst = *src;
}
pub fn clone_s1_array(dst: &mut [S1; 8], src: & [S1; 8]) {
*dst = src.clone();
}
pub fn copy_s1_array(dst: &mut [S1; 8], src: & [S1; 8]) {
*dst = *src;
}
pub fn clone_s1_slice(dst: &mut [S1; 8], src: & [S1; 8]) {
dst.clone_from_slice(src);
}
pub fn copy_s1_slice(dst: &mut [S1; 8], src: & [S1; 8]) {
dst.copy_from_slice(src);
}
```
([Playground](https://play.rust-lang.org/?version=stable&mode=release&edition=2018&gist=820411986f8028ac6ad00a93ac953898))
|
A-LLVM,I-slow,C-enhancement,T-compiler
|
low
|
Major
|
593,966,839 |
rust
|
Switching from `impl` to `dyn` changes lifetime inference (maybe?)
|
I'm not really sure how to report this bug, as I'm not even sure it actually is a bug in Rust, due to my not understanding enough how lifetimes are at work in this example.
So I'll just give the code and try to explain (it's quite long, so I've just put it on the playground). The interesting code is at the bottom of the playground links.
#### Failing code
https://play.rust-lang.org/?version=stable&mode=debug&edition=2018&gist=5bb42243a4f952045f4991f280c101e0
This code doesn't compile. It is basically an extraction of the commented code, which compiles.
I originally assumed that this was a mistake on my part of providing the correct lifetimes.
However…
#### Working code
https://play.rust-lang.org/?version=stable&mode=debug&edition=2018&gist=b7f082cd3912f0ad1d7176c0adb51928
Here, the code does compile. The only change is to replace `Pin<Box<dyn 'static + Send + Future<...>>>` with `Pin<Box<RetFut>>` with `RetFut: 'static + Send + Future<...>`
It appears quite weird that this code compiles but not the previous one, and I'm thus reporting this bug.
As usual, thank you for all your work on Rust!
|
C-enhancement,A-diagnostics,A-lifetimes,T-compiler
|
low
|
Critical
|
593,978,378 |
flutter
|
Add keyline to Material Scaffold
|
The spacing methods in material spec appear to intentionally be open to the designers interpretation. In particular, Material offers *keylines* for adjusting spacing for child widgets, as described at the bottom of the [Spacing section](https://material.io/design/layout/spacing-methods.html#spacing) and illustrated in the animation below:
This was originally discussed in #4460 and reemerged in #21026 as part of a discussion about padding.
/cc @lukepighetti
|
c: new feature,framework,f: material design,a: quality,c: proposal,P3,team-design,triaged-design
|
low
|
Minor
|
593,995,263 |
godot
|
Godot crashed and won't start again. Renaming the folder or exe resolves the problem
|
**Godot version:** 3.2.1-stable_win64
**OS/device including version:** Windows 10 1909 (OS Build 18363.729)
**Issue description:**
Hi,
I am unsure if this issue is with Godot or with Windows, but anyway:
- Godot was running.
- (probably unrelated but for completeness) I downloaded robisplash.zip from https://docs.godotengine.org/en/stable/getting_started/step_by_step/animations.html and extracted it into my project folder
- I clicked "Quit to project list"
- Godot crashed and when trying to start it again, the below error was displayed, a blank window appeared for a second and directly closed.
Starting the binary from another path or just renaming the folder it is in or the executable itself resolved the problem.
Renaming it back causes the error again.
Other things that did not help:
- Renaming the project folder (C:\code\godot) in this case.
- Renaming %APPDATA%\Godot
```
C:\apps\godot>Godot_v3.2.1-stable_win64.exe -v
Godot Engine v3.2.1.stable.official - https://godotengine.org
Using GLES3 video driver
OpenGL ES 3.0 Renderer: GeForce GTX 1060 6GB/PCIe/SSE2
WASAPI: wFormatTag = 65534
WASAPI: nChannels = 6
WASAPI: nSamplesPerSec = 48000
WASAPI: nAvgBytesPerSec = 1152000
WASAPI: nBlockAlign = 24
WASAPI: wBitsPerSample = 32
WASAPI: cbSize = 22
WASAPI: detected 6 channels
WASAPI: audio buffer frames: 1962 calculated latency: 44ms
WASAPI: wFormatTag = 65534
WASAPI: nChannels = 2
WASAPI: nSamplesPerSec = 48000
WASAPI: nAvgBytesPerSec = 384000
WASAPI: nBlockAlign = 8
WASAPI: wBitsPerSample = 32
WASAPI: cbSize = 22
WASAPI: detected 2 channels
WASAPI: audio buffer frames: 1962 calculated latency: 44ms
ERROR: get: FATAL: Index p_index = 0 is out of bounds (size() = 0).
At: ./core/cowdata.h:152
```
The following two entries were added to the windows event log:
```
Faulting application name: Godot_v3.2.1-stable_win64.exe, version: 3.2.1.0, time stamp: 0x00000000
Faulting module name: Godot_v3.2.1-stable_win64.exe, version: 3.2.1.0, time stamp: 0x00000000
Exception code: 0xc000001d
Fault offset: 0x00000000009e6614
Faulting process id: 0x3ce0
Faulting application start time: 0x01d60aab7fd68f8a
Faulting application path: C:\apps\godot\Godot_v3.2.1-stable_win64.exe
Faulting module path: C:\apps\godot\Godot_v3.2.1-stable_win64.exe
Report Id: 227a8bfb-63ed-4d04-ae8a-dfe3db138ee7
Faulting package full name:
Faulting package-relative application ID:
```
```
Windows cannot access the file for one of the following reasons: there is a problem with the network connection, the disk that the file is stored on, or the storage drivers installed on this computer; or the disk is missing. Windows closed the program Godot Engine because of this error.
Program: Godot Engine
File:
The error value is listed in the Additional Data section.
User Action
1. Open the file again. This situation might be a temporary problem that corrects itself when the program runs again.
2. If the file still cannot be accessed and
- It is on the network, your network administrator should verify that there is not a problem with the network and that the server can be contacted.
- It is on a removable disk, for example, a floppy disk or CD-ROM, verify that the disk is fully inserted into the computer.
3. Check and repair the file system by running CHKDSK. To run CHKDSK, click Start, click Run, type CMD, and then click OK. At the command prompt, type CHKDSK /F, and then press ENTER.
4. If the problem persists, restore the file from a backup copy.
5. Determine whether other files on the same disk can be opened. If not, the disk might be damaged. If it is a hard disk, contact your administrator or computer hardware vendor for further assistance.
Additional Data
Error value: 00000000
Disk type: 0
```
**Steps to reproduce:**
Unknown unfortunately
**Minimal reproduction project:**
Unknown unfortunately
|
bug,topic:editor,crash
|
low
|
Critical
|
594,013,578 |
rust
|
Missing write coalescing
|
Function `make_constant()` writes a constant to `S`. Optimally it would be a single 16-byte write to memory, but in fact it generates four separate writes. Function `make()` writes a nonconstant and a constant. Ideally it would generate two writes (one nonconstant, one constant), or perhaps an OR followed by a single write. Instead, it also generates four writes.
```asm
playground::make:
movq %rdi, %rax
movq %rsi, (%rdi)
movb $1, 8(%rdi)
movl $0, 9(%rdi)
movl $0, 12(%rdi)
retq
playground::make_constant:
movq %rdi, %rax
movq $5, (%rdi)
movb $1, 8(%rdi)
movl $0, 9(%rdi)
movl $0, 12(%rdi)
retq
```
```rust
pub struct S(u64, bool, [u8; 7]);
pub fn make(x: u64) -> S {
S(x, true, [0; 7])
}
pub fn make_constant() -> S {
S(5, true, [0; 7])
}
```
([Playground](https://play.rust-lang.org/?version=stable&mode=release&edition=2018&gist=66237b264939df6f7670b6398c60fdd1))
|
A-LLVM,C-enhancement,A-codegen,T-compiler
|
low
|
Minor
|
594,085,965 |
tensorflow
|
Named Dimensions
|
**System information**
- TensorFlow version (you are using): 2.1
- Are you willing to contribute it (Yes/No): No
**Describe the feature and the current behavior/state.**
Currently, TensorFlow ops receive axes as integer indices, e.g. `tf.reduce_sum(tensor, axis=(2, 3))`. Keep track of axes after applying several operations on them and passing tensors around through multiple functions can be challenging.
I've seen many researchers and developers grab a sheet of paper and manually work through how axes should change. We can avoid this slow and often frustrating process by supporting optional axes names for tensors and having the most common ops annotate their outputs:
```python
image = tf.name_axes(image, ('B', 'H', 'W', 'C')) # assert len(image.shape) == 4
print(image) # <tf.Tensor dtype=uint8 shape={B: 500, H: 64, W: 64, C: 3}>
grid = tf.repeat(tf.repeat(image, 4, axis='H'), 4, axis='W')
print(grid) # <tf.Tensor dtype=uint8 shape={B: 500, H: 256, W: 256, C: 3}>
gray = tf.reduce_mean(grid, axis='C')
print(gray) # <tf.Tensor dtype=uint8 shape={B: 500, H: 256, W: 256}>
flipped = tf.transpose(gray[0], ('W', 'H'))
print(flipped) # <tf.Tensor dtype=uint8 shape={W: 256, H: 256}>
```
A function starting with the line `tf.name_axes(...)` for its input tensors will immediately be easier to read. Moreover, developers could place `assert tensor.shape.names == ('B', 'F')` statements throughout their code so it becomes immediately clear from the written code how shapes are changing.
An advanced version of named axes would be to keep track of combined axes:
<details>
```python
video = tf.name_axes(image, ('B', 'T', 'H', 'W', 'C'))
print(video) # <tf.Tensor dtype=uint8 shape={B: 500, T: 100, H: 64, W: 64, C: 3}>
frames = tf.reshape(frames, ('B*T', 'H', 'W', C'))
print(frames) # <tf.Tensor dtype=uint8 shape={B*T: 50000, H: 64, W: 64, C: 3}>
logits = model(frames)
print(logits) # <tf.Tensor dtype=float32 shape={B*T: 50000, 10}>
logits = tf.reshape(logits, ('B', 'T', 10))
print(logits) # <tf.Tensor dtype=float32 shape={B: 500, T: 100, 10}>
logits = tf.name_axes(logits, ('B, 'T', 'K'))
print(logits) # <tf.Tensor dtype=float32 shape={B: 500, T: 100, K: 10}>
```
</details>
**Will this change the current api? How?**
This would extend the current API in a fully backward compatible way. Common low-level TensorFlow operations (e.g. reduce_sum, concat, slice, tile) have to check for input tensors with named axes to annotate their outputs correctly.
**Who will benefit with this feature?**
Named axes have the potential to increase productivity of both researchers and engineers. Especially when developing large TensorFlow applications, named axes can help keep the overview. For product development with multiple developers, it can increase readability.
**Any Other info.**
JAX supports axes names. PyTorch 1.3 added experimental support for [named dimensions](https://pytorch.org/docs/stable/named_tensor.html).
|
stat:awaiting tensorflower,type:feature,comp:ops
|
low
|
Major
|
594,116,650 |
java-design-patterns
|
Rule engine design pattern
|
**Description:**
The Rules design pattern is a behavioral design pattern that enables the encapsulation of business rules into isolated, reusable units. This pattern allows for the dynamic application of business logic at runtime and promotes the separation of concerns, making the system more modular and easier to maintain.
**Main Elements of the Pattern:**
1. **Rule Interface**: Defines the contract for all rules, typically including a method to evaluate the rule and another to execute the rule's action.
2. **Concrete Rules**: Implementations of the Rule interface, each encapsulating a specific piece of business logic.
3. **Rule Engine**: Responsible for managing and executing rules. It evaluates which rules should be applied based on the given context and executes them accordingly.
4. **Context**: The data or state that the rules operate on. This is passed to the rule engine and rules during execution.
**References:**
- [Rules Design Pattern by Michael Whelan](https://www.michael-whelan.net/rules-design-pattern/)
- [Project Contribution Guidelines](https://github.com/iluwatar/java-design-patterns/wiki)
**Acceptance Criteria:**
1. Create a Rule interface with methods to evaluate and execute a rule.
2. Implement at least three concrete rule classes demonstrating different business logic scenarios.
3. Develop a RuleEngine class capable of evaluating and executing a set of rules against a provided context.
4. Include unit tests for the RuleEngine and the concrete rule classes to ensure correctness and robustness.
|
info: help wanted,epic: pattern,type: feature
|
medium
|
Major
|
594,138,176 |
flutter
|
"Welcome to Flutter" and "Waiting for lock" messages break --machine commands
|
`sudo flutter channel --version --machine` is responding with a non json output, this break other toolkit/ci-cd that are relying on this `--machine` flag https://github.com/go-flutter-desktop/go-flutter/issues/384.
```sh
$ sudo flutter channel --version --machine
Woah! You appear to be trying to run flutter as root.
We strongly recommend running the flutter tool without superuser privileges.
/
📎
{
"frameworkVersion": "1.16.4-pre.88",
"channel": "master",
"repositoryUrl": "https://github.com/flutter/flutter.git",
"frameworkRevision": "ab14307e0c77e4a03ea23ccaa68436d9128a445d",
"frameworkCommitDate": "2020-04-03 00:56:01 -0400",
"engineRevision": "f5127cc07a762de774b7e941fe6eb402c97aa1b3",
"dartSdkVersion": "2.8.0 (build 2.8.0-dev.19.0 fae35fca47)"
}
╔════════════════════════════════════════════════════════════════════════════╗
║ Welcome to Flutter! - https://flutter.dev ║
║ ║
║ The Flutter tool uses Google Analytics to anonymously report feature usage ║
║ statistics and basic crash reports. This data is used to help improve ║
║ Flutter tools over time. ║
║ ║
║ Flutter tool analytics are not sent on the very first run. To disable ║
║ reporting, type 'flutter config --no-analytics'. To display the current ║
║ setting, type 'flutter config'. If you opt out of analytics, an opt-out ║
║ event will be sent, and then no further information will be sent by the ║
║ Flutter tool. ║
║ ║
║ By downloading the Flutter SDK, you agree to the Google Terms of Service. ║
║ Note: The Google Privacy Policy describes how data is handled in this ║
║ service. ║
║ ║
║ Moreover, Flutter includes the Dart SDK, which may send usage metrics and ║
║ crash reports to Google. ║
║ ║
║ Read about data we send with crash reports: ║
║ https://flutter.dev/docs/reference/crash-reporting ║
║ ║
║ See Google's privacy policy: ║
║ https://policies.google.com/privacy ║
╚════════════════════════════════════════════════════════════════════════════╝
```
|
tool,P2,team-tool,triaged-tool
|
low
|
Critical
|
594,143,696 |
rust
|
Compiler error references non-existent/undefined lifetimes?
|
I'm trying to box the `impl Future` returned from an `async fn`; to that end, I tried wrapping it in a closure that does the boxing:
```rust
let _clousre = |r: &Request<'_>| Box::new(index_get(r));
```
(`index_get` is an `async fn`.)
I expected to see this happen: Well, I had hoped it would compile! I do not know if it should, however.
Instead, this happened:
```
error: lifetime may not live long enough
--> src/main.rs:19:38
|
19 | let _clousre = |r: &Request<'_>| Box::new(index_get(r));
| - - ^^^^^^^^^^^^^^^^^^^^^^ returning this value requires that `'1` must outlive `'2`
| | |
| | return type of closure is std::boxed::Box<impl std::future::Future>
| let's call the lifetime of this reference `'1`
error: lifetime may not live long enough
--> src/main.rs:19:38
|
19 | let _clousre = |r: &Request<'_>| Box::new(index_get(r));
| -- - ^^^^^^^^^^^^^^^^^^^^^^ returning this value requires that `'3` must outlive `'4`
| | |
| | return type of closure is std::boxed::Box<impl std::future::Future>
| let's call this `'3`
```
Here, the error message references lifetimes `'1` and `'2`; it defines `'1`, **but what is `'2`?**
(It also repeats the message twice, which seems odd.)
`rustc --version --verbose`:
```
» rustc --version --verbose
rustc 1.40.0 (73528e339 2019-12-16)
binary: rustc
commit-hash: 73528e339aae0f17a15ffa49a8ac608f50c6cf14
commit-date: 2019-12-16
host: x86_64-unknown-linux-gnu
release: 1.40.0
LLVM version: 9.0
```
While this isn't the latest stable, I can also replicate this in the playground, using both nightly and stable there. [Playground Example](https://play.rust-lang.org/?version=nightly&mode=debug&edition=2018&gist=098996eb7fdd4e8850986f59e7cd9b7a)
|
A-diagnostics,A-lifetimes,A-closures,T-compiler,C-bug,D-confusing
|
low
|
Critical
|
594,153,875 |
TypeScript
|
Use a data object instead of function closures for oldPrograms
|
## Search Terms
structured clone algorithm, oldProgram, function closure
## Suggestion
Create some transformer that transforms a `Program` to an `OldProgram`, which, instead of using function closures, would store the data, necessary for "rehydrating" a program, directly on the `OldProgram` object.
The `createProgram` function would then take a `Program|OldProgram` as the parameter for `oldProgram`.
The `OldProgram` object should be compatible with the [structured clone algorithm](https://developer.mozilla.org/en-US/docs/Web/API/Web_Workers_API/Structured_clone_algorithm), which would involve creating similar "serialized++" versions of all of the transitive objects contained in the data of `OldProgram`
## Use Cases & Motivation
When caching a `Program` for use in a later compiler step, it contains many functions not consumed when taking in an `oldProgram`, these functions are needlessly holding variables in their closure/memory.
By trimming the methods down to only those used, the memory footprint is reduced, but they still have a memory cost of about 15MB for most of our projects (this number calcuated by observing ~2300MB freed after deleting 150 Programs from a cache). In our case this becomes ~22GB for all our projects. It would be nice to create a more lightweight version of these that can still be efficiently rehydrated for quick compiler actions.
This could also allow users in a specific use case to trim down the data more than is prudent to allow them to cache all of the Programs.
As an added bonus, if the `OldProgram` type could be made compatible with the structured clone algorithm, then it would be much easier to share a single, much larger, cache between a dozen-or-so compiler hosts.
As noted in #37701, compile times are slow when not able to reuse an oldProgram, this is aimed at helping users to be able to reuse an oldProgram more often.
|
Suggestion,In Discussion
|
low
|
Major
|
594,211,043 |
rust
|
BinaryHeap::peek_mut borrow check error (previously worked with -Z polonius)
|
The code below ([playground](https://play.rust-lang.org/?version=nightly&mode=debug&edition=2018&gist=8129d2dbcf17e4027fe851e877a31988)) gives an apparently spurious borrow check error. Intuitively, I would expect the first mutable borrow of `heap` to end when `p` is consumed by `PeekMut::pop`. I probably wouldn’t be filing this as a bug except that `rustc -Z polonius` agreed with my intuition until a few days ago.
Works: rustc 1.44.0-nightly (699f83f52 2020-03-29) with `-Z polonius`
Fails: rustc 1.44.0-nightly (211365947 2020-03-30) with `-Z polonius`
[Commits in `699f83f52`...`211365947`](https://github.com/rust-lang/rust/compare/699f83f525c985000c1f70bf85117ba383adde87...2113659479a82ea69633b23ef710b58ab127755e)
```rust
use std::collections::binary_heap::{BinaryHeap, PeekMut};
fn main() {
let mut heap = BinaryHeap::from(vec![1, 2, 3]);
if let Some(p) = heap.peek_mut() {
PeekMut::pop(p);
heap.push(4);
};
}
```
```
error[E0499]: cannot borrow `heap` as mutable more than once at a time
--> src/main.rs:7:9
|
5 | if let Some(p) = heap.peek_mut() {
| ---------------
| |
| first mutable borrow occurs here
| a temporary with access to the first borrow is created here ...
6 | PeekMut::pop(p);
7 | heap.push(4);
| ^^^^ second mutable borrow occurs here
8 | };
| - ... and the first borrow might be used here, when that temporary is dropped and runs the destructor for type `std::option::Option<std::collections::binary_heap::PeekMut<'_, i32>>`
```
### Meta
<!--
If you're using the stable version of the compiler, you should also check if the
bug also exists in the beta or nightly versions.
-->
`rustc --version --verbose`:
```
rustc 1.44.0-nightly (853c4774e 2020-04-04)
binary: rustc
commit-hash: 853c4774e26ea97b45fe74de9a6f68e526784323
commit-date: 2020-04-04
host: x86_64-unknown-linux-gnu
release: 1.44.0-nightly
LLVM version: 9.0
```
|
A-borrow-checker,T-compiler,C-bug,NLL-polonius
|
low
|
Critical
|
594,252,932 |
create-react-app
|
Line numbers reported in TypeScript syntax errors can be incorrect
|
Sometimes syntax errors have the wrong line numbers.
Needless to say, this is confusing.
### Environment
```
Environment Info:
current version of create-react-app: 3.4.1
running from /Users/sophiebits/.npm/_npx/57068/lib/node_modules/create-react-app
System:
OS: macOS High Sierra 10.13.6
CPU: (12) x64 Intel(R) Core(TM) i7-8850H CPU @ 2.60GHz
Binaries:
Node: 8.12.0 - ~/.nvm/versions/node/v8.12.0/bin/node
Yarn: 1.16.0 - /usr/local/bin/yarn
npm: 6.4.1 - ~/.nvm/versions/node/v8.12.0/bin/npm
Browsers:
Chrome: 80.0.3987.162
Firefox: 71.0
Safari: 13.1
npmPackages:
react: ^16.13.1 => 16.13.1
react-dom: ^16.13.1 => 16.13.1
react-scripts: 3.4.1 => 3.4.1
npmGlobalPackages:
create-react-app: Not Found
```
### Steps to reproduce
<!--
How would you describe your issue to someone who doesn’t know you or your project?
Try to write a sequence of steps that anybody can repeat to see the issue.
-->
(Write your steps here:)
1. Run `npx create-react-app --template typescript bug`
2. Replace `bug/src/App.tsx` with the following:
```
const x = useReducer<A, B>((state, action) => {
here();
are();
some();
lines();
to();
demonstrate();
the();
line();
number();
can();
be();
way();
off();
case 'hello':
});
```
### Expected behavior
When loading the project in a browser, expected the error message:
```
Unexpected token (15:2)
13 | way();
14 | off();
> 15 | case 'hello':
| ^
16 | });
17 |
```
### Actual behavior
Got the error message:
```
./src/App.tsx
SyntaxError: /path/to/src/App.tsx: Unexpected token, expected ";" (1:25)
> 1 | const x = useReducer<A, B>((state, action) => {
| ^
2 | here();
3 | are();
4 | some();
```
NB: `yarn tsc` does output the right line number in this case, so I don't think it's a TS bug:
```
$ yarn tsc
src/App.tsx:15:3 - error TS1128: Declaration or statement expected.
15 case 'hello':
```
Interestingly, the bug goes away if you remove the type param `<A, B>`.
### Reproducible demo
https://github.com/sophiebits/cra-typescript-line-numbers-bug
|
issue: needs investigation,issue: bug report
|
low
|
Critical
|
594,258,068 |
flutter
|
AnnotatedRegion not working on iOS (to set status bar icon color)
|
Internal: b/153083915 [Videos in attached bug]
I am currently trying to use the suggested AnnotatedRegion widget in order to control the status bar icon color.
Basically, after going to a new page which sets the status bar icon color, then going back to the original page, the original page status bar icon color is not reverted to the original one set by the AnnotatedRegion widget.
To reproduce:
```dart
import 'package:flutter/material.dart';
import 'package:flutter/services.dart';
void main() {
runApp(MaterialApp(
title: 'Named Routes Demo',
// Start the app with the "/" named route. In this case, the app starts
// on the FirstScreen widget.
initialRoute: '/',
routes: {
// When navigating to the "/" route, build the FirstScreen widget.
'/': (context) => FirstScreen(),
// When navigating to the "/second" route, build the SecondScreen widget.
'/second': (context) => SecondScreen(),
},
));
}
class FirstScreen extends StatelessWidget {
@override
Widget build(BuildContext context) {
final systemUiOverlayStyle = const SystemUiOverlayStyle(
statusBarColor: Colors.transparent,
statusBarIconBrightness: Brightness.dark,
);
return AnnotatedRegion<SystemUiOverlayStyle>(
value: systemUiOverlayStyle,
child: Container(
color: Colors.white,
child: Center(
child: RaisedButton(
child: Text('Launch screen'),
onPressed: () {
// Navigate to the second screen using a named route.
Navigator.pushNamed(context, '/second');
},
),
),
),
);
}
}
class SecondScreen extends StatelessWidget {
@override
Widget build(BuildContext context) {
return Scaffold(
appBar: AppBar(
brightness: Brightness.dark,
backgroundColor: Colors.black,
title: Text("Second Screen"),
),
body: Center(
child: RaisedButton(
onPressed: () {
// Navigate back to the first screen by popping the current route
// off the stack.
Navigator.pop(context);
},
child: Text('Go back!'),
),
),
);
}
}
```
It only occurs on iOS, on Android it works as expected (the status bar icon color reverts to original color when returning to the original page).
|
platform-ios,framework,customer: money (g3),a: layout,has reproducible steps,P2,found in release: 3.3,found in release: 3.7,team-ios,triaged-ios
|
low
|
Critical
|
594,269,964 |
pytorch
|
torch.multinomial is misnamed.
|
The [multinomial distribution](https://en.wikipedia.org/wiki/Multinomial_distribution) has two inputs: weights <i><b>p</b></i> and the number of trials <i>N</i>, and one output the number of results in each category <b>n</b>. Note that the concept of with or without replacement doesn't enter the picture, and the output as the same shape as the weights, but with a different data type (an integer that sums to <i>N</i> along the axis that the weights sum to 1).
There is a distribution that related to the multinomial distribution by the question of replacement, and that's the [multi-variate hypergeometric distribution](https://en.wikipedia.org/wiki/Hypergeometric_distribution#Multivariate_hypergeometric_distribution), but that, too, will have an output with the same shape as weights (and weights will have to contain integers, instead of floats).
The hypergeometric distribution is about drawing a fixed number of objects from a population that has a fixed number of each type of object without replacing between individual draw, and then counting how many of each type you drew. A multinomial distribution can be gotten from the same process where the objects are replaced after each draw.
Regardless, that's not what torch.multinomial is returning. When you're returning the result of each draw, you're repeatedly sampling either the [categorical distribution](https://en.wikipedia.org/wiki/Categorical_distribution) (the with replacement variant), or a distribution I cannot find a name for (draw?). Although the constraint that the number of draws be less than the number of categories means that the without replacement variant doesn't correspond to any probability distribution I know of, at all.
So, I would suggest renaming torch.multinomial to torch.categorical.
cc @vincentqb @fritzo @neerajprad @alicanb @vishwakftw @nikitaved
|
module: distributions,triaged,module: ux
|
low
|
Minor
|
594,281,795 |
vue
|
.once modifier did'nt perform as expected if my event handler return null
|
### Version
2.6.11
### Reproduction link
[https://jsfiddle.net/q0v6m8dr/1/](https://jsfiddle.net/q0v6m8dr/1/)
### Steps to reproduce
1. bind a handler to an event which with a '.once' modifier
3. set null for the hanlder's return value
2. trigger the event, and trigger again
### What is expected?
only the first trigger will invoke handler, under '.once' influence
### What is actually happening?
subsequent trigger will also invoke handler beside the first one
<!-- generated by vue-issues. DO NOT REMOVE -->
|
improvement
|
low
|
Minor
|
594,287,465 |
go
|
x/image/tiff: add 32bit float grayscale support
|
Please add support for 32bit float grayscale tiff images. This is a common format in scientific and medical imaging.
|
NeedsInvestigation,FeatureRequest
|
low
|
Major
|
594,350,903 |
godot
|
[TRACKER] List of unused signals
|
**Godot version:**
4.0.dev.custom_build. 321ce4d4c
**Issue description:**
With help of https://github.com/qarmin/godot_signal_checker (which could be rewritten to python to be able to put it inside CI in Godot repository) I found lot of potential unused signals with this methodology:
Added signals have at the beginning of line `ADD_SIGNAL(MethodInfo(\"`
Emitted signals have at the beginning of line `emit_signal(CoreStringNames::get_singleton()->`, `emit_signal(SceneStringNames::get_singleton()->` or `emit_signal(\"`
Connected signals have at the beginning of line `->connect(\"` or `connect_compat(\"`
**Signals which are emitted and connected but never added**
- [x] - dpi_changed
- [x] - drop_attempted
- [x] - modal_closed (#41920)
**Signals which are added but never emitted or connected**
- [x] - bake_finished
- [x] - clear_request
- [x] - copy_request
- [x] - item_custom_button_pressed
- [x] - item_group_status_changed
- [x] - item_lock_status_changed
- [x] - pause_pressed
- [x] - search
**Signals which are added and connected but never emitted**
- [ ] - autoload_changed
- [ ] - export_presets_updated
- [ ] - group_edited
- [x] - inputmap_changed
- [ ] - localization_changed
- [ ] - multiple_properties_changed
- [ ] - node_added
- [ ] - node_removed
- [ ] - preview_invalidated
- [ ] - property_edited
- [x] - removed_from_graph
|
bug,topic:core,topic:editor,tracker
|
low
|
Major
|
594,356,205 |
rust
|
Consider renaming (`Re`)`EarlyBound` to `ReParam` and (`Re`)`LateBound` to `Bound`.
|
This convention would be slightly less confusing due to aligning better with `Ty` and `ty::Const`.
On the other hand, the early-bound vs late-bound terminology is so baked-in that this might be hard.
cc @rust-lang/wg-traits @matthewjasper
|
C-cleanup,A-type-system,A-lifetimes,T-compiler,T-types
|
low
|
Minor
|
594,455,718 |
pytorch
|
[JIT] Huge delay (1274s vs 0.031s) when running scripted model
|
## 🐛 Bug
When a model that worked properely on a pure python is scripted, it works with a huge delay (1274s vs 0.031s before scripting), while the result at last is the same.
Below are timings for processing 2 copies of the same image. Note that sometimes it is the 1st image that is processed very slowly, while sometimes it is the 2nd one. I've tested 2 models: the simple, wherer delay is about 10s vs 0.008s and more complex where delay is about 1000s vs 0.056s, tests were done on 2 environments: latest nightly pytorch 1.5.0a0+2dc2933 on 1080Ti and on pytorch 1.4.0 on NVIDIA 1050.
Results:
1.5.0a0+2dc2933 on 1080Ti
cfg/yolov3-tiny.cfg weights/yolov3-tiny.pt (more simple model)
Model Summary: 37 layers, 8.85237e+06 parameters, 8.85237e+06 gradients
before scripting:
image 1/2 data\samples\TEST_AO_2_2_4300001.jpeg: Done. (0.748s)
image 2/2 data\samples\TEST_AO_2_2_4300002.jpeg: Done. (0.008s) _(...to compare with...)_
after scripting:
image 1/2 data\samples\TEST_AO_2_2_4300001.jpeg: Done. (0.170s)
image 2/2 data\samples\TEST_AO_2_2_4300002.jpeg: Done. (11.391s) **(not good...)**
cfg/yolov3.cfg weights/yolov3.pt (more complex model)
Model Summary: 222 layers, 6.19491e+07 parameters, 6.19491e+07 gradients
before scripting:
image 1/2 data\samples\TEST_AO_2_2_4300001.jpeg: Done. (0.768s)
image 2/2 data\samples\TEST_AO_2_2_4300001copy.jpeg: Done. (0.031s) _(...to compare with...)_
after scripting:
image 1/2 data\samples\TEST_AO_2_2_4300001.jpeg: Done. (1.392s)
image 2/2 data\samples\TEST_AO_2_2_4300001copy.jpeg: Done. (579.728s) **(... terrible)**
1.4.0 on 1050:
cfg/yolov3-tiny.cfg weights/yolov3-tiny.pt (more simple)
Model Summary: 37 layers, 8.85237e+06 parameters, 8.85237e+06 gradients
before scripting:
image 1/2 data\samples\TEST_AO_2_2_4300001.jpeg: Done. (0.862s) _(...to compare with...)_
image 2/2 data\samples\TEST_AO_2_2_4300001copy.jpeg: Done. (0.014s)
after scripting:
image 1/2 data\samples\TEST_AO_2_2_4300001.jpeg: Done. (5.997s) **(not good...)**
image 2/2 data\samples\TEST_AO_2_2_4300001copy.jpeg: Done. (0.012s)
cfg/yolov3.cfg weights/yolov3.pt (more complex model)
Model Summary: 222 layers, 6.19491e+07 parameters, 6.19491e+07 gradients
before scripting:
image 1/2 data\samples\TEST_AO_2_2_4300001.jpeg: Done. (1.009s) _(...to compare with...)_
image 2/2 data\samples\TEST_AO_2_2_4300001copy.jpeg: Done. (0.056s)
after scripting:
image 1/2 data\samples\TEST_AO_2_2_4300001.jpeg: Done. (1274.444s) **(... terrible)**
image 2/2 data\samples\TEST_AO_2_2_4300001copy.jpeg: Done. (0.053s)
## To Reproduce
Sorry that I didn't produce more localized example.
Steps to reproduce the behavior:
1. git clone https://github.com/IlyaOvodov/yolov3.git
2. git checkout performance_problem_test
3. Download weights yolov3-tiny.pt (for simple model) and yolov3.pt (for complex one) from https://drive.google.com/drive/folders/1LezFG5g3BCW6iYaV89B2i64cqEUZD7e0
and place them into weights directory
4. python test_timing.py
This run a simple model test (with has a 5-10s delay)
For a complex model test uncomment line 26 in test_timing.py: `cfg, weights = ('cfg/yolov3.cfg', 'weights/yolov3.pt')`
## Environment
1.
PyTorch version: 1.5.0a0+2dc2933
Is debug build: No
CUDA used to build PyTorch: 10.0
OS: Microsoft Windows 10 Pro
GCC version: Could not collect
CMake version: version 3.12.2
Python version: 3.6
Is CUDA available: Yes
CUDA runtime version: 10.0.130
GPU models and configuration: GPU 0: GeForce GTX 1080 Ti
Nvidia driver version: 441.87
cuDNN version: C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\bin\cudnn64_7.dll
Versions of relevant libraries:
[pip3] numpy==1.17.2
[pip3] pytorch-ignite==0.2.0
[pip3] pytorch-toolbelt==0.2.1
[pip3] torch==1.5.0a0+2dc2933
[pip3] torchvision==0.4.1a0+d94043a
[conda] blas 1.0 mkl
[conda] cuda92 1.0 0 pytorch
[conda] mkl 2019.0 pypi_0 pypi
[conda] mkl-include 2019.0 pypi_0 pypi
[conda] mkl_fft 1.0.6 py36hdbbee80_0
[conda] mkl_random 1.0.1 py36h9258bd6_0
[conda] numpy 1.17.2 pypi_0 pypi
[conda] pytorch-ignite 0.2.0 dev_0 <develop>
[conda] pytorch-toolbelt 0.2.1 dev_0 <develop>
[conda] torch 1.5.0a0+2dc2933 dev_0 <develop>
[conda] torchvision 0.4.1a0+d94043a dev_0 <develop>
2.
Collecting environment information...
PyTorch version: 1.4.0
Is debug build: No
CUDA used to build PyTorch: 10.1
OS: Microsoft Windows 7 Џа®дҐббЁ® «м п
GCC version: Could not collect
CMake version: Could not collect
Python version: 3.6
Is CUDA available: Yes
CUDA runtime version: 10.0.130
GPU models and configuration: GPU 0: GeForce GTX 1050
Nvidia driver version: 441.22
cuDNN version: C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\bin\cudnn64_7.dll
Versions of relevant libraries:
[pip3] efficientnet-pytorch==0.6.3
[pip3] numpy==1.18.0
[pip3] torch==1.4.0
[pip3] torchvision==0.5.0
[conda] efficientnet-pytorch 0.6.3 <pip>
[conda] mkl 2019.0 <pip>
[conda] mkl-include 2019.0 <pip>
[conda] numpy 1.18.0 <pip>
[conda] torch 1.4.0 <pip>
[conda] torchvision 0.5.0 <pip>
cc @ezyang @gchanan @zou3519 @suo
|
high priority,oncall: jit,triaged
|
low
|
Critical
|
594,459,259 |
pytorch
|
[DISCUSSION] Better user experience for debugging on Windows
|
## Background
With https://github.com/pytorch/pytorch/pull/36039, we are now printing the traceback of exceptions in the command prompt. For example, before this change, if we throw an exception `c10::Error(SourceLocation, msg)`, it will only print something like `(no backtrace available)` But now we will get something below.
```cpp
00007FF6731716B9 00007FF6731716B0 bound_shape_inference [source.cpp @ 223]
00007FFBFAF3FAFF 00007FFBFAF3FA50 crt_at_quick_exit [C:\WINDOWS\System32\ucrtbase.dll @ <unknown line number>]
00007FFBFD1C7BD4 00007FFBFD1C7BC0 BaseThreadInitThunk [C:\WINDOWS\System32\KERNEL32.DLL @ <unknown line number>]
00007FFBFD32CED1 00007FFBFD32CEB0 RtlUserThreadStart [C:\WINDOWS\SYSTEM32\ntdll.dll @ <unknown line number>]
```
## Problem
However, even with https://github.com/pytorch/pytorch/pull/36039, it won't be that useful if PDBs are not distributed with the binaries. For example, you'll get something below without the PDBs.
```cpp
00007FF6731716B9 <unknown symbol address> <unknown symbol name> [source.exe @ <unknown line number>]
00007FFBFAF3FAFF 00007FFBFAF3FA50 crt_at_quick_exit [C:\WINDOWS\System32\ucrtbase.dll @ <unknown line number>]
00007FFBFD1C7BD4 00007FFBFD1C7BC0 BaseThreadInitThunk [C:\WINDOWS\System32\KERNEL32.DLL @ <unknown line number>]
00007FFBFD32CED1 00007FFBFD32CEB0 RtlUserThreadStart [C:\WINDOWS\SYSTEM32\ntdll.dll @ <unknown line number>]
```
So the major problem here are listed below:
1. Should we distribute the PDBs in the release / CI builds?
2. How do we distribute them?
## Plans
We definitely should distribute the PDBs when users need some debugging info. The possible options are listed below.
1. Distribute in the release builds. (use `/DEBUG:FULL` in the linker flags)
This is the easiest option, both for devs and users. However, the size of the binaries will be much larger. You may expect the package size to be nearly 1GB after the change.
2. Distribute them separately
It is based on the idea that when users needs debugging, they may download them and install them to the existing package. However, we may need to design the build job to generate them.
Any thoughts? cc @peterjc123 @malfet @ezyang
|
module: build,module: windows,triaged
|
low
|
Critical
|
594,464,609 |
godot
|
GDScript cannot always infer types
|
**Godot version:**
3.2.1
**OS/device including version:**
Windows 10
The gdscript editor cannot seem to infer object types most of the time making it impossible to get popup help for near enough all objects. Without this functionality, coding is near impossible, especially for newcomers or those wishing to ensure the method/constant they are typing is correct.
i.e. you expect on pressing '.' that the editor knows what the object type is and provide the correction information. Most other dynamically typed language editors provide this.
Example:
In the image below a collider was retrieve, yet on typing 'collider.' only top level functions were provided. It was only when the type was explicitly declared as with collider2, that the extra functions, specifically 'is_in_group' in my test case became available.
|
topic:gdscript
|
low
|
Major
|
594,482,184 |
create-react-app
|
create-react-app should allow TypeScript imports outside `src`
|
### Is your proposal related to a problem?
Code reuse is a good thing. It is not uncommon that a backend and a frontend can use the same code, and live in the same repository. The need for sharing code is particularly strong with TypeScript type definitions, because developers on the backend and the frontend want to make sure that they are building against a common types.
### Describe the solution you'd like
I don't know the CRA internals, but maybe one of the following:
- Allow `baseUrl` to point outside `.` to enable absolute imports like `import * as foo from 'common/foo``.
- Enable relative imports outside `src` like `import * as foo from '../../common/foo'`.
### Describe alternatives you've considered
There are [work-arounds](https://stackoverflow.com/a/61043925/1804173) using a combination of third-party cra-patching libraries. They are tedious to setup though (I worked 3 evenings on coming up with a solution), potentially fragile, and the use case seems so fundamental that I'm very surprised that it is not to supported out-of-the-box.
|
issue: proposal,needs triage
|
high
|
Critical
|
594,490,729 |
rust
|
Running rustdoc on code using quote generates false lint warnings for unused_braces
|
I tried this code:
```toml
[package]
name = "repro"
version = "0.1.0"
edition = "2018"
[dependencies]
quote = "=1.0.3"
```
```rust
use quote::quote;
pub fn repro() {
let many = [1, 2, 3];
let _together = quote! {
#(#many),*
};
}
```
I think this is a basic and common usage of the [quote](https://crates.io/crates/quote) crate.
I expected to see no warnings:
```
% RUSTFLAGS=-Dwarnings cargo +nightly-2020-04-04 build
Finished dev [unoptimized + debuginfo] target(s) in 0.01s
% RUSTFLAGS=-Dwarnings cargo +nightly-2020-04-04 doc
Documenting repro v0.1.0 (/private/tmp/repro)
Finished dev [unoptimized + debuginfo] target(s) in 0.82s
```
Instead, warnings are generated (and my CI build fails due to the `deny`):
```
% RUSTFLAGS=-Dwarnings cargo +nightly-2020-04-04 build
Finished dev [unoptimized + debuginfo] target(s) in 0.01s
% RUSTFLAGS=-Dwarnings cargo +nightly-2020-04-04 doc
Documenting repro v0.1.0 (/private/tmp/repro)
warning: unnecessary braces around block return value
|
= note: `#[warn(unused_braces)]` on by default
Finished dev [unoptimized + debuginfo] target(s) in 0.82s
```
Note that the warning only appears when running rustdoc.
### Meta
`rustc +nightly-2020-04-04 --version --verbose`:
```
rustc 1.44.0-nightly (74bd074ee 2020-04-03)
binary: rustc
commit-hash: 74bd074eefcf4915c73d1ab91bc90859664729e6
commit-date: 2020-04-03
host: x86_64-apple-darwin
release: 1.44.0-nightly
LLVM version: 9.0
```
|
A-lints,P-medium,T-compiler,regression-from-stable-to-stable,C-bug,ICEBreaker-Cleanup-Crew
|
medium
|
Critical
|
594,494,222 |
go
|
encoding/xml: XMLName value of embedded structs type not used in Marshaller
|
### What version of Go are you using (`go version`)?
<pre>
$ go version
go version go1.14.1 windows/amd64
</pre>
### Does this issue reproduce with the latest release?
Yes
### What operating system and processor architecture are you using (`go env`)?
<details><summary><code>go env</code> Output</summary><br><pre>
$ go env
set GO111MODULE=
set GOARCH=amd64
set GOBIN=
set GOCACHE=C:\Users\Constantine\AppData\Local\go-build
set GOENV=C:\Users\Constantine\AppData\Roaming\go\env
set GOEXE=.exe
set GOFLAGS=
set GOHOSTARCH=amd64
set GOHOSTOS=windows
set GOINSECURE=
set GONOPROXY=
set GONOSUMDB=
set GOOS=windows
set GOPATH=D:\Develop\golib
set GOPRIVATE=
set GOPROXY=https://proxy.golang.org,direct
set GOROOT=D:\Develop\go1.14.1
set GOSUMDB=sum.golang.org
set GOTMPDIR=
set GOTOOLDIR=D:\Develop\go1.14.1\pkg\tool\windows_amd64
set GCCGO=gccgo
set AR=ar
set CC=gcc
set CXX=g++
set CGO_ENABLED=1
set GOMOD=
set CGO_CFLAGS=-g -O2
set CGO_CPPFLAGS=
set CGO_CXXFLAGS=-g -O2
set CGO_FFLAGS=-g -O2
set CGO_LDFLAGS=-g -O2
set PKG_CONFIG=pkg-config
set GOGCCFLAGS=-m64 -mthreads -fmessage-length=0 -fdebug-prefix-map=C:\Users\CONSTA~1\AppData\Local\Temp\go-build125187114=/tmp/go-build -gno-record-gcc-switches
</pre></details>
### What did you do?
Try to use XMLName in embedded type [play](https://play.golang.org/p/UyMZcLQG8R5) [stackoverflow](https://stackoverflow.com/questions/61041521/why-xmlname-value-of-embedded-structs-type-not-used-in-marshaller)
<pre>
type XMLForm struct {
XMLName xml.Name
}
type Test struct {
XMLForm
}
var t Test
t.XMLName.Local = "c:test"
xml.Marshal(&t)
</pre>
### What did you expect to see?
`<c:test></c:test>`
### What did you see instead?
`<Test></Test>`
It fails on go1.14.1\src\encoding\xml\marshal.go: can't cast to xml.Name
```
482: } else if v, ok := xmlname.value(val).Interface().(Name); ok && v.Local != "" {
483: start.Name = v
484: }
```
I also find a reason for this error https://golang.org/src/encoding/xml/typeinfo.go
Need to add idx of embed type to slice
```
78: if tinfo.xmlname == nil {
++ && inner.xmlname != nil {
79: tinfo.xmlname = inner.xmlname
++ tinfo.xmlname.idx = append([]int{i}, inner.xmlname.idx...)
80: }
```
|
NeedsInvestigation
|
low
|
Critical
|
594,513,291 |
electron
|
MacOS: "Ask for a Review" functionality using SKStoreReviewController API
|
Update: you can find a paid bounty for this issue here: https://bountify.co/electron-call-macos-requestreview-api
First working solution gets the bounty. Payout will continue to increase until a solution is found. I will make it public domain if the submitter doesn't.
---
For electron on MacOs, how do we call the "requestReview()" method in Apple's native SKStoreReviewController API?
In recent MacOS and iOS apps, Apple gave developers an easy way to just "ask for a review" using a simple requestReview() API call. It's an absolutely necessary feature, since getting ratings and reviews can be the difference-maker between a successful app and a failure. Apple has made it extremely seamless for users to rate your app in a single touch without leaving your app and going to the app store, which has been shown to make a significant impact between getting app reviews or not. However I have no idea how to implement it using Electron. It's such an essential feature in 2020.
> You can ask users to rate and review your app at appropriate times throughout the user experience. Make the request when users are most likely to feel satisfaction with your app, such as when they’ve completed an action, level, or task. Make sure not to interrupt their activity.
> To give users an easy way to provide feedback on the App Store or the Mac App Store, use the SKStoreReviewController API. You can prompt for ratings up to three times in a 365-day period. Users will submit a rating through the standardized prompt, and can write and submit a review without leaving the app.
>
It's a simple API call with not much back and forth needed, and then you have seamless app review functionality. How do we implement this? It can be the difference maker between Electron being a successful app development framework for Mac App Store developers.
You can read more about the SKStoreReviewController API here:
https://developer.apple.com/app-store/ratings-and-reviews/
For example, if this is something that can be done using node's child_process.execFile, is there any documentation that shows where to store compiled native code inside the .App and call them from electron using node's child process?
|
enhancement :sparkles:,platform/macOS
|
medium
|
Critical
|
594,546,665 |
godot
|
Switching views overlays 3D image on 2D view
|
Current master branch: 94fab213482fd944debc50e255e7b395d391cc33
Windows 10 (latest NVIDIA drivers)
As in GIF. Of course that's not proper use of views, nodes I've put are for 3D usage, but I can't recall that previous versions preserved similar behavior of caching 3D image on top of 2D view.
|
bug,topic:rendering,topic:editor
|
low
|
Minor
|
594,581,603 |
youtube-dl
|
operadeparis.fr
|
<!--
######################################################################
WARNING!
IGNORING THE FOLLOWING TEMPLATE WILL RESULT IN ISSUE CLOSED AS INCOMPLETE
######################################################################
-->
## Checklist
<!--
Carefully read and work through this check list in order to prevent the most common mistakes and misuse of youtube-dl:
- First of, make sure you are using the latest version of youtube-dl. Run `youtube-dl --version` and ensure your version is 2020.03.24. If it's not, see https://yt-dl.org/update on how to update. Issues with outdated version will be REJECTED.
- Make sure that all provided video/audio/playlist URLs (if any) are alive and playable in a browser.
- Make sure that site you are requesting is not dedicated to copyright infringement, see https://yt-dl.org/copyright-infringement. youtube-dl does not support such sites. In order for site support request to be accepted all provided example URLs should not violate any copyrights.
- Search the bugtracker for similar site support requests: http://yt-dl.org/search-issues. DO NOT post duplicates.
- Finally, put x into all relevant boxes (like this [x])
-->
- [x] I'm reporting a new site support request
- [x] I've verified that I'm running youtube-dl version **2020.03.24**
- [x] I've checked that all provided URLs are alive and playable in a browser
- [x] I've checked that none of provided URLs violate any copyrights
- [x] I've searched the bugtracker for similar site support requests including closed ones
## Example URLs
<!--
Provide all kinds of example URLs support for which should be included. Replace following example URLs by yours.
-->
- Single video: https://www.operadeparis.fr/magazine/le-lac-des-cygnes-replay
## Description
<!--
Provide any additional information.
If work on your issue requires account credentials please provide them or explain how one can obtain them.
-->
WRITE DESCRIPTION HERE
The page contains a link to a youtube trailer, that youtube-dl finds
The page contains also a 2-hour video hosted by akamai that youtube-dl does not find. The video tag seems to be
<video playsinline="true" style="width: 100%; height: 100%; position: absolute; background-color: black;" muted="false" src="blob:https://embedftv-a.akamaihd.net/364fc357-3dc7-df49-a398-3f64fc1a7829"></video>
Note: this page will be soon disabled but please provide support for similar pages.
|
site-support-request
|
low
|
Critical
|
594,610,642 |
rust
|
Wasteful duplication between incremental build dirs and normal artifacts.
|
Incremental compilation reuses object files by storing *a copy* of them in the incremental cache directory, which is then copied into the `rlib`.
This results to build dirs which are twice as big as they need to be (which then gets multiplied by the number of stale artifacts that aren't removed by any tool AFAICT).
There are several ways we could resolve this, but they can be split into two categories:
* incremental cache holds the object files
* Cargo could have a mode in which it instructs `rustc` to emit object files but only in the incremental cache and then downstream `rustc` to use that incremental cache
* this would allow non-Cargo tooling to keep working, but we'd have to support `rlib`s as the same time as the new system and we might unknowingly break them if Cargo doesn't use them
* the `rlib` *doesn't have to be* a real archive, we could have a different format that references the files in the incremental cache, or even make the `rlib` a directory full of hardlinks, or a hardlink itself etc.
* `rlib` artifact holds the object files
* not clear what's possible at all for other crate types, for now they'd keep the duplication
* we probably want to hardlink/symlink it from the incremental cache and hash each object file so we can check it's still the same and only reuse it then
* we might already be doing that hashing anyway, in case the incremental dir is corrupted
* requires no change in tooling AFAICT, including custom non-Cargo setups
cc @rust-lang/compiler
|
C-enhancement,T-compiler,A-incr-comp,I-heavy
|
low
|
Minor
|
594,614,832 |
react
|
Provide a renderer-agnostic equivalent of setNativeProps()
|
Dan asked me to open up an issue: https://twitter.com/dan_abramov/status/1246883821477339139
My proposal is to extend React with a small hook that allows us to mutate nodes without causing render. React has no official means to deal with fast occurring updates and libraries like react-spring and framer-motion already do something similar but in a way that forces them to carry a lot of burden.
```jsx
import React, { useMutation }
function A() {
const [specialRef, set] = useMutation()
useEffect(() => {
// the following would execute sync and without causing render
// going through the same channel as a regular props update with all
// the internal interpolation (100 --> "100px")
set({ style: { left: 100 } })
}, [])
return <div ref={specialRef} ... />
```
It uses the fact that reconcilers know how to handle props, something we don't know in userland unless we cause render to set fresh props, which is not at all optimal for animation or anything frame based. react-dom for instance knows what `margin: 3px` is, react-three-fiber knows what `position: [1,2,3]` is, and so on. These details are defined in the reconciler:
```jsx
commitUpdate(instance: any, updatePayload: any, type: string, oldProps: any, newProps: any, fiber: Reconciler.Fiber)
```
If libraries could use this knowledge from outside they could deal with any platform. Animation libraries like react-spring or framer-motion would turn x-platform in one strike, they could animate everything: dom nodes, react native views, meshes, hardware diodes. We could finally write libraries that are not reliant on platforms.
|
Type: Feature Request,Type: Discussion
|
medium
|
Critical
|
594,618,948 |
flutter
|
Flutter Build iOS Fails - /tmp/Runner.dst/Applications/Runner.app does not exist
|
## Command
flutter build iOS fails. I search the Internet but unable to find a solution. I have attached a build fail.txt file located at
https://github.com/churchevent/ChurchEvents/issues/1#issue-594615884
The last thing I did before building the project was 'Clean Build Folder' from Xcode, and 'Flutter Clean' from the command line.
## Logs
```
_Exception: Exception: Source directory "/tmp/Runner.dst/Applications/Runner.app" does not exist, nothing to copy
#0 FileSystemUtils.copyDirectorySync (package:flutter_tools/src/base/file_system.dart:61:7)
#1 buildXcodeProject (package:flutter_tools/src/ios/mac.dart:388:23)
<asynchronous suspension>
#2 BuildIOSCommand.runCommand (package:flutter_tools/src/commands/build_ios.dart:83:43)
#3 _rootRunUnary (dart:async/zone.dart:1192:38)
#4 _CustomZone.runUnary (dart:async/zone.dart:1085:19)
#5 _FutureListener.handleValue (dart:async/future_impl.dart:141:18)
#6 Future._propagateToListeners.handleValueCallback (dart:async/future_impl.dart:682:45)
#7 Future._propagateToListeners (dart:async/future_impl.dart:711:32)
#8 Future._completeWithValue (dart:async/future_impl.dart:526:5)
#9 _AsyncAwaitCompleter.complete (dart:async-patch/async_patch.dart:36:15)
#10 _completeOnAsyncReturn (dart:async-patch/async_patch.dart:306:13)
#11 ApplicationPackageStore.getPackageForPlatform (package:flutter_tools/src/application_package.dart)
#12 _rootRunUnary (dart:async/zone.dart:1192:38)
#13 _CustomZone.runUnary (dart:async/zone.dart:1085:19)
#14 _FutureListener.handleValue (dart:async/future_impl.dart:141:18)
#15 Future._propagateToListeners.handleValueCallback (dart:async/future_impl.dart:682:45)
#16 Future._propagateToListeners (dart:async/future_impl.dart:711:32)
#17 Future._completeWithValue (dart:async/future_impl.dart:526:5)
#18 _AsyncAwaitCompleter.complete (dart:async-patch/async_patch.dart:36:15)
```
```
[✓] Flutter (Channel master, v1.18.1-pre.10, on Mac OS X 10.15.3 19D76, locale en-US)
• Flutter version 1.18.1-pre.10 at /Users/carlton/tools/flutter
• Framework revision 6d52bf9884 (46 minutes ago), 2020-04-05 13:21:01 -0400
• Engine revision 2cc6a6d66d
• Dart version 2.8.0 (build 2.8.0-dev.20.0 80ae6ed91d)
[✓] Android toolchain - develop for Android devices (Android SDK version 29.0.3)
• Android SDK at /Users/carlton/Library/Android/sdk
• Platform android-29, build-tools 29.0.3
• Java binary at: /Applications/Android Studio.app/Contents/jre/jdk/Contents/Home/bin/java
• Java version OpenJDK Runtime Environment (build 1.8.0_212-release-1586-b4-5784211)
• All Android licenses accepted.
[✓] Xcode - develop for iOS and macOS (Xcode 11.3.1)
• Xcode at /Applications/Xcode.app/Contents/Developer
• Xcode 11.3.1, Build version 11C505
• CocoaPods version 1.9.1
[✓] Android Studio (version 3.6)
• Android Studio at /Applications/Android Studio.app/Contents
• Flutter plugin version 44.0.2
• Dart plugin version 192.7761
• Java version OpenJDK Runtime Environment (build 1.8.0_212-release-1586-b4-5784211)
[✓] VS Code (version 1.43.2)
• VS Code at /Applications/Visual Studio Code.app/Contents
• Flutter extension version 3.9.0
[✓] Connected device (2 available)
• iPhone 11 Pro Max • DAD1B264-6F9C-4FAF-B7F0-122130454D04 • ios • com.apple.CoreSimulator.SimRuntime.iOS-13-3
(simulator)
• iPhone 11 • 274CAF9A-91C7-4455-AA65-B69DB6177000 • ios • com.apple.CoreSimulator.SimRuntime.iOS-13-0
(simulator)
• No issues found!
```
## Flutter Application Metadata
```
**Type**: app
**Version**: 2.0.2+4
**Material**: true
**Android X**: false
**Module**: false
**Plugin**: false
**Android package**: null
**iOS bundle identifier**: null
**Creation channel**: stable
**Creation framework version**: 68587a0916366e9512a78df22c44163d041dd5f3
### Plugins
flutter_keyboard_visibility-0.7.0
flutter_local_notifications-1.2.2
flutter_plugin_android_lifecycle-1.0.5
geolocator-5.2.1
google_api_availability-2.0.2
google_maps_flutter-0.5.22+3
image_picker-0.6.3+1
location_permissions-2.0.4+1
path_provider-1.6.0
shared_preferences-0.5.6+1
shared_preferences_macos-0.0.1+4
shared_preferences_web-0.1.2+3
url_launcher-5.4.2
url_launcher_macos-0.0.1+4
url_launcher_web-0.1.1+1
webview_media-0.1.1+3
```
|
c: crash,platform-ios,tool,t: xcode,P3,team-ios,triaged-ios
|
low
|
Minor
|
594,673,423 |
godot
|
Exported variable setter function called during typing
|
**Godot version:**
3.2.1 stable
**OS/device including version:**
Windows 10 18363
**Issue description:**
If an exported variable has a a setter function and its script is running in tool mode, the setter function is called every time the user types a new letter. For example, typing the value "icon" calls the setter function with inputs "i", "ic", "ico" and "icon".
Maybe this is the intended behaviour, but in my opinion, the function should only be called after the user presses Enter or the input box loses focus.
I prepared a minimal project that has three sprites, "icon.png", "icon1.png" and "icon2.png". Also, the Sprite node has an exported variable called "Icon Name", which defines the icon the Sprite should load. Typing "icon1" on the input throws 3 errors, because "i.png", "ic.png" and "ico.png" do not exist, then the script loads "icon.png", which exists, and finally loads the correct texture "icon1.png".
**Steps to reproduce:**
1. Open minimal reproduction project
2. Open Node2D.tscn scene
3. Click the Sprite node
4. Type "icon1" in the Icon Name exported variable.
5. Open the Output console and look how Godot tried to load "i.png", "ic.png" and "ico.png" before loading "icon.png", which doesn't throw an error and then "icon1.png".
**Minimal reproduction project:**
[ExportSetter-BugDemo.zip](https://github.com/godotengine/godot/files/4434989/ExportSetter-BugDemo.zip)
|
discussion,topic:editor
|
low
|
Critical
|
594,704,933 |
flutter
|
Overscrolling outer scrollview / SliverAppBar.stretch does not work in NestedScrollView
|
<details>
<summary>code sample</summary>
```bash
DefaultTabController(
length: 2,
child: NestedScrollView(
physics: BouncingScrollPhysics(),
headerSliverBuilder: (context, innerScrolled) => <Widget>[
SliverOverlapAbsorber(
handle: NestedScrollView.sliverOverlapAbsorberHandleFor(context),
child: SliverAppBar(
pinned: true,
stretch: true,
title: Text('username'),
expandedHeight: 325,
flexibleSpace: FlexibleSpaceBar(
stretchModes: <StretchMode>[
StretchMode.zoomBackground,
StretchMode.blurBackground,
],
background: Image.asset('images/splash.png', fit: BoxFit.cover)
),
bottom:
TabBar(
tabs: <Widget>[
Text('test1'),
Text('test2')
]
)
),
)
],
body: TabBarView(
children: [
Builder(
builder: (context) => CustomScrollView(
slivers: <Widget>[
SliverOverlapInjector(
handle: NestedScrollView.sliverOverlapAbsorberHandleFor(context)),
SliverFixedExtentList(
delegate: SliverChildBuilderDelegate(
(_, index) => Text('not working'),
childCount: 100),
itemExtent: 25
)
],
),
),
Text('working')
]
)));
```
</details>
<details>
<summary>flutter doctor -v</summary>
```bash
[✓] Flutter (Channel stable, v1.12.13+hotfix.9, on Mac OS X 10.15.1 19B88, locale en-CN)
• Flutter version 1.12.13+hotfix.9 at /Users/lalawila/flutter
• Framework revision f139b11009 (6 days ago), 2020-03-30 13:57:30 -0700
• Engine revision af51afceb8
• Dart version 2.7.2
[!] Android toolchain - develop for Android devices (Android SDK version 29.0.2)
• Android SDK at /Users/lalawila/Library/Android/sdk
• Android NDK location not configured (optional; useful for native profiling support)
• Platform android-29, build-tools 29.0.2
• Java binary at: /Library/Java/JavaVirtualMachines/jdk-10.0.2.jdk/Contents/Home/bin/java
• Java version Java(TM) SE Runtime Environment 18.3 (build 10.0.2+13)
✗ Android license status unknown.
Try re-installing or updating your Android SDK Manager.
See https://developer.android.com/studio/#downloads or visit https://flutter.dev/setup/#android-setup for detailed instructions.
[✓] Xcode - develop for iOS and macOS (Xcode 11.2.1)
• Xcode at /Applications/Xcode.app/Contents/Developer
• Xcode 11.2.1, Build version 11B53
• CocoaPods version 1.8.0
[!] Android Studio (not installed)
• Android Studio not found; download from https://developer.android.com/studio/index.html
(or visit https://flutter.dev/setup/#android-setup for detailed instructions).
[✓] VS Code (version 1.43.2)
• VS Code at /Applications/Visual Studio Code.app/Contents
• Flutter extension version 3.9.0
[✓] Connected device (1 available)
• iPhone 11 • 0BFEAC24-164A-4F59-B67D-EBDBF1A499D1 • ios • com.apple.CoreSimulator.SimRuntime.iOS-13-2 (simulator)
! Doctor found issues in 2 categories.
```
</details>
|
framework,f: material design,f: scrolling,has reproducible steps,P2,workaround available,team-design,triaged-design,found in release: 3.16,found in release: 3.20
|
low
|
Critical
|
594,710,930 |
node
|
Always call Socket#setTimeout in HTTP ClientRequest#setTimeout
|
<!--
Thank you for suggesting an idea to make Node.js better.
Please fill in as much of the template below as you're able.
-->
**Is your feature request related to a problem? Please describe.**
Hello there.
Currently, `ClientRequest#setTimeout` and `Socket#setTimeout` has different behavior.
`ClientRequest#setTimeout` calls `Socket#setTimeout` right after connection has established.
I am not sure it is intended or not, but it's impossible to detect socket inactivity before connection establishment when using `ClientRequest#setTimeout`. It makes users harder to handle connect timeout.
**Describe the solution you'd like**
Always call `Socket#setTimeout` regardless of connection establishment.
**Describe alternatives you've considered**
None.
|
http
|
low
|
Minor
|
594,840,066 |
pytorch
|
Half type promotion with Numpy arrays is incorrect
|
Sample script:
```
import torch
t = torch.tensor((1,), dtype=torch.float16)
arr = np.array((1,), dtype=np.int32)
print(t + arr)
```
Result:
`tensor([2.], dtype=torch.float64)`
The result of these operations should be the widest float type. E.g. np.int64 + torch.float16 should produce a torch.float16 tensor, and np.float32 + torch.float16 should produce a torch.float32 tensor. Symmetry should apply, too: torch.float16 + np.int64 should produce a torch.float16 tensor.
Half behavior is incorrect when interacting with the following types: np.bool, np.uint8, np.int8, np.int16, np.int32, np.int64, np.float16.
See related testing PR https://github.com/pytorch/pytorch/pull/35945.
|
triaged,module: numpy
|
low
|
Minor
|
594,866,095 |
pytorch
|
Some tutorials cannot be found from both side panel and tutorial welcome page
|
## 🐛 Bug
Some tutorials cannot be accessed via Tutorial Home Page or PyTorch Tutorial side panel. Their links are missing. To access these tutorials, I search Google
## To Reproduce
Steps to reproduce the behavior:
1. Go to https://pytorch.org/tutorials/
1. These tutorials are not found in both side panel, and in the welcome page:
1. "WRITING CUSTOM DATASETS, DATALOADERS AND TRANSFORMS.
Link: https://pytorch.org/tutorials/beginner/data_loading_tutorial.html
1. WORD EMBEDDINGS: ENCODING LEXICAL SEMANTICS
Link: https://pytorch.org/tutorials/beginner/nlp/word_embeddings_tutorial.html
## Expected behavior
<!-- A clear and concise description of what you expected to happen. -->
## Environment
Please copy and paste the output from our
[environment collection script](https://raw.githubusercontent.com/pytorch/pytorch/master/torch/utils/collect_env.py)
(or fill out the checklist below manually).
You can get the script and run it with:
```
wget https://raw.githubusercontent.com/pytorch/pytorch/master/torch/utils/collect_env.py
# For security purposes, please check the contents of collect_env.py before running it.
python collect_env.py
```
- PyTorch Version (e.g., 1.0):
- OS (e.g., Linux):
- How you installed PyTorch (`conda`, `pip`, source):
- Build command you used (if compiling from source):
- Python version:
- CUDA/cuDNN version:
- GPU models and configuration:
- Any other relevant information:
## Additional context
<!-- Add any other context about the problem here. -->
|
module: docs,triaged
|
low
|
Critical
|
594,872,172 |
go
|
proposal: net/http, net/http/httptrace: add mechanism for tracing request serving
|
For the purpose of accurate request tracing when serving HTTP requests (for both HTTP/1.x and HTTP/2), it would be ideal if `http.Handler`s could obtain a `time.Time` representing the time of first byte of the request.
In this context, time of first byte would be the best estimate of the instant when the go process reads the first byte of the HTTP request line from the socket (i.e. ignoring the time spent in hardware queues and in kernel buffers).
Between the time of first byte and the time the first `http.Handler` is invoked for that request quite some time can pass since in that interval all headers must be received (and parsed) and the request body must start. If the client is slow then this interval can be extremely long.
Furthermore, when using the middleware pattern for tracing, the order of middleware can influence the recorded time at which the request starts to be handled. This can skew metrics and the accuracy of tracing, leading to hard-to-debug performance issues.
With this proposal implemented, all of the above issues would be solved:
- request parsing would be correctly accounted for
- slow clients can be correctly accounted for
- request start time can be reliably detected irrespective of middleware order
This information could be included either in the `http.Request` struct, returned by a method on `http.Request`, or attached to the request `context.Context` (perhaps via something like the `httptrace.ServerTrace` suggestion [below](https://github.com/golang/go/issues/38270#issuecomment-609725677) by @odeke-em).
---
update: rephrased according to @odeke-em's suggestions
|
Proposal,Proposal-Hold
|
medium
|
Critical
|
594,902,977 |
excalidraw
|
excalidraw CLI
|
Hi there,
I was wondering, what if we have a CLI for excalidraw?
```bash
excalidraw my-drawing.excalidraw
```
This will open a browser. If `my-drawing.excalidraw` already exists, it will load it.
When user clicks "save" button, it won't show the dialog but quietly save to `my-drawing.excalidraw`.
```bash
excalidraw my-drawing.excalidraw --export
```
with `--export` given, whenever user saves, the PNG version is exported as well.
```bash
excalidraw my-drawing.excalidraw --export images/my-drawing.png
```
---
I guess this can unblock new workflows, for example, for developers who want to have drawings in their git repository.
What do you think?
|
enhancement
|
medium
|
Critical
|
594,957,481 |
ant-design
|
fixed Table 勾選列後，欄位位置錯誤
|
- [ ] I have searched the [issues](https://github.com/ant-design/ant-design/issues) of this repository and believe that this is not a duplicate.
### Reproduction link
[https://imgur.com/mdLgADn](https://imgur.com/mdLgADn)
### Steps to reproduce
點擊checkbox
### What is expected?
https://imgur.com/peLrVBk
### What is actually happening?
點擊checkbox，欄位就跑掉
| Environment | Info |
|---|---|
| antd | 4.1.1 |
| React | 16.12.0 |
| System | win10 |
| Browser | IE 11 |
<!-- generated by ant-design-issue-helper. DO NOT REMOVE -->
|
Internet Explorer,Inactive
|
low
|
Major
|
594,965,035 |
PowerToys
|
Windows File Explorer folder global keyboard shortcuts
|
# Summary of the new feature/enhancement
System wide Explorer keyboard shortcuts for User folders (Desktop, Documents, Downloads, ...).
This would allow to make the current Explorer window go to that Folder instantly, instead, of clicking/tabbing etc.
This behavior is available on macOS Finder for some folders (CMD+Shift+D for Desktop).
# Proposed technical implementation details (optional)
When a File Explorer window or File Explorer dialog (Open/Save/etc) is open, pressing one of these shortcuts will directly switch to the default User folder:
- Desktop: Win+Shift+D
- Downloads: Win+Shift+T
- Documents: Win+Shift+F
(etc.)
|
Idea-New PowerToy,Product-File Explorer
|
low
|
Major
|
594,969,511 |
flutter
|
VerticalDivider is unusable out of the box
|
## Steps to Reproduce
<!-- You must include full steps to reproduce so that we can reproduce the problem. -->
1. Run `flutter create bug`.
2. Update the files as follows: main.dart
```
import 'package:flutter/material.dart';
void main() => runApp(MyApp());
class MyApp extends StatelessWidget {
@override
Widget build(BuildContext context) {
return MaterialApp(
title: 'Flutter Demo',
theme: ThemeData(
primarySwatch: Colors.blue,
),
home: MyHomePage(title: 'Flutter Demo Home Page'),
);
}
}
class MyHomePage extends StatefulWidget {
MyHomePage({Key key, this.title}) : super(key: key);
final String title;
@override
_MyHomePageState createState() => _MyHomePageState();
}
class _MyHomePageState extends State<MyHomePage> {
int _counter = 0;
@override
Widget build(BuildContext context) {
return Scaffold(
appBar: AppBar(
title: Text(widget.title),
),
body: Center(
child: Column(
children: <Widget>[
Row(
children: <Widget>[
Text('qwe'),
VerticalDivider(),
Text('qwe'),
],
),
IntrinsicHeight(
child: Row(
children: <Widget>[
Text('qwe'),
VerticalDivider(),
Text('qwe'),
],
),
),
],
),
),
);
}
}
```
3. ... <!-- describe how to reproduce the problem -->
**Expected results:** Divider shown in all rows
**Actual results:** First row divider not shown
<details>
```
D:\dev\pavement>flutter doctor -v
[√] Flutter (Channel unknown, v1.12.13+hotfix.7, on Microsoft Windows [Version 10.0.17134.1304], locale en-GB)
• Flutter version 1.12.13+hotfix.7 at C:\tools\flutter
• Framework revision 9f5ff2306b (2 months ago), 2020-01-26 22:38:26 -0800
• Engine revision a67792536c
• Dart version 2.7.0
[√] Android toolchain - develop for Android devices (Android SDK version 29.0.3)
• Android SDK at c:\tools\Android-SDK
• Android NDK location not configured (optional; useful for native profiling support)
• Platform android-29, build-tools 29.0.3
• ANDROID_HOME = c:\tools\Android-SDK
• Java binary at: C:\tools\android-studio\jre\bin\java
• Java version OpenJDK Runtime Environment (build 1.8.0_212-release-1586-b04)
• All Android licenses accepted.
[√] Android Studio (version 3.6)
• Android Studio at C:\tools\android-studio
• Flutter plugin version 45.0.1
• Dart plugin version 192.7761
• Java version OpenJDK Runtime Environment (build 1.8.0_212-release-1586-b04)
[!] VS Code, 32-bit edition (version 1.43.2)
• VS Code at C:\Program Files (x86)\Microsoft VS Code
X Flutter extension not installed; install from
https://marketplace.visualstudio.com/items?itemName=Dart-Code.flutter
[√] Connected device (1 available)
• LG H873 • LGH8736bb28167 • android-arm64 • Android 8.0.0 (API 26)
! Doctor found issues in 1 category.
```
</details>
|
framework,f: material design,a: quality,has reproducible steps,found in release: 3.3,found in release: 3.7,team-design,triaged-design
|
low
|
Critical
|
594,970,013 |
vscode
|
Terminal canvas context loss is not handled
|
<!-- ⚠️⚠️ Do Not Delete This! bug_report_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: https://code.visualstudio.com/insiders/ -->
<!-- Use Help > Report Issue to prefill these. -->
- VSCode Version: 1.43.2
- OS Version: macOS 10.15.4
<img width="843" alt="2" src="https://user-images.githubusercontent.com/11925432/78546082-716d3580-782f-11ea-9eda-57db8abc2e6c.png">
|
bug,upstream,upstream-issue-linked,terminal-rendering
|
low
|
Critical
|
594,991,753 |
react
|
Bug: Server hydration mistmatch and radio group with defaultChecked
|
<!--
Please provide a clear and concise description of what the bug is. Include
screenshots if needed. Please test using the latest version of the relevant
React packages to make sure your issue has not already been fixed.
-->
When hydrating from the server, whenever there's a mismatch in the initially checked input in a radio button group and the inputs use `defaultChecked` (uncontrolled), there is no warning of that mismatch and the component behaves in a buggy way for the input that was initially selected in the server payload (the `onChange` callback prop is not fired for the input that was marked as selected in the initial html, when selecting it).
Worth noting that using a controlled input (using `checked`) makes the bug go away.
React version: 16.13.1
## Steps To Reproduce
1. In the example provided below, make sure you refresh the browser within Code Sandbox.
2. Check the first radio button. Verify that the radio is checked, but the text next to it still shows as "not checked" (which means the onChange prop was not triggered)
3. Attempt to select any other option (works fine), and then back the first one (works fine too).
4. Refreshing the page again, choosing any other option other than the first works fine.
<!--
Your bug will get fixed much faster if we can run your code and it doesn't
have dependencies other than React. Issues without reproduction steps or
code examples may be immediately closed as not actionable.
-->
Link to code example: https://codesandbox.io/s/affectionate-stonebraker-3wj68
<!--
Please provide a CodeSandbox (https://codesandbox.io/s/new), a link to a
repository on GitHub, or provide a minimal code example that reproduces the
problem. You may provide a screenshot of the application if you think it is
relevant to your bug report. Here are some tips for providing a minimal
example: https://stackoverflow.com/help/mcve.
-->
## The current behavior
I know that using hydrate doesn't guarantee fixing the mismatches, but it says that it should warn about them in development (https://reactjs.org/docs/react-dom.html#hydrate). In this case, React doesn't warn about the mismatch, and the behavior is buggy, as shown in the example and steps to repro.
## The expected behavior
React should ideally work without the issue, but given that it's not guaranteed to be fixed according to the documentation, it should at least warn about the mismatch.
|
Type: Bug,Component: DOM,Component: Server Rendering
|
low
|
Critical
|
595,056,524 |
flutter
|
CustomScrollView should not scroll when pointer down starts on Sliver App Bar
|
When using SliverAppBar you can scroll the view by only touching the app bar, that is really weird.
Dragging inside SliverAppBar should not affect the scroll of the scroll view. Also it would be interesting to block scrolling also to SliverPersistentHeader when is collapsed.
https://dartpad.dartlang.org/7c10802218f91b56cfa51ae64a27f389
|
c: new feature,framework,f: material design,f: scrolling,a: quality,has reproducible steps,P3,found in release: 3.3,found in release: 3.5,team-design,triaged-design
|
low
|
Minor
|
595,059,784 |
godot
|
.obj do not have correct material colors when imported directly . .
|
**Godot version:**
Godot 3.2 . .
**OS/device including version:**
Windows 10, from Blender 3D . . .
**Issue description:**
When I import a 100 % white material from Blender, on an .obj, which are nice because they are easy to work with, for static objects, props, it's like the ' 100 % white ' becomes slightly gray, or dark . . . When I make a texture, with 100 % white, and put that on the model, as an albedo, it becomes 100 % white, after applying, in Blender . . I use materials for coloring in prototypes, and really like just press export in Blender, and the Godot imports it, but there's a mistake, in the color conversion, for some reason, on materials . . .
**Steps to reproduce:**
Put some materials, with specific colors on them, on an .obj in Blender, before exporting the model, then copy it into Godot game folder, and put it in a meshinstance 3D slot . . Then, take the hex values in Blender, for the colors, and make a 2D texture, after UV unwrapping, it can be a box, something simple . . . When you import the same colors, even if they're the same in Blender, by hex value, they are darker in the materials, in Godot, in the viewport, than the albedo, at least on my computer . . . Can you fix it, exporting .obj's with materials is nice for beginners, and also for prototyping, or so . .
**Minimal reproduction project:**
I am not sure what you mean, I'm also not sure if this is a bug, it just looks weird, applying a texture with the same hex colors, using texture paint, fixes it, but it's annoying, that .obj import doesn't work, it seems to me . . . .
|
discussion,topic:rendering
|
low
|
Critical
|
595,105,699 |
rust
|
Extraneous documentation in src/doc
|
While exploring rustup's doc functionality I found a bunch of outdated documentation, e.g. `guide-strings.html`. Nearly all of documentation in the top level `src/doc` is out of date, both in terms of content and style, and it doesn't seem like anyone is interested in maintaining them. The generated output is about ~800K so it's not adversely affecting the download size. Most of the documentation just tells to go another website.
### Possible Solutions
- Remove the files. This will technically break links that reference this documentation, though I don't know if these a lot of these docs have been actively referenced since 2015–2016.
- Change the implementation to generate redirects, rather than regular pages.
|
C-cleanup,T-rustdoc
|
low
|
Major
|
595,136,528 |
godot
|
Can't access parent's named enum from match statement
|
**Godot version:** 3.2.1
**OS/device including version:** Arch Linux *(kernel 5.5.13)*
**Issue description:** Godot throws an error when I try to access a named enum from a parent inside a `match` statement. If I change to be a classic `if/elif/else` block, it works fine.
The error is `built-in:8 - Parse Error: Invalid operator in pattern. Only index (A.B) is allowed`, which suggests that Godot won't allow to access a field more than once.
Looks related to #10571
**Steps to reproduce:** Have a scene tree like:
```
Parent
└── Child
```
**Parent.gd**:
```gdscript
extends Node
enum Sides {
LEFT,
RIGHT
}
var currentSide = Sides.RIGHT
func _ready():
$Child.foo(currentSide)
```
**Child.gd**
```gdscript
extends Node
func foo(side):
match side:
$"..".Sides.LEFT: # <--- Errors here
print("Left side")
$"..".Sides.RIGHT: # <--- and here
print("Right side")
```
**Minimal reproduction project:** [project.zip](https://github.com/godotengine/godot/files/4438747/project.zip)
|
bug,topic:gdscript
|
low
|
Critical
|
595,138,996 |
pytorch
|
[JIT] support list of nn.Module in torchscript
|
## 🚀 Feature
Let torchsciprt support list of nn.Module
## Motivation
Many projects (like detectron2) use list of modules instead of nn.ModuleList to avoid complex namespace of model parameters, but currently torchscript does not support this property. To make these projects scriptable, we must replace list of modules with nn.ModuleList which will make the pretrained models invalid, since the replacement will change the keys of state_dict, and may increase the number of model parameters.
## Pitch
we may define a scriptable model like:
```
def MyModel(nn.Module):
def __init__(self):
super().__init__()
self.layers: List[nn.Module] = []
for i in range(3):
# this definition is common
self.add_module(str(i), some_layer)
self.layers.append(some_layer)
def forward(self, x):
for m in self.layers:
x = m(x)
return x
```
cc @suo
|
triage review,oncall: jit,triaged,enhancement
|
medium
|
Major
|
595,146,488 |
rust
|
E0597: `lock` does not live long enough (surprising error, the message could be more helpful)
|
I tried compiling this code:
```rust
fn main() {
let lock = std::sync::Mutex::new(10);
if let Ok(_) = lock.try_lock() {}
}
```
Error:
```
error[E0597]: `lock` does not live long enough
--> src/main.rs:3:20
|
3 | if let Ok(_) = lock.try_lock() {}
| ^^^^-----------
| |
| borrowed value does not live long enough
| a temporary with access to the borrow is created here ...
4 | }
| -
| |
| `lock` dropped here while still borrowed
| ... and the borrow might be used here, when that temporary is dropped and runs the destructor for type `std::result::Result<std::sync::MutexGuard<'_, i32>, std::sync::TryLockError<std::sync::MutexGuard<'_, i32>>>`
|
= note: The temporary is part of an expression at the end of a block. Consider adding semicolon after the expression so its temporaries are dropped sooner, before the local variables declared by the block are dropped.
error: aborting due to previous error
For more information about this error, try `rustc --explain E0597`.
error: could not compile `tmp`.
To learn more, run the command again with --verbose.
```
I'm surprised that this does not compile!
The error message suggests adding a semicolon after the expression, but it's not obvious to me what exactly the expression is and where the semicolon should go. A suggestion on where to place the semicolon would be helpful.
The solution is this and compiles just fine:
```rust
fn main() {
let lock = std::sync::Mutex::new(10);
if let Ok(_) = lock.try_lock() {};
}
```
|
A-borrow-checker,T-compiler,C-bug
|
low
|
Critical
|
595,150,642 |
pytorch
|
RuntimeError CUDA error despite CUDA available and GPU supported
|
Got the following error (on Ubuntu 18.04 with PyTorch 1.14, Nvidia driver version 440.64, CUDA toolkit 10.1, GPU GeForce GTX Titan) despite PyTorch saying CUDA and CUDNN are available:
RuntimeError: CUDA error: no kernel image is available for execution on the device
Script to reproduce:
```
import torch
print(torch.__version__)
print(torch.cuda.is_available())
print(torch.backends.cudnn.enabled)
device = torch.device('cuda')
print(torch.cuda.get_device_properties(device))
print(torch.tensor([1.0, 2.0]).cuda())
```
Output:
```
1.4.0
True
True
_CudaDeviceProperties(name='GeForce GTX TITAN', major=3, minor=5, total_memory=6082MB, multi_processor_count=14)
Traceback (most recent call last):
File "test_pytorch_cuda.py", line 7, in <module>
print(torch.tensor([1.0, 2.0]).cuda())
File "/home/deepstation/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/tensor.py", line 159, in __repr__
return torch._tensor_str._str(self)
File "/home/deepstation/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/_tensor_str.py", line 311, in _str
tensor_str = _tensor_str(self, indent)
File "/home/deepstation/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/_tensor_str.py", line 209, in _tensor_str
formatter = _Formatter(get_summarized_data(self) if summarize else self)
File "/home/deepstation/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/_tensor_str.py", line 87, in __init__
nonzero_finite_vals = torch.masked_select(tensor_view, torch.isfinite(tensor_view) & tensor_view.ne(0))
RuntimeError: CUDA error: no kernel image is available for execution on the device
```
cc @ezyang @seemethere @ngimel
|
module: binaries,module: cuda,module: error checking,triaged
|
low
|
Critical
|
595,159,513 |
flutter
|
Cupertino route/transition should block touch events
|
I think block any touch events during Cupertino transition is reasonable.
|
platform-ios,framework,a: fidelity,f: cupertino,f: routes,c: proposal,P2,has partial patch,team-design,triaged-design
|
low
|
Major
|
595,163,882 |
youtube-dl
|
Evisco link download
|
<!--
######################################################################
WARNING!
IGNORING THE FOLLOWING TEMPLATE WILL RESULT IN ISSUE CLOSED AS INCOMPLETE
######################################################################
-->
## Checklist
<!--
Carefully read and work through this check list in order to prevent the most common mistakes and misuse of youtube-dl:
- First of, make sure you are using the latest version of youtube-dl. Run `youtube-dl --version` and ensure your version is 2020.03.24. If it's not, see https://yt-dl.org/update on how to update. Issues with outdated version will be REJECTED.
- Make sure that all provided video/audio/playlist URLs (if any) are alive and playable in a browser.
- Make sure that site you are requesting is not dedicated to copyright infringement, see https://yt-dl.org/copyright-infringement. youtube-dl does not support such sites. In order for site support request to be accepted all provided example URLs should not violate any copyrights.
- Search the bugtracker for similar site support requests: http://yt-dl.org/search-issues. DO NOT post duplicates.
- Finally, put x into all relevant boxes (like this [x])
-->
- [x] I'm reporting a new site support request
- [x] I've verified that I'm running youtube-dl version **2020.03.24**
- [x] I've checked that all provided URLs are alive and playable in a browser
- [x] I've checked that none of provided URLs violate any copyrights
- [x] I've searched the bugtracker for similar site support requests including closed ones
## Example URLs
<!--
Provide all kinds of example URLs support for which should be included. Replace following example URLs by yours.
-->
- Single video: https://api.video.evisco.com/html5/html5lib/v2.78.2/mwEmbedFrame.php/p/104/uiconf_id/23448508/entry_id/0_pgoho8ms?wid=_104
embedded link from:
https://operlive.de/parsifal/
## Description
<!--
Provide any additional information.
If work on your issue requires account credentials please provide them or explain how one can obtain them.
-->
The link will only be available until April 11. I would like to download it to be able to watch later.
No login needed.
/usr/local/bin/youtube-dl -v https://api.video.evisco.com/html5/html5lib/v2.78.2/mwEmbedFrame.php/p/104/uiconf_id/23448508/entry_id/0_pgoho8ms?wid=_104
[debug] System config: []
[debug] User config: []
[debug] Custom config: []
[debug] Command-line args: [u'-v', u'https://api.video.evisco.com/html5/html5lib/v2.78.2/mwEmbedFrame.php/p/104/uiconf_id/23448508/entry_id/0_pgoho8ms?wid=_104']
[debug] Encodings: locale UTF-8, fs utf-8, out UTF-8, pref UTF-8
[debug] youtube-dl version 2020.03.24
[debug] Python version 2.7.16 (CPython) - Darwin-19.4.0-x86_64-i386-64bit
[debug] exe versions: none
[debug] Proxy map: {}
[generic] 0_pgoho8ms?wid=_104: Requesting header
WARNING: Falling back on generic information extractor.
[generic] 0_pgoho8ms?wid=_104: Downloading webpage
[generic] 0_pgoho8ms?wid=_104: Extracting information
ERROR: Unsupported URL: https://api.video.evisco.com/html5/html5lib/v2.78.2/mwEmbedFrame.php/p/104/uiconf_id/23448508/entry_id/0_pgoho8ms?wid=_104
Traceback (most recent call last):
File "/usr/local/bin/youtube-dl/youtube_dl/extractor/generic.py", line 2375, in _real_extract
doc = compat_etree_fromstring(webpage.encode('utf-8'))
File "/usr/local/bin/youtube-dl/youtube_dl/compat.py", line 2551, in compat_etree_fromstring
doc = _XML(text, parser=etree.XMLParser(target=_TreeBuilder(element_factory=_element_factory)))
File "/usr/local/bin/youtube-dl/youtube_dl/compat.py", line 2540, in _XML
parser.feed(text)
File "/System/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/xml/etree/ElementTree.py", line 1659, in feed
self._raiseerror(v)
File "/System/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/xml/etree/ElementTree.py", line 1523, in _raiseerror
raise err
ParseError: not well-formed (invalid token): line 5, column 118
Traceback (most recent call last):
File "/usr/local/bin/youtube-dl/youtube_dl/YoutubeDL.py", line 797, in extract_info
ie_result = ie.extract(url)
File "/usr/local/bin/youtube-dl/youtube_dl/extractor/common.py", line 530, in extract
ie_result = self._real_extract(url)
File "/usr/local/bin/youtube-dl/youtube_dl/extractor/generic.py", line 3352, in _real_extract
raise UnsupportedError(url)
|
site-support-request
|
low
|
Critical
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.