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 |
|---|---|---|---|---|---|---|
559,495,663 | flutter | Specify tab size in Text widget | Hi, is there any parameter to specify tab size in text widget such as `Text` and `RichText`?
Seems it is 8 by default. In some cases developers may want to specify it as 4 or 2, for example [code synatx highlighting lib](https://github.com/pd4d10/highlight), like the [`tab-size` prop](https://developer.mozilla.org/en-US/docs/Web/CSS/tab-size) in CSS
### Screenshot
### Reproduce steps
Paste the code below to https://dartpad.dev/flutter
```dart
import 'package:flutter/material.dart';
final Color darkBlue = Color.fromARGB(255, 18, 32, 47);
void main() {
runApp(MyApp());
}
class MyApp extends StatelessWidget {
@override
Widget build(BuildContext context) {
return MaterialApp(
theme: ThemeData.dark().copyWith(scaffoldBackgroundColor: darkBlue),
debugShowCheckedModeBanner: false,
home: Scaffold(
body: Center(
child: MyWidget(),
),
),
);
}
}
class MyWidget extends StatelessWidget {
@override
Widget build(BuildContext context) {
return Column(
crossAxisAlignment: CrossAxisAlignment.start,
children: <Widget>[
Text(
'#\tTab in Text',
style: TextStyle(fontFamily: 'Courier New'),
),
RichText(
text: TextSpan(
text: '#\tTab in RichText',
style: TextStyle(fontFamily: 'Courier New'),
),
),
Text.rich(
TextSpan(text: '#\tTab in Text.rich'),
style: TextStyle(fontFamily: 'Courier New'),
),
],
);
}
}
``` | c: new feature,framework,engine,a: typography,P3,team-engine,triaged-engine | low | Critical |
559,520,192 | godot | editor/editor_node.cpp:5268 - An EditorPlugin build callback failed | **Godot version:** 3.2 mono
**OS/device including version:** Ubuntu 18.04
**Issue description:** Projects fail to run, error message is editor/editor_node.cpp:5268 - An EditorPlugin build callback failed. This happened on an imported project, and later happened in a blank project as well. On the blank project, I can run it once, then the bug appears. Deleting the .sln file seems to fix the problem, restoring it causes the error to pop up again.
**Steps to reproduce:** Do not know.
**Minimal reproduction project:** https://github.com/TZubiri/godot-mcve-bug
**Related issues:** https://github.com/godotengine/godot/issues/23562 https://github.com/godotengine/godot/issues/29232 | bug,topic:editor,confirmed,topic:dotnet | medium | Critical |
559,583,674 | rust | Misleading error when incorrectly using crate as a module. | I had a source file used as a module in my binary crate. At some point, I have decided to turn this file into a second binary crate instead. But I forgot to remove the corresponding `mod` statement in the first crate. It took me a while to figure it out because the error message was very misleading.
Here is the situation boiled down:
`Cargo.toml`
```toml
[package]
name = "rust_test"
version = "0.1.0"
authors = ["Iago-lito"]
edition = "2018"
default-run = "first_bin"
[[bin]]
name = "first_bin"
path = "src/first_bin.rs"
[[bin]]
name = "second_bin"
path = "src/second_bin.rs"
```
`src/first_bin.rs`
```rust
mod common_module;
mod second_bin; // That is very wrong..
fn main() {
common_module::common_function();
}
```
`src/second_bin.rs`
```rust
mod common_module; // .. but the error blames this line.
fn main() {
common_module::common_function();
}
```
`src/common_module.rs`
```rust
pub fn common_function() {}
```
I'd expect getting something like:
```
error[EXXX]: binary crate root file cannot be used as a module.
--> src/first_bin:2:5
|
2 | mod second_bin;
| ^^^^^^^^^^
|
= help: remove this statement or consider not declaring `src/second_bin.rs` as the root of a binary crate in `Cargo.toml`
```
But the actual error is:
```
error[E0583]: file not found for module `common_module`
--> src/second_bin.rs:1:5
|
1 | mod common_module;
| ^^^^^^^^^^^^^
|
= help: name the file either second_bin/common_module.rs or second_bin/common_module/mod.rs inside the directory "src"
```
Because of this, I had to dig the whole way through [Chapter 7 of the Book](https://doc.rust-lang.org/book/ch07-00-managing-growing-projects-with-packages-crates-and-modules.html) again before I figured out that there was nothing wrong with the location of `src/common_module.rs`. Then I had to scroll the other files line by line to eventually spot the mistake.
Would this be considered a bug? Is it the correct place to report it? | A-diagnostics,T-compiler,D-terse | low | Critical |
559,667,319 | godot | HTTPClient is_response_chunked (seemingly) always returns false | **Godot version:**
3.2 Release
**OS/device including version:**
Linux Mint 19.3
**Issue description:**
`httpclient.is_response_chunked() == false` even though the response is split into chunks
**Steps to reproduce:**
Load a file >4096 bytes and read the output. It will be chunked regardless of `is_response_chunked()`
```swift
const MY_HOST := "https://gist.githubusercontent.com"
const MY_EP := "/The5heepDev/a15539b297a7862af4f12ce07fee6bb7/raw/7164813a9b8d0a3b2dcffd5b80005f1967887475/entire_bee_movie_script"
func _ready() -> void:
var httpclient := HTTPClient.new()
assert(httpclient.connect_to_host(MY_HOST, -1, true) == OK)
while httpclient.get_status() == HTTPClient.STATUS_RESOLVING or httpclient.get_status() == HTTPClient.STATUS_CONNECTING:
assert(httpclient.poll() == OK)
yield(Engine.get_main_loop(), "idle_frame")
print(MY_HOST + MY_EP)
assert(httpclient.request(HTTPClient.METHOD_GET, MY_EP, []) == OK)
while httpclient.get_status() == HTTPClient.STATUS_REQUESTING:
assert(httpclient.poll() == OK)
yield(Engine.get_main_loop(), "idle_frame")
print()
print("Chunked?: " + str(httpclient.is_response_chunked()))
print("Full Body: " + str(httpclient.get_response_body_length()))
print()
while httpclient.get_status() == HTTPClient.STATUS_BODY:
var size := httpclient.read_response_body_chunk().size()
if size != 0:
print("Chunk: " + str(size))
```
**Minimal reproduction project:**
[Test7.zip](https://github.com/godotengine/godot/files/4153043/Test7.zip) | enhancement,documentation,topic:network | low | Minor |
559,680,978 | kubernetes | When a lot of persistentvolumes are created together, POST persistentvolume request latency grows significantly | **What happened**:
I was running a scale test where I created ~8 PV/s for some period of time. I observed that POST persistentvolume latency started growing
I debugged this already:
* kube-controller-manager created PD objects in GCE
* due to rate limited error, it failed to fetch labels for new disk: https://github.com/kubernetes/kubernetes/blob/master/pkg/volume/gcepd/gce_util.go#L191
* PersistentVolumeLabel admission controller tried to add missing GCE labels, but failed due to the same rate limited error (this also caused significant delay in request processing)
**What you expected to happen**:
POST persistentvolume latency to be ~constant during the test.
This can be achieved by:
1. removing cloud.GetAutoLabelsForZonalPD call from https://github.com/kubernetes/kubernetes/blob/master/pkg/volume/gcepd/gce_util.go#L188
* this way, there is no API call that can fail after object is created
* since we just successfully created a PD, we know all parameters for that PD (zone, region, potentially in future some other disk's properties as well) and we can use this information for labels computation and there is no point in validating if it still exists.
2. Implementing proper retry strategy for cloud.GetAutoLabelsForPD calls so that we do not move load to kube-apiserver when kube-controller-manager is overloaded.
3. (partially) by simply passing zone argument in https://github.com/kubernetes/kubernetes/blob/master/pkg/volume/gcepd/gce_util.go#L188 we can reduce number of calls there 3-4x times in case of zonal disks, but this is a mitigation only
For me, 1 is the most reasonable thing to do (and I have some draft PR for that), but I wanted to hear your opinion on this before I start cleaning the PR.
Potential drawbacks of this approach are:
* Right now, if disk with given name already exists, we simply reuse that disk (https://github.com/kubernetes/kubernetes/blob/master/staging/src/k8s.io/legacy-cloud-providers/gce/gce_disks.go#L732). In this case, we in fact don't know what replicationZones have been used for creation of that PD, so we do need to get them in that case. We can limit get calls to only this case.
**How to reproduce it (as minimally and precisely as possible)**:
Just create a lot of pvs together.
**Anything else we need to know?**:
**Environment**:
- Kubernetes version (use `kubectl version`):
- Cloud provider or hardware configuration:
- OS (e.g: `cat /etc/os-release`):
- Kernel (e.g. `uname -a`):
- Install tools:
- Network plugin and version (if this is a network-related bug):
- Others:
/assign @saad-ali
/cc @wojtek-t | kind/bug,sig/scalability,sig/storage,lifecycle/frozen | low | Critical |
559,693,366 | vscode | Allow to reveal an empty TreeView programmatically | Currently, it i possible to reveal/expand a TreeView only when there is an element in the TreeView. it is oputting revealing this element in the TreeView.
it would useful to be abel to expand the TreeView even when there are no elements in the Tree.
Use case:
- when testing https://github.com/camel-tooling/vscode-camelk/pull/291
- when writing a tutorial guiding users https://github.com/redhat-developer/vscode-didact/issues/4#issuecomment-557648413
asked on StackOverflow https://stackoverflow.com/questions/59983711/how-to-programmatically-make-a-treeview-visible | feature-request,api,tree-views,api-proposal | medium | Critical |
559,699,002 | rust | rustc generates a lot of LLVM IR when using a match in struct field initialization | In https://github.com/rusterlium/rustler/issues/299, we detected compile time to increase non-linearly when using a rather simple proc-macro. In https://github.com/rusterlium/rustler/issues/299#issuecomment-579213528, we found that our decoder function results in a lot of IR. We were able to fix the issue in https://github.com/rusterlium/rustler/pull/300 by making a simple transformation for generated struct initialization.
## Transformation
The change to our proc-macro resulted in the following transformation of the resulting code.
Before:
```rust
Ok(SomeStruct {
field1: match Decoder::decode(term[1usize])? { ... }
// .. more fields
})
```
After:
```rust
let field1 = match Decoder::decode(term[1usize]) { ... };
// .. more bindings
Ok(SomeStruct {
field1,
// .. more fields
})
```
## Simplification
I simplified our case in https://github.com/evnu/rust-slow-compilation. I haven't been able to simplify this further: when I tried to get rid of the rustler-specific `Decoder`, or when using `map_err()`, compilation became very fast. The [README](https://github.com/evnu/rust-slow-compilation#run-the-benchmark) of the repository indicates compilation times on my machine for specific revisions of the repository.
## Meta
```
$ 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
``` | C-enhancement,A-codegen,I-compiletime,T-compiler | low | Major |
559,716,176 | terminal | Ability (perhaps shortcut?) to evenly resize splits | Can we have added functionality that (via a keystroke) automatically resizes any multiple rows/columns equally within the window, please.
| Help Wanted,Area-Settings,Product-Terminal,Issue-Task | low | Major |
559,742,648 | pytorch | which pytorch do i install for running this project in my windows.link for this is :: https://github.com/xiaojunxu/SQLNet .IN this they have used python 2.7 ,but i am unable to install pytorch on python 2.7 environment .help me with this | ## ❓ Questions and Help
### Please note that this issue tracker is not a help form and this issue will be closed.
We have a set of [listed resources available on the website](https://pytorch.org/resources). Our primary means of support is our discussion forum:
- [Discussion Forum](https://discuss.pytorch.org/)
cc @ezyang | module: binaries,triaged | low | Minor |
559,786,792 | tensorflow | How can I clear GPU memory in tensorflow 2? |
### System information
- Custom code; nothing exotic though.
- Ubuntu 18.04
- installed from source (with pip)
- tensorflow version v2.1.0-rc2-17-ge5bf8de
- 3.6
- CUDA 10.1
- Tesla V100, 32GB RAM
I created a model, nothing especially fancy in it. When I create the model, when using nvidia-smi, I can see that tensorflow takes up nearly all of the memory. When I try to fit the model with a small batch size, it successfully runs. When I fit with a larger batch size, it runs out of memory. Nothing unexpected so far.
However, the only way I can then release the GPU memory is to restart my computer. When I run nvidia-smi I can see the memory is still used, but there is no process using a GPU. Also, If I try to run another model, it fails much sooner.
Nothing in the first five pages of google results works. (and most solutions are for TF1)
Is there any way to release GPU memory in tensorflow 2? | stat:awaiting tensorflower,type:bug,comp:gpu,TF 2.7 | high | Critical |
559,817,583 | terminal | Add "Run As Administrator" menu option to Start Menu and Taskbar | # Description of the new feature/enhancement
Ability, primarily in Windows, to easily open the terminal app in an elevated session. Either via the Start Menu icon, as the PowerShell app does it, or via the taskbar icon.
Example for Start Menu:
Example for Taskbar icon:
| Area-UserInterface,Product-Terminal,Issue-Task | high | Critical |
559,818,952 | react-native | Android: textInput: secureTextEntry toggle causes keyboard change; keyboard does not automatically come up | Description:
When enabling the user to toggle the secure text entry flag on a text input, on Android, two issues are occurring:
1. The keyboard does not automatically come up when rendering; the user must tap on the field.
2. Tapping the 'eye' icon to toggle the secureTextEntry flag from false to true switches the keyboard type.
React Native version:
System: OS: Windows 10 10.0.17134
CPU: (4) ia32 Intel(R) Pentium(R) CPU G4560T @ 2.90GHz
Memory: 1.65 GB / 7.43 GB
Binaries:
Node: 12.10.0 - C:\Program Files (x86)\nodejs\node.EXE
Yarn: 1.10.1 - C:\Users\mjfiandaca\AppData\Roaming\npm\yarn.CMD
npm: 6.13.4 - C:\Program Files (x86)\nodejs\npm.CMD
SDKs:
Android SDK:
API Levels: 23, 25, 26, 27, 28
Build Tools: 23.0.1, 25.0.1, 26.0.3, 27.0.3, 28.0.0, 28.0.3
System Images: android-23 | Intel x86 Atom, android-23 | Google APIs Intel x86 Atom, android-23 | Google APIs Intel x86 Atom_64, android-26 | Google APIs Intel x86 Atom, android-26 | Google APIs Intel x86 Atom_64, android-27 | Google APIs Intel x86 Atom, android-27 | Google Play Intel x86 Atom, android-28 | Intel x86 Atom, android-28 | Intel x86 Atom_64, android-28 | Google APIs Intel x86 Atom, android-28 | Google APIs Intel x86 Atom_64, android-28 | Google Play Intel x86 Atom
IDEs:
Android Studio: Version 3.5.0.0 AI-191.8026.42.35.6010548
npmPackages:
react: 16.8.6 => 16.8.6
react-native: 0.60.6 => 0.60.6
## Steps To Reproduce
1. Sign onto the provided snack.
2. Intermittently, the keyboard will not come up (I tested this several times; sometimes, no keyboard, sometimes, app starts, flashes, and then the keyboard comes up).
3. Tap on the 'eye' icon to toggle the secure text entry.
4. Notice the keyboard changes.
## Expected Results
The keyboard should not change based on the secure text entry value.
## Snack, code example, screenshot, or link to a repository:
https://snack.expo.io/@mjfiandaca/toggle-securetextentry | Platform: iOS,Issue: Author Provided Repro,Component: TextInput,Platform: Android,Needs: React Native Team Attention | medium | Critical |
559,833,252 | material-ui | [Select] VoiceOver & Chrome on native select is not announcing the newly chosen option when the select menu is closed | <!-- Provide a general summary of the issue in the Title above -->
<!--
Thank you very much for contributing to Material-UI by creating an issue! ❤️
To avoid duplicate issues we ask you to check off the following list.
-->
<!-- Checked checkbox should look like this: [x] -->
- [x] The issue is present in the latest release.
- [x] I have searched the [issues](https://github.com/mui-org/material-ui/issues) of this repository and believe that this is not a duplicate.
## Current Behavior 😯
<!-- Describe what happens instead of the expected behavior. -->
When using the native select and I'm closing any of the dropdowns, I'm hearing the previous selection being announced rather than what's being shown currently. If I navigate away with tab, and back again, the announcement is correct.
This doesn't seem to be an issue with VoiceOver on Safari (Mac OS) or ChromeVox on Chrome (Acer Chromebook 14).
## Expected Behavior 🤔
<!-- Describe what should happen. -->
I would expect the new selection to be announced when I closed the select menu, as per https://www.w3schools.com/tags/tryit.asp?filename=tryhtml_select.
## Steps to Reproduce 🕹
Steps:
1. Turn on VoiceOver and open https://material-ui.com/components/selects/#native-select in Chrome.
2. Tab navigate to the first example.
3. "Age, collapsed, button" will be announced.
4. Press Space, "menu 4 items tick" will be announced.
5. Arrow key down to "Ten" and press Space key - "Ten, Age, collapsed, button" will be announced.
6. Open the menu with Space, arrow key down, choose another option "Twenty". "Closing menu, Ten, Age, collapsed, button" will be announced.
7. Open the menu with Space, arrow key down, choose another option "Thirty". "Closing menu, Twenty, Age, collapsed, button" will be announced.
## Context 🔦
<!--
What are you trying to accomplish? How has this issue affected you?
Providing context helps us come up with a solution that is most useful in the real world.
-->
It's misleading for a screenreader user to announce the old selection after interacting with the select.
## Your Environment 🌎
| Tech | Version |
| ----------- | ------- |
| Material-UI | v4.9.1 |
| MacOS | Catalina 10.15.3 |
| Chrome | Version 79.0.3945.130 (Official Build) (64-bit) | | accessibility,component: select | low | Minor |
559,876,561 | pytorch | [JIT] pytorch 1.4 breaks torch.jit.script(LSTM/GRU) | ## 🐛 Bug
### LSTM/GRU cannot be torch scripted in new release (`1.4.0`)
## To Reproduce
Steps to reproduce the behaviour:
```shell
conda create -n pytorch1.4 python=3.7 pytorch=1.4 -c pytorch -y
conda activate pytorch1.4
python3
```
In `python3` interpreter
```python
import torch
input_size = 3
num_layers = 2
seq_len = 3
batch = 3
args = (torch.randn(seq_len, batch, input_size),)
scripted_lstm = torch.jit.script(torch.nn.LSTM(input_size, 3, num_layers))
scripted_lstm(*args)
```
Throws error message:
```shell
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/julian/miniconda3/envs/pytorch1.4/lib/python3.7/site-packages/torch/nn/modules/module.py", line 532, in __call__
result = self.forward(*input, **kwargs)
File "/home/julian/miniconda3/envs/pytorch1.4/lib/python3.7/site-packages/torch/jit/__init__.py", line 1678, in __getattr__
return super(RecursiveScriptModule, self).__getattr__(attr)
File "/home/julian/miniconda3/envs/pytorch1.4/lib/python3.7/site-packages/torch/jit/__init__.py", line 1499, in __getattr__
return super(ScriptModule, self).__getattr__(attr)
File "/home/julian/miniconda3/envs/pytorch1.4/lib/python3.7/site-packages/torch/nn/modules/module.py", line 576, in __getattr__
type(self).__name__, name))
AttributeError: 'RecursiveScriptModule' object has no attribute 'forward'
```
## Expected behaviour
The expected behaviour is for the `scripted_lstm(*args)` not to output an error and return the tensor.
I get this behaviour when I use `1.3` and replace the first x2 bash lines above with:
```shell
conda create -n pytorch1.3 python=3.7 pytorch=1.3 -c pytorch -y
conda activate pytorch1.3
```
## Environment
```shell
Collecting environment information...
PyTorch version: 1.4.0
Is debug build: No
CUDA used to build PyTorch: 10.1
OS: Ubuntu 18.04.3 LTS
GCC version: (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0
CMake version: version 3.10.2
Python version: 3.7
Is CUDA available: Yes
CUDA runtime version: Could not collect
GPU models and configuration: GPU 0: GeForce GTX 1050 Ti with Max-Q Design
Nvidia driver version: 430.64
cuDNN version: Could not collect
Versions of relevant libraries:
[pip3] numpy==1.13.3
[pip3] torch-stft==0.1.4
[conda] blas 1.0 mkl
[conda] mkl 2019.4 243
[conda] mkl-service 2.3.0 py37he904b0f_0
[conda] mkl_fft 1.0.15 py37ha843d7b_0
[conda] mkl_random 1.1.0 py37hd6b4f25_0
[conda] pytorch 1.4.0 py3.7_cuda10.1.243_cudnn7.6.3_0 pytorch
```
## Additional context
N/A
cc @ezyang @gchanan @zou3519 @suo | triage review,oncall: jit,triaged,has workaround | low | Critical |
559,901,166 | You-Dont-Know-JS | sync & async: research event loop | https://www.youtube.com/watch?v=cCOL7MC4Pl0 | for second edition | medium | Minor |
559,931,199 | rust | Spurious 'conflicting implementations' error when specializing a type with a compilation error | The following code:
```rust
#![feature(specialization)]
struct BadStruct {
err: MissingType
}
trait MyTrait<T> {
fn foo();
}
impl<T, D> MyTrait<T> for D {
default fn foo() {}
}
impl<T> MyTrait<T> for BadStruct {
fn foo() {}
}
```
gives the following errors:
```rust
error[E0412]: cannot find type `MissingType` in this scope
--> src/lib.rs:4:10
|
4 | err: MissingType
| ^^^^^^^^^^^ not found in this scope
error[E0119]: conflicting implementations of trait `MyTrait<_>` for type `BadStruct`:
--> src/lib.rs:15:1
|
11 | impl<T, D> MyTrait<T> for D {
| --------------------------- first implementation here
...
15 | impl<T> MyTrait<T> for BadStruct {
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ conflicting implementation for `BadStruct`
error: aborting due to 2 previous errors
```
For some reason, the fact that the definition of `BadStruct` has an error (`MissingType` is not defining) causes a 'conflicting implementations' error to be emitted when `BadStruct` has a specialized impl. If `MissingType` is changed to a type which actually exists (e.g. `()`), the 'conflicting implementations' error disappears.
I found this when working on `rustc` - a missing `use` statement caused 30 spurious specialization-related errors to be emiited, which all disappeared when I added the missing import. | A-diagnostics,E-needs-test,T-compiler,A-specialization,C-bug,F-specialization,D-verbose | low | Critical |
559,965,321 | godot | Calling signal callback with wrong parameter types raises confusing "Method not found" error. | [Networking.zip](https://github.com/godotengine/godot/files/4155965/Networking.zip)
**Godot version:** 3.1.2
**OS/device including version:** Windows 10
**Issue description:**
When the the server (specific godot project) received an event (signal) with the name "network_peer_packet", the system should fire the method "ReceivedPacket" in the class Server.cs.
But instead it doesn't found it.
`ERROR: emit_signal: Error calling method from signal 'network_peer_packet': 'Node2D(MinimalServer.cs)::ReceivedPacket': Method not found.
At: core/object.cpp:1236`
**Steps to reproduce:**
> Create a server with NetworkedMultiplayerENet
> Connect signal "network_peer_packet" to a method
> Create a client
> SendByte to the server
**Minimal reproduction project:**
Server side :
```
public class MinimalServer : Node {
public readonly int PORT = 4041;
public readonly int MAX_USERS = 2;
private NetworkedMultiplayerENet host;
public override void _Ready() {
this.host = new NetworkedMultiplayerENet();
Error error = host.CreateServer(PORT, MAX_USERS);
if (error != Error.Ok) {
GD.PrintErr(error); return;
}
this.GetTree().SetNetworkPeer(host);
// HERE: Bind
GetTree().Multiplayer.Connect("network_peer_packet", this, nameof(ReceivedPacket));
}
public void ReceivedPacket() {
GD.Print("test");
}
}
```
Client side :
```
public class MinimalClient : Node {
private readonly int PORT = 4041;
private readonly string IP = "127.0.0.1";
public NetworkedMultiplayerENet client;
public override void _Ready() {
this.client = new NetworkedMultiplayerENet();
Error error = client.CreateClient(IP, PORT);
if (error != Error.Ok) {
GD.PrintErr(error); return;
}
this.GetTree().SetNetworkPeer(client);
GD.Print("Connected to server " + IP + " and port " + PORT);
}
public override void _Input(InputEvent @event) {
if(this.client == null)
return;
if (!(@event is InputEventKey))
return;
InputEventKey e = (InputEventKey) @event;
if (e.IsPressed() && e.Scancode == (int) KeyList.Enter) {
String payload = "Hello world";
GD.Print("[CLIENT] Send tchat message with payload : " + payload);
GetTree().Multiplayer.SendBytes(Encoding.UTF8.GetBytes(payload));
}
}
}
```
**Action**: PRESS ENTER to send "Hello world" to server.
+ You can find a ZIP with the server and the client project.
+ In the client project, the UI is not important ( could be deleted )
[Networking.zip](https://github.com/godotengine/godot/files/4155967/Networking.zip)
| enhancement,topic:core,usability | low | Critical |
559,979,678 | go | x/pkgsite: invalidate cache for pkg.go.dev/<import-path> when a new latest version is available | ### What is the URL of the page with the issue?
https://pkg.go.dev/github.com/sethvargo/go-githubactions?tab=doc
### What is your user agent?
<pre>
Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.66 Safari/537.36
</pre>
### Screenshot
<img width="1223" alt="Screen Shot 2020-02-04 at 21 02 48" src="https://user-images.githubusercontent.com/3374574/73786763-d8b82c00-4791-11ea-88a5-25f10f1ddd65.png">
### What did you do?
Visited the above URL, shortly after `v0.1.1` was released (at the time the screenshot was taken: 2020-02-04 21.02.48 UTC)
### What did you expect to see?
Refreshing the page to take me to the latest version. `v0.1.1`.
### What did you see instead?
I was left on the `v0.1.0` page with a red link (shown in the screenshot) that prompted me to click through for the latest version.
To my mind the latest version should always be shown for the bare, non-canonical URL.
The red "latest" link should then show on the canonical version pages that are not the latest version. e.g.
https://pkg.go.dev/github.com/sethvargo/[email protected]?tab=doc
Related to #36807
| NeedsInvestigation,pkgsite | low | Minor |
559,997,792 | flutter | Investigate/Remove usage of sync WebDriver APIs | The flutter_driver web implementation currently relies on sync sockets: https://github.com/flutter/flutter/blob/master/packages/flutter_driver/lib/src/driver/web_driver.dart#L12
We probably shouldn't be using these, since it can hang the tool or test script to block on a socket connection. It seems like the async versions of web_driver incur additional overhead on closing, but that could be investigated as a bug on its own.
+ @angjieli did you ever discover the reason for the 15 second delay? | tool,t: flutter driver,c: proposal,P3,team-tool,triaged-tool | low | Critical |
560,013,379 | TypeScript | Narrowed is fail when use as `T` | <!-- 🚨 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:** 3.7.x-dev.201xxxxx
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:**
**Code**
```ts
export type IGunEpubNode = {
timestamp: number,
exists: false,
} | {
timestamp: number,
exists: true,
filename: string,
base64: string,
}
```
### this will fail
Error:(12, 10) TS2339: Property 'base64' does not exist on type 'IGunEpubNode'.
Error:(12, 18) TS2339: Property 'filename' does not exist on type 'IGunEpubNode'.
```ts
export function checkGunData<T extends IGunEpubNode>(data: T)
{
if (data && data.timestamp)
{
if (data.exists)
{
let { base64, filename, exists, timestamp } = data;
return true;
}
}
return null;
}
```
### this will work
```ts
checkGunData(data: IGunEpubNode)
```
**Expected behavior:**
both code should narrowed
```ts
checkGunData<T extends IGunEpubNode>(data: T)
checkGunData(data: IGunEpubNode)
```
**Actual behavior:**
this failed
```ts
checkGunData<T extends IGunEpubNode>(data: T)
```
**Playground Link:** http://www.typescriptlang.org/play/#code/KYDwDg9gTgLgBDAnmYcCSBxArgOwKJhYBGAchACaoC8cA3gLABQAkDAJYC2wAzjAIYcwALjg4sHIsCgAaJs1Bte3EQDM+AG27BZjAL5wAPnTnsuvAcNHjJMuQqUiYULNqZyVbdcBwDgI3lBsOADmOsxEfFoAbAAs-k5BoUy6boygkLBwKrgAxuwQOHA5ABbAOQDW2DgAInz8ADwAKnCgMN7k3OhVBMRklAB8ABTkdXwijQCUTAwsbCpww6NwAGTLcCP8AHSmPPyCUywzzMxzCxt8m-Yw3AfHR8de8LRwEdEx0lme3r4fV9wfO3Mgjg+ho5wA3KljswoMAYFgoIUnC5ISxmCkWBi5LD4YirOp1KjdEA
**Related Issues:** <!-- Did you find other bugs that looked similar? -->
| Bug | low | Critical |
560,033,032 | pytorch | [jit] Dict set item type mismatch error doesn't say the type was inferred | This error doesn't include the fact that the compiler inferred the type of `out_dict` to be `Dict[str, Tensor]`, but it should have
```python
def fn():
out_dict = {}
M = torch.ones(2)
out_dict[0] = M
return out_dict
```
```
RuntimeError:
Arguments for call are not valid.
The following variants are available:
aten::_set_item(Tensor[](a!) l, int idx, Tensor(b -> *) el) -> (Tensor[](a!)):
Expected a value of type 'List[Tensor]' for argument 'l' but instead found type 'Dict[str, Tensor]'.
aten::_set_item(int[](a!) l, int idx, int el) -> (int[](a!)):
Expected a value of type 'List[int]' for argument 'l' but instead found type 'Dict[str, Tensor]'.
aten::_set_item(float[](a!) l, int idx, float el) -> (float[](a!)):
Expected a value of type 'List[float]' for argument 'l' but instead found type 'Dict[str, Tensor]'.
aten::_set_item(bool[](a!) l, int idx, bool el) -> (bool[](a!)):
Expected a value of type 'List[bool]' for argument 'l' but instead found type 'Dict[str, Tensor]'.
aten::_set_item(t[](a!) l, int idx, t(b -> *) el) -> (t[](a!)):
Could not match type Dict[str, Tensor] to List[t] in argument 'l': Cannot match List[t] to Dict[str, Tensor].
aten::_set_item(Dict(str, t)(a!) l, str(b -> *) idx, t(c -> *) v) -> ():
Expected a value of type 'str' for argument 'idx' but instead found type 'int'.
aten::_set_item(Dict(int, t)(a!) l, int(b -> *) idx, t(c -> *) v) -> ():
Expected a value of type 'Dict[int, t]' for argument 'l' but instead found type 'Dict[str, Tensor]'.
aten::_set_item(Dict(float, t)(a!) l, float(b -> *) idx, t(c -> *) v) -> ():
Expected a value of type 'Dict[float, t]' for argument 'l' but instead found type 'Dict[str, Tensor]'.
aten::_set_item(Dict(Tensor, t)(a!) l, Tensor(b -> *) idx, t(c -> *) v) -> ():
Expected a value of type 'Dict[Tensor, t]' for argument 'l' but instead found type 'Dict[str, Tensor]'.
The original call is:
File "../test.py", line 34
out_dict = {}
M = torch.ones(2)
out_dict[0] = M
~~~~~~~~~~~~~~~ <--- HERE
return out_dict
```
cc @suo | oncall: jit,triaged | low | Critical |
560,054,435 | rust | MBE (macro_rules!) pattern-matching is unnecessarily O(n) even in simple cases. | There are some crates which do something like this (e.g. [`string_cache_codegen`](https://github.com/servo/string-cache/blob/af2c7707e797768660b3db90066b80218dbca6f7/string-cache-codegen/lib.rs#L271-L275)):
```rust
macro_rules! foo {
("a") => (A);
("b") => (B);
("c") => (C);
// ... etc. (maybe hundreds more)
}
```
(@nnethercote came across this when profiling `html5ever`'s compilation)
Also, prefixing patterns with `@foo` to create "internal rules" is common.
But the current implementation (AFAICT) has to check each of the patterns in turn.
Instead, we could build something like a decision tree ahead of time, from the constant (i.e. `$`-less) prefix of all arms, using maps for operators/idents/literals, to make the pattern-matching of the prefix amortize to being linear in the number of input tokens.
We could keep it simple by limiting the constant prefix to leaf tokens (no `(...)`, `[...]` or `{...}`).
For the example above, we'd pre-compute something like this:
```rust
DecisionTree::Match {
op: {},
ident: {},
literal: {
"a": DecisionTree::Done([0]),
"b": DecisionTree::Done([1]),
"c": DecisionTree::Done([2]),
// ...
},
}
```
Whereas for something using `@foo` rules it could be:
```rust
DecisionTree::Match {
op: {
'@': DecisionTree::Match {
op: {},
ident: {
"foo": DecisionTree::Done([0, 1]),
"bar": DecisionTree::Done([2, 3, 4]),
"baz": DecisionTree::Done([5]),
},
literal: {},
}
},
ident: {},
literal: {},
}
```
(where `DecisionTree::Done` indicates the arms to continue with)
cc @rust-lang/compiler (this might need a design meeting?) | C-enhancement,I-compiletime,A-macros,T-compiler | low | Major |
560,060,341 | kubernetes | say namespace of pods that could not be evicted to avoid search through all namespaces | when draining a node I often see errors and then have to grep through all pods in all namespaces to find out what the problem is (since `get pod foo --all-namespaces` does not work)
```
error when evicting pod \"foo\" (will retry after 5s): Cannot evict pod as it would violate the pod's disruption budget.
```
please add the namespace so it says `error when evicting pod \"foo\" from namespace \"bar\"` or something like that | kind/feature,sig/cli,lifecycle/frozen | low | Critical |
560,070,130 | TypeScript | Variable aliasing this should be treated the same as this | <!-- 🚨 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
this alias
## Suggestion
<!-- A summary of what you'd like to see added or changed -->
If I do `const me = this;` then `me` should be treated the same as `this` especially since it's a constant and cannot be updated to another value
## Use Cases
<!--
What do you want to use this for?
What shortcomings exist with current approaches?
-->
I'm using `--allowJs` and `--checkJs`. Right now TypeScript makes us do this:
```
constructor(chart, datasetIndex) {
this.chart = chart;
this._ctx = chart.ctx;
this.index = datasetIndex;
this._cachedAnimations = {};
this._cachedDataOpts = {};
this._cachedMeta = this.getMeta();
this._type = this._cachedMeta.type;
this._config = undefined;
this._parsing = false;
this._data = undefined;
this._dataCopy = undefined;
this._objectData = undefined;
this._labels = undefined;
this._scaleStacked = {};
this.initialize();
}
```
## Examples
<!-- Show how this would be used and what the behavior would be -->
I would like to be able to do this:
```
constructor(chart, datasetIndex) {
const me = this;
me.chart = chart;
me._ctx = chart.ctx;
me.index = datasetIndex;
me._cachedAnimations = {};
me._cachedDataOpts = {};
me._cachedMeta = this.getMeta();
me._type = this._cachedMeta.type;
me._config = undefined;
me._parsing = false;
me._data = undefined;
me._dataCopy = undefined;
me._objectData = undefined;
me._labels = undefined;
me._scaleStacked = {};
me.initialize();
}
```
This would make our transition to TypeScript much easier since our code follows this pattern throughout the codebase. It would also result in a smaller file size, which is why we'd adopted this pattern.
## 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 | Critical |
560,074,540 | rust | Support for fatal compile_error that suppresses subsequent errors | If a crate includes something like:
```rust
#[cfg(not(target_os = "linux"))]
compile_error!("This crate only runs on Linux")
```
then it isn't helpful for rustc to report that error but then go on to emit many more cascading errors about failed `use` statements and types that don't exist and similar.
It's especially unhelpful when those additional errors scroll the initial error off the screen. | A-diagnostics,T-compiler,C-feature-request | low | Critical |
560,093,543 | pytorch | torch batchwise max with indices | ## 🚀 Feature
A batchwise verison of torch.max() or torch.argmax() or any other OP like sum etc.
## Motivation
I have an array with elements coming from multiple examples where each example contributes an independent number of elements. I would like to have an efficient way to find the max element along with the index.
## Pitch
torch.max() should support batches of max operations instead of a single tensor
## Alternatives
A hacky alternative
c = torch.stack([torch.argmax(a[batch==i],dim=0,keepdim=True) for i in range(2)],dim=0) | triaged,module: batching,function request,module: reductions | low | Major |
560,120,128 | flutter | TextField “onChanged” was called multi times when i use Chinese inputting mode! | ## Steps to Reproduce
```dart
import 'package:flutter/material.dart';
class SearchInputWidget extends StatefulWidget {
final String preInputText;
final Function(String text) onSubmitted;
const SearchInputWidget({
@required this.onSubmitted,
this.preInputText
});
@override
State<StatefulWidget> createState() {
return new _SearchInputWidgetState();
}
}
class _SearchInputWidgetState extends State<SearchInputWidget> {
bool showCancel = false;
String inputText = "";
@override
void initState() {
super.initState();
this.inputText = (widget.preInputText != null ? widget.preInputText : "");
}
@override
Widget build(BuildContext context) {
print("searchInput build "+ this.inputText);
TextEditingController inputController = TextEditingController.fromValue(TextEditingValue(
text: inputText,
selection: TextSelection.fromPosition(TextPosition(
affinity: TextAffinity.downstream,
offset: '${inputText}'.length)
)
));
return new Container(
//margin: EdgeInsets.only(left: 40, top: 40),
alignment: Alignment(0, 0),
height: 44,
width: 300,
decoration: new BoxDecoration(
color: Colors.grey.shade200,
borderRadius: BorderRadius.all(Radius.circular(12.0)),
//border: new Border.all(width: 1, color: Colors.red),
),
child: new TextField(
//maxLines: 1,
controller: inputController,
maxLength: 30,
cursorColor: Colors.black,
style: TextStyle(fontSize: 16, color: Colors.black),
onChanged: (text){
print("searchInput onChanged "+ text);
setState(() {
print("searchInput setState "+ text);
inputText = text;
showCancel = (text.length > 0);
});
},
onSubmitted: (text) {
widget.onSubmitted != null && widget.onSubmitted(text);
},
decoration: InputDecoration(
prefixIcon: Icon(
Icons.search,
color: Colors.black45,
size: 24,
),
suffixIcon: new Offstage(
offstage: !showCancel,
child: new IconButton(
icon: Icon(
Icons.cancel,
color: Colors.black45,
size: 20,
),
onPressed: _onClickClear),
),
border: InputBorder.none,
/* fillColor: Colors.red,
filled: true,*/
counterText: "",
),
),
);
}
void _onClickClear() {
setState(() {
inputText = "";
showCancel = false;
});
}
}
```
but the same code run well if i input in english inputting mode.
```bash
flutter: searchInput build
flutter: searchInput onChanged q
flutter: searchInput setState q
**Actual results:** <!-- what did you see? -->
when i input "q" in Chinese inputting mode,the text in the TextField is "qq",here is the log.
flutter: searchInput onChanged q
flutter: searchInput setState q
flutter: searchInput build q
flutter: searchInput onChanged qq
flutter: searchInput setState qq
flutter: searchInput build qq
```
| a: text input,platform-ios,engine,a: internationalization,has reproducible steps,P2,found in release: 3.7,found in release: 3.10,team-ios,triaged-ios | low | Major |
560,154,802 | flutter | TextOverflow.ellipsis not applied properly with multi-line Text with maxLines set | ```
Container(
constraints: BoxConstraints(
maxWidth: MediaQuery.of(context).size.width * 0.55,
),
child: Text(
"some text that changes lines \n\n\n\n a new line",
style: TextStyle(
color: Colors.black,
fontSize: 14,
),
maxLines: 1,
overflow: TextOverflow.ellipsis
),
);
```
shows the following:
with no ellipsis
the maxLines property states:
An optional maximum number of lines for the text to span, wrapping if necessary. If the text exceeds the given number of lines, it will be truncated according to [overflow].
so it should show the ellipsis.
this is reproducible in https://dartpad.dev/ | framework,engine,a: typography,has reproducible steps,P2,found in release: 3.7,found in release: 3.9,found in release: 3.10,team-engine,triaged-engine | low | Critical |
560,178,125 | TypeScript | Accumulating type in TypeScript like powerful instrument for organizing code | ## Main Idea
Accumulating type is a type that combines the types of declarations that occur in different parts of the code with the same name. Type merging is done according to the OR principle.
Principles and dogmas:
* related code in one place
* organizing separate and independent modules
* write once use everywhere
* intermodular interaction through types
## Search Terms
accumulate type, type system
## Suggestion
Add the ability to declare **accumulative** types
## Examples
```typescrypt
// file: moduleAFolder/moduleA.ts
accumulate type ModuleNames = "ModuleA"; // type to be "ModuleA" | "ModuleB"
// file: moduleAFolder/moduleAHelper.ts
import {ModuleNames} from "moduleAFolder/moduleA";
// ^ powerful part of example: importing type ONLY from moduleA, NOT from ModuleB
// ^ accessible value for ModuleNames here are "ModuleA" and "ModuleB"
const a: ModuleNames = "ModuleA"; // it's ok
const b: ModuleNames = "ModuleB"; // it's ok.
const c: ModuleNames = "Unknown"; // it's error
// file: moduleBFolder/moduleB.ts
accumulate type ModuleNames = "ModuleB"; // type to be "ModuleA" | "ModuleB";
```
Example above is very simple but illustrates all principles good code. In more difficalt cases we have 100 modules and 1000 types. Some of them has types that need in another modules.
Yes, we can create folder with name 'shared' and move common types to this folder:
* but then we destroy the integrity of the code inside the separate module. Everything that looked whole before that became fragmented;
* but then we make the “shared” folder a trash or garbage;
Yes, we can copy common types from one module to another:
* but then we make smelling bad code;
* but then we make bug prone code;
## More difficult example
```typescrypt
// file: moduleAFolder/moduleA.ts
accumulate type CrossModuleTypes = {
"ModuleA": {
id: string;
data: {
value: string;
}
}
}
// file: moduleBFolder/moduleB.ts
accumulate type CrossModuleTypes = {
"ModuleB": {
id: string;
segment: {
part: string;
}
}
}
// file: moduleCFolder/moduleC.ts
// no imports here!
accumulate type CrossModuleTypes; // nothing merge, only usaging;
type UnionToIntersection<U> = (U extends any ? (k: U)=>void : never) extends ((k: infer I)=>void) ? I : never
type ModuleBType = UnionToIntersection<CrossModuleTypes>["ModuleB"];
const cachingData: ModuleBType = this.loadFromCache("ModuleB");
console.log("Caching segment from ModuleB is ", cachingData.segment.part); // no errors here.
```
## 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,In Discussion | low | Critical |
560,187,127 | flutter | Need a line/word breaker for custom text rendering | This is a more specific issue that is a part of #35994. For doing custom text rendering on non-rectangular or non-rtl/ltr surfaces, developers need to measure the text that can fit in each line. In order to measure the text, developers need to know appropriate locations in the text where line breaks can occur. This functionality already exists in LibTxt at the Flutter Engine layer, but it is not exposed at the Dart layer.
## Use cases
Anywhere that non-rectangular or non-rtl/ltr text rendering needs to be done (that is, anywhere that the Flutter `Paragraph` cannot be used to render the entire text directly):
### Example 1: Vertical CJK and Mongolian
Image adapted from [here](https://www.w3.org/International/articles/vertical-text/).
### Example 2: Filling a non-rectangular shape with text
[Image source](https://www.raywenderlich.com/5960-text-kit-tutorial-getting-started)
### Example 3: Text around exclusion paths
[Image source](https://developer.apple.com/library/archive/documentation/StringsTextFonts/Conceptual/TextAndWebiPhoneOS/CustomTextProcessing/CustomTextProcessing.html)
## Problem
Given a long string, how does one know where appropriate places to line wrap are?
If the words in the string are separated by spaces like this
```
final myString = 'This is a long string in which each word is separated by spaces.';
```
Then it is easy to find break locations. Any letter with a space before it is fine. However, there are lots of other situations that are not so easy. Here is one example:
```
final myString = 'This-is_a,long.string(in)-which{each}word[is]NOT~separated#by+spaces.';
```
Writing the logic to find the breaks becomes much more complex, especially if [Unicode conformance](https://unicode.org/reports/tr14/) is required.
## Proposal
In order not to require developers to reinvent the wheel when the solution already exists in the Flutter engine, I propose that Flutter expose the LibTxt Line/Word breaker functionality at the Dart level.
| c: new feature,framework,engine,a: internationalization,a: typography,c: proposal,P2,team-engine,triaged-engine | low | Major |
560,212,366 | go | x/review/git-codereview: package documentation for multi-commit branch workflow does not match behavior | The package documentation for `git-codereview` [says](https://godoc.org/golang.org/x/review/git-codereview#hdr-Multiple_Commit_Work_Branches):
> The 'git codereview change' command amends the top commit in the stack (HEAD).
However, [CL 20049](https://golang.org/cl/20049) appears to have removed this behavior, instead stating that some combination `git commit --amend` and `git rebase` should be used instead.
I'd appreciate it if these docs were updated to better explain the current behavior (potentially with an example).
I'm working on a multi-change contribution, and am finding this a little bit confusing to figure out. | Documentation,help wanted,NeedsFix | low | Minor |
560,217,367 | TypeScript | TypeScript: improve refactor-rename behavior for merged declarations | <!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Describe the feature you'd like. -->
Currently, if you have e.g. a function-namespace merged declaration, or an interface-namespace merged declaration, attempting to refactor-rename one component of the merged declaration keeps the other components name unchanged. For example, if you rename the function, the merged namespace doesn't rename.
The result is that the merged declaration becomes unmerged.
An unfortunate consequence of this is that other files referencing the merged declaration don't seem to be able to decide which name to keep, with the current behavior being keeping the name in its usage (e.g. a function call).
My proposal is that the behavior change to renaming every component of the merged declaration, such that the overall merged property of the declaration is not destroyed.
My justification for it is that the developer has intentionally merged the declaration, and the rename action should preserve this intention. If the developer intentionally chooses to break the merge, it is that developer's responsibility to manually ensure all references `import` the right former merge components' new names after the unmerge, this being something the machine likely cannot accurately decide on the human's behalf.
### Use cases
I often do merge declarations to keep declarations and implementations in the same file while keeping the export space relatively tidy. The intention is to scope names that are particular to some principal functionality instead of having e.g. `myFunction` , `ResultForMyFunction`, `ArgForMyFunction` polluting the export space (and thereby the auto-import suggestions).
In all cases below, it makes sense that a refactor-rename attempt would rename all components of the merged declarations.
Here is an example for a React component and its props.
```ts
export namespace SomeReactComponent {
export interface Props {
// ...
}
}
export function SomeReactComponent(props: SomeReactComponent.Props){
/* ... */
}
```
Here is another example of an API payload schema and its runtime declaration
```ts
export namespace SomeApiPostPayload {
export function validate(value: unknown): value is SomeApiPostPayload {
//
}
}
export interface SomeApiPostPayload {
// ...
}
```
This is also especially useful when you have a function that relies on some kind of external context for dependency inversion and you would like to employ interface segregation hierarchically so that what you need to provide just kind-of "emerges":
```ts
// foo.ts
export namespace foo {
export interface Context {}
}
export function foo(/* ..., */ context: foo.Context){
// ...
}
```
```ts
// bar.ts
export namespace bar {
export interface Context {}
}
export function bar(/* ..., */ context: bar.Context){
// ...
}
```
```ts
// foobar.ts
import { foo } from "./foo.ts";
import { bar } from "./bar.ts";
export namespace foobar {
export interface Context extends foo.Context, bar.Context {}
}
export function foobar(/* ..., */ context: foobar.Context){
// ...
}
```
Or in the cases where result and error types are particular to a module rather than a project-wide shared data structure.
```ts
export namespace foo {
export interface Result { /* ... */ }
export interface Error { /* ... */ }
}
// assume type Failable<Success, Failure> =
// ({ ok: true } & Success) | ({ ok: false } & Failure)
// or some such notion
export function foo(): Failable<foo.Result, foo.Error> {
/* ... */
}
``` | Suggestion,Awaiting More Feedback | low | Critical |
560,261,161 | godot | ColorRect flickers when shown by mouse_entered signal | **Godot version:**
3.2 Stable
**OS/device including version:**
macOS Catalina 10.15.3 / MacBookPro11,1
**Issue description:**
When calling show() inside a function on a ColorRect node, from mouse_entered signals, the ColorRect node flickers.
**Steps to reproduce:**
Use the Godot 3.2 project attached.
**Minimal reproduction project:**
[Godot 3.2.zip](https://github.com/godotengine/godot/files/4158725/Godot.3.2.zip)
Video:
[mouse_entered_flicker.mov.zip](https://github.com/godotengine/godot/files/4158730/mouse_entered_flicker.mov.zip)
| discussion,regression,topic:gui | low | Major |
560,290,336 | rust | Improve diagnostics for serializing 'static type from Deserialize<'de> | With this code:
```rust
use serde_json;
use serde::Deserialize;
fn foo<'de, D>(b: Vec<u8>) where D: Deserialize<'de> {
let _ = serde_json::from_slice::<D>(b.as_ref()).unwrap();
}
```
We get this error message:
```rust
error[E0597]: `b` does not live long enough
--> src/lib.rs:5:41
|
4 | fn foo<'de, D>(b: Vec<u8>) where D: Deserialize<'de> {
| --- lifetime `'de` defined here
5 | let _ = serde_json::from_slice::<D>(b.as_ref()).unwrap();
| ----------------------------^----------
| | |
| | borrowed value does not live long enough
| argument requires that `b` is borrowed for `'de`
6 | }
| - `b` dropped here while still borrowed
```
([Playground](https://play.rust-lang.org/?version=stable&mode=debug&edition=2018&gist=707cd7c8bdd48c7f2f79ffe4fa211416).)
My colleague came to me yesterday asking me to explain this error/help fix the problem. Eventually, I figured out that method call tries to propagate the lifetime bound and fails because the target doesn't have any lifetime bounds. I solved it by applying a HRTB (`D: for<'de> Deserialize<'de>`), which worked out okay.
The errors here are kind of misleading, because the borrowed value does actually live long enough, and in this example `b` is not actually still borrowed at the end of the function.
cc @estebank | C-enhancement,A-diagnostics,A-lifetimes,T-compiler,D-confusing,D-newcomer-roadblock | low | Critical |
560,370,257 | flutter | Add FloatingActionButton to the end of FlexibleSpaceBar | <!-- Thank you for using Flutter!
Please check out our documentation first:
* https://flutter.dev/
* https://api.flutter.dev/
If you can't find the answer there, please consider asking a question on
the Stack Overflow Web site:
* https://stackoverflow.com/questions/tagged/flutter?sort=frequent
Please don't file a GitHub issue for support requests. GitHub issues are
for tracking defects in the product. If you file a bug asking for help, we
will consider this a request for a documentation update.
-->
I'm using SliverAppBar with FlexibleSpaceBar as 'flexibleSpace' in my Flutter-app. I want to add FloatingActionButton to the end of FlexibleSpaceBar just like in native android development like on the screenshot below. The problem is that I don't know how to do it. I haven't found solution in docs.
| c: new feature,framework,f: material design,d: stackoverflow,c: proposal,P3,team-design,triaged-design | low | Critical |
560,382,882 | pytorch | matmul: no warning when contracting differently named dimensions | ## 🐛 Bug
According to [documentation](https://pytorch.org/docs/stable/named_tensor.html) named tensors "use names to automatically check that APIs are being used correctly at runtime, providing extra safety". However, `matmul` does not warn when the dimensions that are being contracted have different names.
## To Reproduce
Steps to reproduce the behavior:
```python
a = torch.randn((2, 3), names=('B', 'X'))
b = torch.randn((3, 5), names=('C', 'E'))
torch.matmul(a, b)
```
This executes without any warning.
## Expected behavior
A warning about non-matching dimensions might be useful
## Environment
PyTorch version: 1.4.0
Is debug build: No
CUDA used to build PyTorch: 10.1
OS: Ubuntu 18.04.3 LTS
GCC version: (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0
CMake version: version 3.12.0
Python version: 3.6
Is CUDA available: Yes
CUDA runtime version: 10.0.130
GPU models and configuration: GPU 0: Tesla T4
Nvidia driver version: 418.67
cuDNN version: /usr/lib/x86_64-linux-gnu/libcudnn.so.7.6.5
Versions of relevant libraries:
[pip3] numpy==1.17.5
[pip3] torch==1.4.0
[pip3] torchsummary==1.5.1
[pip3] torchtext==0.3.1
[pip3] torchvision==0.5.0
[conda] Could not collect
cc @zou3519 | triaged,module: named tensor | low | Critical |
560,387,258 | go | net: consider rejecting null byte in LookupHost (and similar) on all platforms consistently | This is a followup to issues #31597 and #37031.
It is my current understanding that no valid host can contain the null byte ('\x00'), so any such string input is not valid.
`net.LookupHost` currently operates differently when given an input that contains a null byte. For example, for `net.LookupHost("foo\x00bar")`:
- in Go 1.13, on Windows, it caused a panic (#31597)
- in Go 1.14, on Windows, it returns an error (see [here](https://github.com/golang/go/blob/go1.14beta1/src/net/lookup_windows.go#L107))
- on macOS with cgo DNS resolver, it does the equivalent of `net.LookupHost("foo")`
- on macOS with pure Go DNS resolver, it does something different
It seems that the `getaddrinfo` C API used on many platforms accepts a null-terminated C-style string and cannot be given a string containing null bytes.
As @ianlancetaylor points out, in the `syscall` package we reject strings with embedded null bytes, since they won't work with system calls that expect C strings, see [`syscall.ByteSliceFromString`](https://github.com/golang/go/blob/go1.13.7/src/syscall/syscall.go#L44-L56).
Perhaps this is an opportunity to make Go operate more consistently on such inputs on all platforms by reporting an error when the `name` string contains a null byte. There may be more functions in `net` that could benefit from this input validation, but to be able to make this change, we need to be confident that this change won't break correct Go programs.
/cc @FiloSottile @ianlancetaylor @mikioh @bradfitz | NeedsInvestigation | low | Critical |
560,402,031 | pytorch | Upper/Lower attributes for named dimensions for proper Ricci notation and to generalize matrix operations | ## 🚀 Feature
1. Add an "upper|lower|unspecified" attribute to each (named) tensor dimension.
2. Honor the upper/lower distinction when checking dimension equality (upper!=lower, upper==unspecified, lower==unspecified, and of course lower==lower, upper==upper) in operations such as add, mul
3. Provide a method to match pairs of upper and lower dimensions which share the same name
4. Make all operators and algorithms specified on matrices available to all tensors with more than one upper and lower dimension (transpose, inverses, decompositions,...)
## Motivation
In multilinear/tensor algebra, there is a clear distinction between upper and lower indices, which (among other things) can be understood to distinguish between input and output dimensions.
By having a distinction between inputs and outputs, we can generalize any operations on linear maps (i.e. matrices) to tensors of arbitrary dimensionality, e.g. we can apply the matrix transpose, matrix inverses, matrix decompositions, and in general any algorithm defined on matrices in a mathematically consistent way:
**Flatten:**
Reshapes any tensor into a _matrix_ with row index = cartesian product of all upper indices and column index = cartesian product of all lower indices.
Note that with this definition of flattening, we can generalize any matrix operation to any tensor, e.g. transposes, inverses, decompositions,... by applying it to the flattened tensor and reshaping (folding) it afterwards.
Note 2: The current flat() method basically assumes that all dimension have the same attribute value
**Transpose:**
Flip the upper/lower attribute (upper to lower and lower to upper)
Note: this is the equivalent of swapping row and column in a matrix.
Note: Currently, transpose() reverses the order of dimensions in the underlying data array by default. Dimension order should be an irrelevant implementation detail in a named tensor scheme, though (and maybe left to a compiler for optimization). For future compatibility, reordering could be restricted to tensors with at least one "unspecified" dimension.
**Inverses:**
Invert the flattened tensor, then fold it back into its original shape. Upper/lower attributes get flipped.
Note: It may make sense to implicitly transmogrify some index names of the resulting output tensor, so that contracting the tensor with its inverse yields another tensor (instead of a scalar). for example, the lower indices of a left-sided pseudoinverse would match upper indices of its operand, but upper indices would not match the lower indices of the operand. This asymmetric name change effectively encodes the "left-sidedness" of the inverse, as tensor contraction (in contrast to matrix multiplication) is commutative.
**contraction/tensordot:**
Tensordot sums over pairs of indices with the same name but differing upper/lower attribute. For backwards compatibility, tensordot matches names with "unspecified" attribute with any other name.
(Here, no additional functionality is gained over named indices, but it makes tensordot a proper implementation of the tensor contraction as defined in the Ricci-notation)
**Coordinate transformation / Change of basis**
Coordinate transformation of tensors can be done automatically, as the upper/lower attribute distinguishes contravariant from covariant indices, i.e. it completely specifies how the coordinate values of a tensor need to be transformed under a change of basis. This is an inherent feature of proper tensors and extensively used in Physics and Mechanics,
A complementary purpose of the upper/lower attribute is to explicitly distinguish between maps (inputs and outputs), metrics(only inputs), and multidimensional data (only outputs). Invalid operand combinations (in the Ricci notation and multilinear algebra sense) could raise exceptions.
Unfortunately, we cannot simply use some naming scheme to handle the upper/lower distinction, as the semantics with respect to matching dimensions for tensor contraction differ (names use *equality* while upper/lower uses *inequality*).
## Pitch
Currently, no popular "tensor" library implements tensors such that they are fully compatible with the Ricci-notation, which is the most common notation to specify tensor equations. Pytorch already is well positioned though by the recent addition of named dimensions. Adding (optional) upper/lower attributes to each dimension would allow Pytorch users to implement tensor equations verbatim, improving code readability and avoiding bugs when implementing textbook equations.
From a practical, software engineering perspective, the upper/lower attribute provides an additional check when constructing elaborate computation graphs by distinguishing whether a dimension is supposed to consume data (input dimension, or lower index), or is supposed to provide data (output dimension, or upper index).
For tensors with two dimensions, "upper" and "lower" is equivalent to the row-and-column distinction in matrix notation, so it is a straightforward and robust generalization.
In the far future, the change may also enable Pytorch to natively support algorithms from multilinear algebra, e.g. tensor decomposition algorithms.
## Alternatives
The alternative is to comment-code the input-output direction semantics, similar to how dimension semantics had to be handled before introducing named dimensions. This is obviously error-prone and bad from a code-as-documention perspective.
cc @zou3519 | triaged,module: named tensor | low | Critical |
560,407,894 | TypeScript | TS2536 thown when using keyof on a property of a generic type as key | <!-- Please try to reproduce the issue with `typescript@next`. It may have already been fixed. -->
**TypeScript Version:** 3.8.0-dev.20200204
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:**
* TS2536
* generic typing
* keyof
**Code**
```ts
enum Switch {
A = 'a',
B = 'b',
}
interface ResultA {
keyA1: number;
keyA2: string;
}
type ResultB = number[];
type ResultType = ResultA | ResultB;
interface GenericBase {
switch: Switch;
value: ResultType;
}
interface GenericA extends GenericBase {
switch: Switch.A;
value: ResultA;
}
interface GenericB extends GenericBase {
switch: Switch.B;
value: ResultB;
}
type GenericType = GenericA | GenericB;
function getValue<T extends GenericType, K extends keyof T['value']>(result: T, key: K): T['value'][K] {
// T = GenericA
// T['value'] = ResultA
// keyof ResultA = 'keyA1' | 'keyA2'
// T = GenericB
// T['value'] = ResultB
// keyof ResultB = number | "toString" | [...]
// unexpected: error TS2536: Type 'K' cannot be used to index type 'ResultType'.
return result.value[key];
}
const a: GenericA = {switch: Switch.A, value: {keyA1: 1, keyA2: 'A'}};
const b: GenericB = {switch: Switch.B, value: [1, 2, 3]};
// no errors
console.log(getValue(a, 'keyA1'));
console.log(getValue(a, 'keyA2'));
console.log(getValue(b, 0));
console.log(getValue(b, 3));
// error TS2345: Argument of type '0' is not assignable to parameter of type '"keyA1" | "keyA2"'.
// => expected, and the compiler does know what keys can be used
console.log(getValue(a, 0));
// error TS2345: Argument of type '"keyA1"' is not assignable to parameter of type 'number | "toString" [...]
// => also expected
console.log(getValue(b, 'keyA1'));
```
**Expected behavior:**
As long as the correct keys are used, which is covered by TS2345, the index access should not raise an error.
**Actual behavior:**
TS2536 is thrown when compiling, altough the keys should be known due to the generic typing.
**Playground Link:** [typescriptlang.org](http://www.typescriptlang.org/play/?ts=3.8-Beta&ssl=31&ssc=52&pln=31&pc=57#code/KYOwrgtgBAyg7gSwC4GMAWUDeAoKeoCCUAvFAOQCGZANLvgEInkBGN2AvttgiEsAE4AzCimBQASsADOYADZIiOfFADWwAJ4EAjAC4o4CMwEBuOnjWaATHqlJ+PAOanO2JOoAOYyTPmNSBo34AbQBdUzdPCWk5JAAVDzFSbxiiAB8onyR6U25eAWFRKABxUAEEFHoKKTElfClEVDQ9eGR0U2UANwpZMGA9ZPl4z2cuHj4hETESkDKUImAADz4QABMpYtL7Cqqasyh61qbYBvQAOgJ2-C6evoyUkdzxgqnN8sZF5bWNma3K6qw9gdGs0TmhTtk9tdev1or4HhEXj9ykNEt9Zmk0b8coIwCAUEgEAB7EBQBzAJAANW6vQAPLEoB9QF9prMUdQoABpBlLJnrCyEwRQWJBMhQ4BkEIAPgAFPxYUg9LF2RY9ByAJSKkViiVBDkhAHKAD0hqguMWnnxwBWegE-EJ-CFMEsAFYAMwANkVCXIHLIUBQFBAIEJSCgRlN1RWUCQhKgPBWi2j3rIAziCTIpz2cqQYH4JLlmVOYqCFjCHC4KGJtigFD0LK2RFImCB6BBh3O7LFekwFm0ei0yo0BGs5AIZHY7FMlZA1eYddeFSYzdBbca4M71NuQQHUEs7NdIUnXGN+ljtvtUmw06khNkwFOskJDmlZMpm+lFHZZF7WjIarVU5Vre96Ps+r5UjcH5fr2lh-gBV5AXeD5Pi+5IQb00rMOyAAM-6ATOwHIWBaHvlhUCunhx4mueDqxE6roACzOnoBD8A4kCgKGApJpEZDYX6CDrMGoZVFICAOCAFDMHe0axu4FD8BQEDkgIUDcQi5AAEQ-ppUDpNpQ6WJpGbYCexCStyFp8Cs7KBlGSBoGIlYQO4CB3g6KyEtIqjBnAUBwGgFChhY6wBiS4ZgJGCEEUhoGoW+kGflAuHwaZ1H8HatH0UxLFsRxvBqYKGlkAZmhaMZcZCSGNZSGJElSTJMZQPJinKeMhU8WIZABKp+kxjAdiOLpQSnKNIRpSQFndDelnAJaKzRTesUoeBpHQUOv6UUAA)
**Related Issues:**
* possibly, but not quite: https://github.com/microsoft/TypeScript/issues/33521
* more likely, but also not quite: https://github.com/microsoft/TypeScript/issues/21760 | Bug | low | Critical |
560,414,866 | TypeScript | JSDoc @param does not work for function variables | <!-- 🚨 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:** 3.5.1, 3.7.5
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:** jsdoc, param, variable, function
**Code**
```ts
function makeFoo() {
return (x : number) => {}
}
/**
* @param x - Comment
*/
const foo = makeFoo();
foo(/* No comment in hover */);
/**
* @param x - Comment
*/
function foo2(x: number) {
}
foo2(/* Comment in hover */)
```
**Expected behavior:**
**Actual behavior:**
**Playground Link:** [Playground](http://www.typescriptlang.org/play/?ssl=1&ssc=1&pln=17&pc=29#code/GYVwdgxgLglg9mABAWwIYGsCmAxOcAUAlIgN4CwAUItYgE6ZQi1L4AeiAXImCMgEaZaxALwA+UgF9KUipQD0AKgWVECxAAEADqlqpkidgFpEAYTjJkmMFBUK5lCAgDOURMDyJhKDDjxEA3JTuBIqIAHJwiI4WVq4wSAAWcABugqpyhIGyFIrKVGpaOnoGiMZmMda29hSgkLAIbngATGxcPPyCxORUNNJBzfih5ZbWiPGISam06YRAA)
**Related Issues:** Couldn't find any
-----
For my use case, I use factories to make functions a lot.
```ts
export const lPad = makeOperator3<string, bigint, string, string>(/*args*/);
export const rPad = makeOperator3<string, bigint, string, string>(/*args*/);
export const lTrim = makeOperator1<string, string>(/*args*/);
export const rTrim = makeOperator1<string, string>(/*args*/);
```
And I'd like to add JSDoc to these variables, because the parameter names aren't very descriptive, by themselves (`left, mid, right, arg, etc.`). But it seems like the JSDoc for `@param` does not show up in hover on VS Code and the Playground. | Needs Investigation | low | Critical |
560,435,976 | go | all: builders and TryBots should check documentation for broken links | We occasionally end up with broken or outdated links in documentation (see #46545, https://github.com/golang/go/issues/37042#issuecomment-582448476, #27860, #21951, #19244).
Testing for broken links would not be entirely trivial, but nonetheless pretty straightforward to implement: we have an HTML parser in `golang.org/x/net/html`, and it is easy enough to issue a `HEAD` request for link targets to see if they resolve. (For links with anchors, we would probably want to also check the `Content-Type` header and then parse the actual linked HTML to ensure that the anchor exists.)
CC @golang/osp-team | Testing,Documentation,help wanted,Builders,NeedsInvestigation | low | Critical |
560,437,899 | kubernetes | Reevaluate flushing unschedulable pods into activeQ | /sig scheduling
#72122 suggested the addition of flushing unschedulable pods into the active queue, to mitigate against unknown race conditions that would keep pods in the unschedulable queue forever. The scenario I can think of is that an event that makes the pod more schedulable comes while we are still in its scheduling cycle. But this should have been addressed with #73078.
In any case, as seen in #86373, flushing indistinctively can produce starvation if there are too many unschedulable pods with high-priority, or if scheduling takes too long due to the use of advanced features (like extenders).
So I suggest any of these 2 options:
- [Extreme] Get rid of the flush, since we don't have evidence that it's actually helping more than #73078.
- Limit the number of pods we flush back. We would always flush the oldest ones. Either a (configurable?) hard limit (always flush X) or a dynamic limit (stop once the activeQ is past a threshold). This should be low risk and easy to do. | sig/scheduling,lifecycle/frozen,needs-triage | high | Critical |
560,462,454 | flutter | Web: Cache-able code splitting, Tree Shaking and Routes grouping | Currently Flutter for web build command generates a big single Javascript file in release mode.
It is proposed that new routing APIs should allow groping of related routes and then the build command should generate separate Javascript files for each group of routes these files can than be pre-fetched separately. This will help to improve Time to Interactive TTI score and defer or remove unnecessary JavaScript work that occurs during initial page load. | c: new feature,framework,c: performance,dependency: dart,f: routes,customer: crowd,platform-web,c: proposal,a: build,P2,team-web,triaged-web | low | Critical |
560,493,241 | go | x/tools/go/analysis/analysistest: support modules | Currently the analysistest helper only supports a GOPATH like tree. This is fine for stand alone analyses, but does not help much for analyses designed to run against a particular module. For example, you have an analysis that checks all code in your module correctly/safely interacts with a specific package in your module. Ideally you would be able to write an analysis test with a stub file that is able to load other packages/dependencies of the current module.
/cc @matloob @heschik
| help wanted,NeedsInvestigation,Tools,Analysis | medium | Critical |
560,499,146 | flutter | RefreshIndicator hidden behind AppBar with `extendBodyBehindAppBar: true` | I found this issue that the `RefreshIndicator` doesn't work properly combined with the Scaffold property `extendBodyBehindAppBar: true`.
Here is the code snippet that causes the error. In total the RefreshIndicator is pulled down from above the status bar and therefore the indicator has a long way till being dragged down to the visible body part. If the user drags only half the way the indicator can also be behind the app bar which is for sure not an intended behaviour.
```
Scaffold(
extendBodyBehindAppBar: true, // true causes a layout error
appBar: AppBar(...),
body: RefreshIndicator(
onRefresh: () {
// some logic
},
child: ListView(
children: <Widget>[
``` | framework,f: material design,has reproducible steps,P2,found in release: 3.3,workaround available,found in release: 3.7,team-design,triaged-design | low | Critical |
560,522,096 | storybook | Add support for Vue SFC syntax to stories | **Is your feature request related to a problem? Please describe.**
When writing a story for Vue components inside mdx files, it differs from the way one would write the code inside a .vue file.
Also when reading the story source inside storybook, the code differs from the code one would normally put inside a .vue file.
So there is this mental switch developers have to do, when using storybook as their component documentation. This is especially troublesome for new developers, which in many teams are the main target group of the documentation.
**Describe the solution you'd like**
In addition to writing a Vue story like this:
```mdx
<Story name='basic' height='400px'>{{
components: { InfoButton },
template: '<info-button label="I\'m a button!"/>',
}}</Story>
```
It would be nice to add support for stories in this format:
```mdx
<Story name='basic' height='400px'>
<template>
<info-button label="I\'m a button!"/>
</template>
<script>
export default {
components: { InfoButton },
}
</script>
</Story>
```
**Are you able to assist bring the feature to reality?**
yes | feature request,vue,addon: docs | medium | Critical |
560,522,320 | flutter | Document how to publish module App and plugins frameworks in dependency management systems like CocoaPods and Swift Package Manager | <!-- 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
Our Flutter application is located in another repository and we use `cocoapods` to manage our dependencies, It is possible to create a flag to generate `podspecs` for all frameworks? It helps us to use our dependencies remotely in separate repositories.
<!--
Please tell us the problem you are running into that led to you wanting
a new feature.
Is your feature request related to a problem? Please give a clear and
concise description of what the problem is.
Describe alternative solutions you've considered. Is there a package
on pub.dev/flutter that already solves this?
-->
## Proposal
<!--
Briefly but precisely describe what you would like Flutter to be able to do.
Consider attaching images showing what you are imagining.
Does this have to be provided by Flutter directly, or can it be provided
by a package on pub.dev/flutter? If so, maybe consider implementing and
publishing such a package rather than filing a bug.
-->
Recently Flutter created `--cocoapods` flag to create a podspec to Flutter. It is possible to create the same way to create podspec for `App.framework`, `plugins` and `FlutterPluginRegistrant.framework`? If possible, create `.zip` file to upload these files in repositories.
**Approach B:** Using [option B](https://flutter.dev/docs/development/add-to-app/ios/project-setup#option-b---embed-frameworks-in-xcode)
```bash
flutter build ios-framework --cocoapods
```
Output:
```text
some/path/MyApp/
└── Flutter/
├── Build Mode/
│ ├── Flutter.podspec
│ ├── App.zip (include App.framework, FlutterPluginRegistrant.framework and all plugins)
│ └── App.podspec (include App.framework, FlutterPluginRegistrant.framework and all plugins)
```
Podfile with locale path:
```ruby
pod 'Flutter', :podspec => 'some/path/MyApp/Flutter/{build_mode}/Flutter.podspec'
pod 'App', :podspec => 'some/path/MyApp/Flutter/{build_mode}/App.podspec'
```
Podfile with remote path:
```ruby
pod 'Flutter', :git => 'repository/Flutter.podspec'
pod 'App', :git => 'repository/App.podspec'
```
**Approach A:** Using [option A](https://flutter.dev/docs/development/add-to-app/ios/project-setup#option-a---embed-with-cocoapods-and-the-flutter-sdk)
Is possible to use `podhelper.rb` remotely?
```podspec
# MyApp/Podfile
target 'MyApp' do
install_all_flutter_pods(git_path)
end
``` | c: new feature,platform-ios,tool,d: api docs,a: existing-apps,c: proposal,P3,team-ios,triaged-ios | medium | Critical |
560,578,165 | godot | Import TextureAtlas as Mesh is not working properly | **Godot version:**
3.2 stable
**OS/device including version:**
Windows10 x64 1809
**Issue description:**
Not sure that issues are the same: #32869 , #35756
Importing TextureAtlas as mesh gives an incorrect results when used in the Sprite and in the AnimatedSprite. The texture is shifted (1) and some frames are distorted and skipped when animation is playing. AnimationFrame panel looks wrong (2)
**Output**:
` drivers/gles3/rasterizer_storage_gles3.cpp:3376 - Condition "!(p_format & VisualServer::ARRAY_FORMAT_VERTEX)" is true.`
**Debugger:**
`mesh_add_surface: Condition "!(p_format & VisualServer::ARRAY_FORMAT_VERTEX)" is true.`
**Steps to reproduce:**
- Import TextureAtlas as mesh
- Use imported textures with Sprite node or AnimatedSprite node as AnimationFrames
**Minimal reproduction project:**
Launch Scene.tscn and switch between the correct AnimatedSprite_texture node (frames imported as textures) and the AnimatedSprite_atlas_mesh node to see the difference:
[atlas.zip](https://github.com/godotengine/godot/files/4161524/atlas.zip)
| bug,topic:core,topic:rendering,confirmed,topic:2d | low | Critical |
560,612,191 | TypeScript | Ensure generated `.d.ts` files compile with no errors under `noImplicitAny` | We recently just had https://github.com/microsoft/TypeScript/issues/36630 opened up on us. This is just a few weeks after #36216.
I'm pretty sure we already check that our generated `.d.ts` files are valid when the input was valid, but apparently we're not checking under `noImplicitAny`. We would've avoided these bugs if we'd checked with `noImplicitAny`. | Bug,Infrastructure | low | Critical |
560,651,524 | flutter | Configured font weights do not align with custom font | I have a typeface with multiple weights I want access to, and font files for each of these weights. I've configured `pubspec.yaml` with those weights, but they are not used when the identical `FontWeight` is used in a `TextStyle`.
pubspec:
```
fonts:
- family: Chivo
fonts:
- asset: assets/fonts/Chivo-Regular.ttf
- asset: assets/fonts/Chivo-Italic.ttf
style: italic
- asset: assets/fonts/Chivo-Light.ttf
weight: 200
- asset: assets/fonts/Chivo-LightItalic.ttf
weight: 200
style: italic
- asset: assets/fonts/Chivo-Bold.ttf
weight: 700
- asset: assets/fonts/Chivo-BoldItalic.ttf
style: italic
weight: 700
- asset: assets/fonts/Chivo-Black.ttf
weight: 900
- asset: assets/fonts/Chivo-BlackItalic.ttf
weight: 900
style: italic
```
If I then used `FontWeight.bold` in a `TextStyle`, the 900 weight is used instead. This does not line up with the documentation on using custom fonts here: https://flutter.dev/docs/cookbook/design/fonts#pubspecyaml-option-definitions
Should the documentation be updated, or is this an issue that should be fixed in the framework?
There is a workaround: two fonts. I configured one font with the "common" weights (400/700), and a second with the "extreme" weights (200/900), treated as 400/700:
```
fonts:
- family: Chivo
fonts:
- asset: assets/fonts/Chivo-Regular.ttf
- asset: assets/fonts/Chivo-Italic.ttf
style: italic
- asset: assets/fonts/Chivo-Bold.ttf
weight: 700
- asset: assets/fonts/Chivo-BoldItalic.ttf
style: italic
weight: 700
- family: 'Chivo Extreme'
fonts:
- asset: assets/fonts/Chivo-Light.ttf
- asset: assets/fonts/Chivo-LightItalic.ttf
style: italic
- asset: assets/fonts/Chivo-Black.ttf
weight: 700
- asset: assets/fonts/Chivo-BlackItalic.ttf
weight: 700
style: italic
``` | framework,d: api docs,a: assets,a: typography,P2,team-framework,triaged-framework | low | Major |
560,672,351 | terminal | closePane vs. closeTab differences? | # Environment
```none
Windows build number: 10.0.18362.592
Windows Terminal version (if applicable): 0.8.10261.0
Any other software?
WSL 2 (Debian 10)
bash 5.0.3
OpenSSH_7.9p1 Debian-10+deb10u1, OpenSSL 1.1.1c
```
Not entirely sure where this lies, but starting here since I can't recreate it elsewhere (putty, cmd.exe)
When closing a WSL session by either middle mouse clicking the tab (NOT the X) or by running the "closeTab" command via a hotkey, my ssh session will terminate and the application that's running receives a SIGHUP.
When closing a WSL session by clicking the X on a tab or by running the "closePane" command via a hotkey, my ssh session will terminate and the application that's running receives both SIGUSR1 (first) and SIGHUP.
This isn't an issue per se, but I'm trying to track down a bug in our own application and part of it seems to be related to how the application was terminated. Our end users don't use Windows Terminal, but in trying to understand how to create this scenario I've been working through different SSH clients.
**What does Windows Terminal do differently when closing a tab using "closePane" versus "closeTab"?** | Help Wanted,Product-Conpty,Area-Server,Issue-Bug,Priority-2 | low | Critical |
560,685,818 | pytorch | Torch not compiled with CUDA enabled | **Hi, I'm trying to run this code from the following repository (https://github.com/CorentinJ/Real-Time-Voice-Cloning), I already installed all requirements, however I'm getting this error when I run the demo_cli.py:**
AssertionError: Torch not compiled with CUDA enabled
**This is the full text I get when I run it:**
runfile('C:/Users/isabel/Desktop/VOICE CLONER/Real-Time-Voice-Cloning-master/demo_cli.py', wdir='C:/Users/isabel/Desktop/VOICE CLONER/Real-Time-Voice-Cloning-master')
Arguments:
enc_model_fpath: encoder\saved_models\pretrained.pt
syn_model_dir: synthesizer\saved_models\logs-pretrained
voc_model_fpath: vocoder\saved_models\pretrained\pretrained.pt
low_mem: False
no_sound: False
Running a test of your configuration...
Your PyTorch installation is not configured to use CUDA. If you have a GPU ready for deep learning, ensure that the drivers are properly installed, and that your CUDA version matches your PyTorch installation. CPU-only inference is currently not supported.
Traceback (most recent call last):
File "<ipython-input-6-e2d49148a56d>", line 1, in <module>
runfile('C:/Users/isabel/Desktop/VOICE CLONER/Real-Time-Voice-Cloning-master/demo_cli.py', wdir='C:/Users/isabel/Desktop/VOICE CLONER/Real-Time-Voice-Cloning-master')
File "C:\Users\isabel\Anaconda3\lib\site-packages\spyder_kernels\customize\spydercustomize.py", line 827, in runfile
File "C:\Users\isabel\Anaconda3\lib\site-packages\spyder_kernels\customize\spydercustomize.py", line 110, in execfile
# =============================================================================
File "C:/Users/isabel/Desktop/VOICE CLONER/Real-Time-Voice-Cloning-master/demo_cli.py", line 47, in <module>
device_id = torch.cuda.current_device()
File "C:\Users\isabel\Anaconda3\lib\site-packages\torch\cuda\__init__.py", line 377, in current_device
_lazy_init()
File "C:\Users\isabel\Anaconda3\lib\site-packages\torch\cuda\__init__.py", line 196, in _lazy_init
_check_driver()
File "C:\Users\isabel\Anaconda3\lib\site-packages\torch\cuda\__init__.py", line 94, in _check_driver
raise AssertionError("Torch not compiled with CUDA enabled")
AssertionError: Torch not compiled with CUDA enabled
**CUDA 10.0 is installed. I already tried uninstalling Pytorch and installing it again with the command "conda install pytorch torchvision cudatoolkit=10.0 -c pytorch", but it didn't work. Can anyone help with this issue?? Thank you so much!**
cc @ezyang @ngimel | awaiting response (this tag is deprecated),module: binaries,module: cuda,triaged | low | Critical |
560,697,466 | svelte | SSR Actions | **Is your feature request related to a problem? Please describe.**
It would be nice if actions could output attributes in SSR context.
**Describe the solution you'd like**
```html
<script>
function myAction(node, params) {
// the normal action stuff for client side
}
myAction.SSR = function (params) {
return {
'data-some-key': params.someKey,
};
}
export let someKey = 'someValue';
</script>
<div use:myAction={{ someKey }}>
<!-- would render <div data-some-key="someValue"></div> -->
```
**How important is this feature to you?**
Somewhat. It's just an idea i had, that could play well with something like [svelte-css-vars](https://github.com/kaisermann/svelte-css-vars). Imagine that svelte-css-vars could return `{ style: '--color: red' }` when rendered on the server. | feature request | low | Major |
560,698,082 | vue | Slot wrapped in `v-if` disappears after re-render | ### Version
2.6.11
### Reproduction link
[https://jsfiddle.net/y9kmcbg6/](https://jsfiddle.net/y9kmcbg6/)
### Steps to reproduce
1. JSFiddle will render both elements.
2. After 3s, it will toggle off
3. After 3s, it will toggle on
4. To see it happen again, rerun the jsfiddle
### What is expected?
At step 3, the bottom slot content should re-appear when ToggleComponent toggles back
### What is actually happening?
The bottom slot never re-appears
---
Seems to be related to the `v-if` on line 21. I'm doing this b/c it's useful to have an optional slot, but I don't want an extra div to appear when the slot isn't used (for layout reasons). I'm fairly new to vue, so I don't know if there's a different recommended way to do this.
Not documented in the example, but if ParentComponent updates, the slot content will re-appear
example code included here for perusal:
```vue
const ToggleComponent = {
data() {
return {
toggleEl: true,
};
},
created() {
setTimeout(() => this.toggleEl = false, 3000);
setTimeout(() => this.toggleEl = true, 6000);
},
render(h) {
return this.toggleEl ? h('div', this.$slots.default) : null
}
};
const SlotComponent = {
template: `
<div>
<slot></slot>
<div v-if="$slots.other">
<slot name="other"></slot>
</div>
</div>
`
}
const ParentComponent = {
template: `
<ToggleComponent>
<SlotComponent>
<template v-slot:default>
I should have an element below me
</template>
<template v-slot:other>
I am the element below you
</template>
</SlotComponent>
</ToggleComponent>
`,
components: {
ToggleComponent,
SlotComponent
},
}
var demo = new Vue({
el: '#demo',
template: `
<ParentComponent/>
`,
components: {
ParentComponent
},
});
```
<!-- generated by vue-issues. DO NOT REMOVE --> | bug,has workaround | low | Minor |
560,712,484 | pytorch | Scripting fails to preserve attribute aliasing | Script fails to preserve aliasing between attributes. As shown in example below:
def test_module_with_aliased_attr(self):
class M(torch.nn.Module):
def __init__(self):
super(M, self).__init__()
self.a = [1, 2, 3, 4, 5, 6]
self.b = self.a
def forward(self, x):
self.b[1] += 10
return self.a
m = M()
m_s = torch.jit.script(m)
m_s.eval()
inp = torch.tensor([5])
out = m_s.forward(inp)
expected = m.forward(inp)
print("expected: ", expected, "out ", out)
self.assertEqual(out, expected)
Prints: expected: [1, 12, 3, 4, 5, 6] out [1, 2, 3, 4, 5, 6]
cc @suo | oncall: jit,triaged | low | Minor |
560,713,764 | pytorch | Support batch linear transformation | ## 🚀 Feature
https://github.com/pytorch/pytorch/blob/1b746b95fb6f3df45bd2883045de9fbd82b72100/torch/nn/functional.py#L1362
expects the weight and bias tensors to have shapes `(out_features, in_features)` and `(out_features)`, respectively. We would like for this function to optionally support `(N, out_features, in_features)` and `(N, out_features)`, as well.
This request extends to other `torch.nn.functional` functions, but this is the painful one to us right now.
## Motivation
It's currently a bit of a pain to use `torch.nn.functional` with batch networks.
## Pitch
This request is effectively asking to replace
https://github.com/pytorch/pytorch/blob/1b746b95fb6f3df45bd2883045de9fbd82b72100/torch/nn/functional.py#L1379 with
```py
output = input.matmul(weight.transpose(-1, -2))
```
## Alternatives
Writing our own `torch.nn.functional.linear`
## Additional context
In writing this request, github helpfully pointed out https://github.com/pytorch/pytorch/issues/7500 as being similar. I suppose this is a request to take that beyond algebra operators and into nn operators.
cc @albanD @mruberry | module: nn,triaged,module: batching,function request | low | Minor |
560,721,433 | go | x/crypto/ssh: Semantics of Conn.Wait and Conn.Close unclear in docs | The semantics of Conn.Wait and Conn.Close are unclear as of the docs currently.
https://pkg.go.dev/golang.org/x/crypto/ssh?tab=doc
```
// Close closes the underlying network connection
Close() error
// Wait blocks until the connection has shut down, and returns the
// error causing the shutdown.
Wait() error
```
Is it necessary to call both, and in which order, or is just Close enough? | Documentation,NeedsInvestigation | low | Critical |
560,776,613 | rust | Poor error message when indexing result of dbg!() | Consider:
```rust
fn main() {
dbg!(1, 3)[1];
}
```
On nightly (2020-02-05) this yields the error:
```
error[E0608]: cannot index into a value of type `({integer}, {integer})`
--> src/main.rs:2:3
|
2 | dbg!(1, 3)[1];
| ^^^^^^^^^^^^^ help: to access tuple elements, use: `($ ($ crate :: dbg ! ($ val)), +,).1`
```
Maybe the `help: to access tuple elements, use` part should either avoid printing the left-hand side of the dot operator, or use the span of the source file like `help: to access tuple elements, use: 'dbg!(1, 3).1'` | C-enhancement,A-diagnostics,T-compiler | low | Critical |
560,786,315 | godot | Control focus inside viewport will get stuck after mouse wheel scroll (e.g. in a scroll container) | **Godot version:**
3.2-stable
**OS/device including version:**
Distributor ID: Ubuntu
Description: Pop!_OS 19.10
Release: 19.10
Codename: eoan
(Happens on windows as well)
**Issue description:**
When the mouse wheel is used to scroll a scroll container inside a viewport, the currently focused control will steal focus and block any further mouse movement events from reaching controls inside the viewport. (gui.mouse_focus gets set and never released)
**Steps to reproduce:**
Create a viewport scene (like the '2d in 3d demo' and add a scroll container with overflowing content such as a vboxcontainer full of buttons. Use the mouse wheel and notice that you can't click on any other buttons until after the mouse is moved out of the window or the viewport's mouse focus is reset some other way.
*NOTE* I've discovered that it only happens when the viewport has `handle_input_locally=false`
This may mean that this as 'not a bug' but I'll still leave this report for future generations ;)
**Minimal reproduction project:**
[gui_in_3d_mfocus_scrollwheel_repro.zip](https://github.com/godotengine/godot/files/4163347/gui_in_3d_mfocus_scrollwheel_repro.zip) | bug,confirmed,topic:gui | low | Critical |
560,809,097 | go | go.dev: How to get listed in "In-person training"? | There are 4 companies mentioned in the "In-person training" section at https://learn.go.dev/.
It's not clear why these companies where selected and now to get listed there. IMO there should be clear guidelines and instructions on how to get listed there.
Also it might be a good idea to split these companies by country.
Disclaimer: I want my company to be listed there as well. | NeedsInvestigation,website | low | Minor |
560,821,553 | flutter | Error message should be more verbose when Navigator.pop() is called accidentally | ## Steps to Reproduce
<!-- You must include full steps to reproduce so that we can reproduce the problem. -->
1. Run `flutter create bug`.
2. Add `Navigator.pop()` to the `onPressed` method of the `FloatingActionButton`.
3. Run the app and press the button.
4. Hot reload the app.
**Expected results:** Verbose error message in the console that mentions that you may have accidentally called `Navigator.pop()` in the wrong context.
**Actual results:**
`'package:flutter/src/widgets/navigator.dart': Failed assertion: line 2855 pos 12: '_history.isNotEmpty': is not true.`
<details>
<summary>Logs</summary>
<!--
Run your application with `flutter run --verbose` and attach all the
log output below between the lines with the backticks. If there is an
exception, please see if the error message includes enough information
to explain how to solve the issue.
-->
```
════════ Exception caught by widgets library ═══════════════════════════════════
The following assertion was thrown building Navigator-[GlobalObjectKey<NavigatorState> _WidgetsAppState#3f9ff](dirty, state: NavigatorState#69d35(tickers: tracking 0 tickers)):
'package:flutter/src/widgets/navigator.dart': Failed assertion: line 2855 pos 12: '_history.isNotEmpty': 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
The relevant error-causing widget was
MaterialApp lib/main.dart:XX
When the exception was thrown, this was the stack
#2 NavigatorState.build package:flutter/…/widgets/navigator.dart:2855
#3 StatefulElement.build package:flutter/…/widgets/framework.dart:4604
#4 ComponentElement.performRebuild package:flutter/…/widgets/framework.dart:4492
#5 StatefulElement.performRebuild package:flutter/…/widgets/framework.dart:4660
#6 Element.rebuild package:flutter/…/widgets/framework.dart:4216
...
════════════════════════════════════════════════════════════════════════════════
```
```
```
<!-- Finally, paste the output of running `flutter doctor -v` here. -->
```
[✓] Flutter (Channel master, v1.14.7-pre.138, on Linux, locale en_US.UTF-8)
• Flutter version 1.14.7-pre.138 at /.../flutter
• Framework revision 49be146953 (5 hours ago), 2020-02-05 18:02:20 -0800
• Engine revision 81dffd1241
• Dart version 2.8.0 (build 2.8.0-dev.7.0 c8ed304e97)
[✓] Android toolchain - develop for Android devices (Android SDK version 29.0.2)
• Android SDK at /.../Android/Sdk
• Android NDK location not configured (optional; useful for native profiling support)
• Platform android-29, build-tools 29.0.2
• Java binary at: /snap/android-studio/81/android-studio/jre/bin/java
• Java version OpenJDK Runtime Environment (build 1.8.0_202-release-1483-b49-5587405)
• All Android licenses accepted.
[✓] Linux toolchain - develop for Linux desktop
• clang++ 9.0.0
• GNU Make 4.2.1
[✓] Android Studio (version 3.5)
• Android Studio at /snap/android-studio/81/android-studio
• Flutter plugin version 41.1.2
• Dart plugin version 191.8593
• Java version OpenJDK Runtime Environment (build 1.8.0_202-release-1483-b49-5587405)
[!] IntelliJ IDEA Ultimate Edition (version 2019.2)
• IntelliJ at /.../idea-IU-192.6262.58
✗ Flutter plugin not installed; this adds Flutter specific functionality.
✗ Dart plugin not installed; this adds Dart specific functionality.
• For information about installing plugins, see
https://flutter.dev/intellij-setup/#installing-the-plugins
[✓] Connected device (2 available)
• ONEPLUS A6013 • 93a8e251 • android-arm64 • Android 10 (API 29)
• Linux • Linux • linux-x64 • Linux
! Doctor found issues in 1 category.
```
</details>
| framework,f: routes,a: error message,has reproducible steps,P2,found in release: 3.7,found in release: 3.8,team-framework,triaged-framework | low | Critical |
560,854,488 | vscode | Add a task instancePolicy to task runOptions. | When a task has reached the max number of instances it can have (default is 1, can be changed with `instanceLimit`) the `instancePolicy` determines what happens when another instance is started.
History: https://github.com/microsoft/vscode/issues/32264
PR for `insanceLimit` which is relevant to anyone interested in implementing `instancePolicy`: https://github.com/microsoft/vscode/pull/89872 | help wanted,feature-request,tasks | high | Critical |
560,892,095 | tensorflow | Loading Tensorflow models from disk is instant while loading from gcs is very slow | **System information**
- OS Platform and Distribution: Both MacOS and Ubuntu 18 (Tensorflow/serving)
- TensorFlow version (use command below): 2.0.0
- Python version: 3.6
**Describe the current behavior**
I want to use Tensorflow serving to load multiple Models. If I mount a directory containing the model, loading everything is done in an instant, while loading them from a `gs://` path takes around 10 seconds per model.
While researching the issue I discovered this is probably a Tensorflow issue and not a Tensorflow Serving issue as loading them in Tensorflow is a huge difference as well:
```
[ins] In [22]: %timeit tf.saved_model.load('test/1')
3.88 s ± 719 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
[ins] In [23]: %timeit tf.saved_model.load('gs://path/to/test/1')
30.6 s ± 2.66 s per loop (mean ± std. dev. of 7 runs, 1 loop each)
```
Then it could be that downloading the model (which is very small) is slow, but I tested this as well:
```
def test_load():
bucket_name = 'path'
folder='test'
delimiter='/'
file = 'to/test/1'
bucket=storage.Client().get_bucket(bucket_name)
blobs=bucket.list_blobs(prefix=file, delimiter=delimiter)
for blob in blobs:
print(blob.name)
destination_uri = '{}/{}'.format(folder, blob.name)
blob.download_to_filename(destination_uri)
[ins] In [31]: %timeit test_load()
541 ms ± 54.9 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
```
Any idea what is happening here?
**Describe the expected behavior**
Around the same load time for external vs local models. The first load from gs can be slow if the auth needs to happen, but even authenticating is still way faster than the models loads.
| stat:awaiting tensorflower,comp:apis,type:performance,TF 2.4 | medium | Major |
560,926,576 | create-react-app | Provide stacktrace for react-scripts/build.js `build` function | ### Is your proposal related to a problem?
<!--
Provide a clear and concise description of what the problem is.
For example, "I'm always frustrated when..."
-->
I've encountered a problem that I think could've been diagnosed more easily if `build` script were more verbose and provided a stack trace aside of error message.
My scenario was pretty simple - I've used CRA, eject, then played a bit with `entry` record:
https://github.com/facebook/create-react-app/blob/687c4ebf211ad30238f2d59e063b8171e015bfc7/packages/react-scripts/config/webpack.config.js#L151
by replacing an array configuration with an object configuration ([which is valid webpack option](https://webpack.js.org/configuration/entry-context/#entry))
After that I couldn't build my app (which is fine) and I coundn't easily find a problem, see attached image:
As you can see, the error message didn't tell me where the problem occured or what the context was, so if I've played a lot around my configuration then I have to either revert my changes line by line to find which exactly broke the build or debug build script itself.
### Describe the solution you'd like
<!--
Provide a clear and concise description of what you want to happen.
-->
Error message was formed here:
https://github.com/facebook/create-react-app/blob/687c4ebf211ad30238f2d59e063b8171e015bfc7/packages/react-scripts/scripts/build.js#L196
If there would've been an additional stacktrace added to that array:
```js
if (err) {
messages.errors.push(err.stack);
}
return reject(new Error(messages.errors.join('\n\n'), ));
```
then I would have a more precise description of a problem that shows me that 'filter of undefined' situation occured exactly at 583 line of my webpack config which I should further work with:
### Describe alternatives you've considered
<!--
Let us know about other solutions you've tried or researched.
-->
There is an obvious concern that stacktraces might not have a place in defined policy of user-friendly error messages, but they probably could be enabled by additional `--verbose` flag.
| issue: proposal,needs triage | low | Critical |
560,930,463 | material-ui | [material-ui][Button] Text not aligned vertically with icon | When button has an icon, text is not aligned vertically.
You can see it at [example page](https://material-ui.com/components/buttons/#buttons-with-icons-and-label).
I made an screenshot and check at graphics designer - button "delete" has 12px from top to text and 14px from text to bottom. While icon has exactly 10px from top and from bottom. | component: button,package: material-ui,ready to take | low | Major |
560,936,566 | material-ui | [Select] Pass rendered content of MenuItem to renderValue function | <!-- Provide a general summary of the feature in the Title above -->
<!--
Thank you very much for contributing to Material-UI by creating an issue! ❤️
To avoid duplicate issues we ask you to check off the following list.
-->
<!-- Checked checkbox should look like this: [x] -->
- [x] I have searched the [issues](https://github.com/mui-org/material-ui/issues) of this repository and believe that this is not a duplicate.
## Summary 💡
In the current situation it's possible to customize the rendered content of a Select component using the renderValue prop. This works really well I just have one gripe. I would like the rendered content of the selected MenuItem to be passed as an argument as well.
As a solution that wont break existing implementations I propose it would be passed as a second function argument, e.g.:
`function(value: any, content: ReactNode) => ReactNode`
When using the multiple prop the content argument would be an array of ReactNodes.
<!-- Describe how it should work. -->
<!--
Provide a link to the Material design specification, other implementations,
or screenshots of the expected behavior.
-->
## Motivation 🔦
Often I'm using customized content inside the MenuItems in my Select, like icons and avatars. When not using the renderValue prop this content is automatically rendered inside the Select when a selecting a value. However when using the renderValue prop you only get passed the value as function parameter, you would have to generate the content again inside the renderValue function.
<!--
What are you trying to accomplish? How has the lack of this feature affected you?
Providing context helps us come up with a solution that is most useful in the real world.
-->
| new feature,component: select,ready to take | low | Major |
560,948,835 | pytorch | [debatable] Better infer dtype in torch.as_tensor | Supporting generator inputs in `torch.tensor` / `torch.as_tensor` would save some keystrokes to the user. Converting the input to list internally would do the trick. On the contrary note: numpy does not support this.
```
>>> torch.as_tensor(list(False for i in range(3)))
tensor([False, False, False])
>>> torch.as_tensor([False for i in range(3)])
tensor([False, False, False])
>>> torch.as_tensor(False for i in range(3))
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
RuntimeError: Could not infer dtype of generator
```
The relevant line is in https://github.com/pytorch/pytorch/blob/62b06b9fae9b151b20d25c8e54ae9ae1081f23ad/torch/csrc/utils/tensor_new.cpp#L200 | triaged,module: tensor creation | low | Critical |
560,953,876 | pytorch | bytearray(tensor) behaves very differently from bytearray(tensor.numpy()) | ```python
import torch
a = torch.ShortTensor([10000])
print(bytearray(a.numpy()))
# bytearray(b"\x10\'")
print(bytearray(a))
# bytearray(b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00 ............')
```
I thought this is the Python prescribed interface of getting byte representation of some object, but this fails with PyTorch arrays. My goal is to pass the obtained byte array or byte string to webrtcvad method working with audio byte frames.
Related: https://discuss.pytorch.org/t/getting-tensor-raw-data/18961/2
cc @mruberry @rgommers @heitorschueroff | module: printing,triaged,module: numpy | low | Minor |
560,954,250 | scrcpy | Fire Stick 4K Low FPS | I am trying to copy my Amazon Fire Stick 4K display on my Laptop using the scrcpy. Everything seems to be working fine, but when I run the scrcpy.exe, the screen I see is super low FPS. I mean it is around 1FPS.
And When I try scrcpy.exe -m640
it is almost 60fps, but it is unusable for me at that resolution. Any resolution above that has very low FPS.
It only happens with this new Fire Stick 4K. My old Fire Stick and my phone seems to be working fine. | performance | low | Major |
560,971,585 | flutter | Flutter Web: JavaScript history.pushState raises an exception in Release Build. | ## Steps to Reproduce
<!-- You must include full steps to reproduce so that we can reproduce the problem. -->
1. Clone repo: https://github.com/gladimdim/push_history_bug.
2. Switch to Flutter beta channel & enable web
```
flutter channel beta
flutter upgrade
flutter doctor
flutter config --enable-web
```
3. Run the *debug* mode:
```
flutter run -d chrome
```
It will show an app with the button. Clicking on the button will add 'google.com' at the end of the URL in address bar. js.history.pushState works OK in debug mode.
4. Run the *release* build :
```
flutter run --release -d chrome
```
5. Press the red button multiple times.
**Expected results:** the url is updated via js.history.pushState bindings from dart:js. Each button click adds 'google.com' at the end of the URL in the address bar.
**Actual results:**: the url is not updated. Exception is thrown in the Chrome console: Another exception was thrown: Instance of 'minified:ar<void>'
<details>
<summary>Logs</summary>
<!--
Run your application with `flutter run --verbose` and attach all the
log output below between the lines with the backticks. If there is an
exception, please see if the error message includes enough information
to explain how to solve the issue.
-->
```
flutter run --verbose -d chrome --release
[ +28 ms] executing: [/Users/i501633/progs/flutter/] git -c log.showSignature=false log -n 1 --pretty=format:%H
[ +48 ms] Exit code 0 from: git -c log.showSignature=false log -n 1 --pretty=format:%H
[ ] fabeb2a16f1d008ab8230f450c49141d35669798
[ ] executing: [/Users/i501633/progs/flutter/] git describe --match v*.*.* --first-parent --long --tags
[ +39 ms] Exit code 0 from: git describe --match v*.*.* --first-parent --long --tags
[ ] v1.14.6-0-gfabeb2a16
[ +7 ms] executing: [/Users/i501633/progs/flutter/] git rev-parse --abbrev-ref --symbolic @{u}
[ +10 ms] Exit code 0 from: git rev-parse --abbrev-ref --symbolic @{u}
[ ] origin/beta
[ ] executing: [/Users/i501633/progs/flutter/] git ls-remote --get-url origin
[ +9 ms] Exit code 0 from: git ls-remote --get-url origin
[ ] https://github.com/flutter/flutter.git
[ +45 ms] executing: [/Users/i501633/progs/flutter/] git rev-parse --abbrev-ref HEAD
[ +11 ms] Exit code 0 from: git rev-parse --abbrev-ref HEAD
[ ] beta
[ +6 ms] executing: sw_vers -productName
[ +16 ms] Exit code 0 from: sw_vers -productName
[ ] Mac OS X
[ ] executing: sw_vers -productVersion
[ +16 ms] Exit code 0 from: sw_vers -productVersion
[ ] 10.15.3
[ ] executing: sw_vers -buildVersion
[ +16 ms] Exit code 0 from: sw_vers -buildVersion
[ ] 19D76
[ +33 ms] Artifact Instance of 'AndroidMavenArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'AndroidGenSnapshotArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'AndroidInternalBuildArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'IOSEngineArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'FlutterWebSdk' is not required, skipping update.
[ +3 ms] Artifact Instance of 'WindowsEngineArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'MacOSEngineArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'LinuxEngineArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'LinuxFuchsiaSDKArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'MacOSFuchsiaSDKArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'FlutterRunnerSDKArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'FlutterRunnerDebugSymbols' is not required, skipping update.
[ +36 ms] executing: /Users/i501633/Library/Android/sdk/platform-tools/adb devices -l
[ +7 ms] Exit code 0 from: /Users/i501633/Library/Android/sdk/platform-tools/adb devices -l
[ ] List of devices attached
[ +15 ms] executing: /Users/i501633/progs/flutter/bin/cache/artifacts/libimobiledevice/idevice_id -h
[ +63 ms] executing: /usr/bin/xcode-select --print-path
[ +8 ms] Exit code 0 from: /usr/bin/xcode-select --print-path
[ ] /Applications/Xcode.app/Contents/Developer
[ ] executing: /usr/bin/xcodebuild -version
[ +595 ms] Exit code 0 from: /usr/bin/xcodebuild -version
[ +3 ms] Xcode 11.3.1
Build version 11C504
[ +4 ms] /usr/bin/xcrun simctl list --json devices
[ +113 ms] Artifact Instance of 'AndroidMavenArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'AndroidGenSnapshotArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'AndroidInternalBuildArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'IOSEngineArtifacts' is not required, skipping update.
[ +2 ms] Artifact Instance of 'WindowsEngineArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'MacOSEngineArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'LinuxEngineArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'LinuxFuchsiaSDKArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'MacOSFuchsiaSDKArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'FlutterRunnerSDKArtifacts' is not required, skipping update.
[ ] Artifact Instance of 'FlutterRunnerDebugSymbols' is not required, skipping update.
[ +216 ms] Generating /Users/i501633/projects/history_push_error/android/app/src/main/java/io/flutter/plugins/GeneratedPluginRegistrant.java
[ +117 ms] Launching lib/main.dart on Chrome in release mode...
[ ] Building application for the web...
[ +185 ms] Generating /Users/i501633/projects/history_push_error/android/app/src/main/java/io/flutter/plugins/GeneratedPluginRegistrant.java
[ +4 ms] Compiling lib/main.dart for the Web...
[ +13 ms] Initializing file store
[ +10 ms] Done initializing file store
[ +72 ms] invalidated build due to missing files: /Users/i501633/projects/history_push_error/packages/flutter_tools/lib/src/build_system/targets/web.dart
[ +16 ms] web_entrypoint: Starting due to {InvalidatedReason.inputMissing}
[ +2 ms] web_entrypoint: Complete
[+1827 ms] dart2js: Starting due to {InvalidatedReason.inputChanged}
[+17424 ms] dart2js: Complete
[ +66 ms] web_release_bundle: Starting due to {InvalidatedReason.inputChanged}
[ +63 ms] web_release_bundle: Complete
[ +72 ms] web_service_worker: Starting due to {InvalidatedReason.inputChanged}
[ +31 ms] web_service_worker: Complete
[ +2 ms] Persisting file store
[ +5 ms] Done persisting file store
[ +15 ms] Compiling lib/main.dart for the Web... (completed in 19.6s)
[ +13 ms] Building application for the web... (completed in 19.8s)
[ +874 ms] Warning: Flutter's support for web development is not stable yet and hasn't
[ ] been thoroughly tested in production environments.
[ ] For more information see https://flutter.dev/web
[ ] 🔥 To hot restart changes while running, press "r". To hot restart (and refresh the browser), press "R".
[ +2 ms] For a more detailed help message, press "h". To quit, press "q".
```
<!--
Run `flutter analyze` and attach any output of that command below.
If there are any analysis errors, try resolving them before filing this issue.
-->
```
flutter analyze
Analyzing history_push_error...
info • Avoid using web-only libraries outside Flutter web plugin packages • lib/main.dart:3:1 • avoid_web_libraries_in_flutter
1 issue found. (ran in 8.2s)
```
<!-- Finally, paste the output of running `flutter doctor -v` here. -->
```
flutter doctor -v
[✓] Flutter (Channel beta, v1.14.6, on Mac OS X 10.15.3 19D76, locale en)
• Flutter version 1.14.6 at /Users/i501633/progs/flutter
• Framework revision fabeb2a16f (9 days ago), 2020-01-28 07:56:51 -0800
• Engine revision c4229bfbba
• Dart version 2.8.0 (build 2.8.0-dev.5.0 fc3af737c7)
[✓] Android toolchain - develop for Android devices (Android SDK version 29.0.2)
• Android SDK at /Users/i501633/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: /Applications/Android Studio.app/Contents/jre/jdk/Contents/Home/bin/java
• Java version OpenJDK Runtime Environment (build 1.8.0_202-release-1483-b49-5587405)
• 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 11C504
• CocoaPods version 1.8.4
[✓] Chrome - develop for the web
• Chrome at /Applications/Google Chrome.app/Contents/MacOS/Google Chrome
[✓] Android Studio (version 3.5)
• Android Studio at /Applications/Android Studio.app/Contents
• Flutter plugin version 43.0.1
• Dart plugin version 191.8593
• Java version OpenJDK Runtime Environment (build 1.8.0_202-release-1483-b49-5587405)
[✓] VS Code (version 1.41.1)
• VS Code at /Applications/Visual Studio Code.app/Contents
• Flutter extension version 3.8.0
[✓] Connected device (2 available)
• Chrome • chrome • web-javascript • Google Chrome 80.0.3987.87
• Web Server • web-server • web-javascript • Flutter Tools
• No issues found!
```
</details>
| c: crash,dependency: dart,platform-web,a: release,has reproducible steps,P2,found in release: 3.3,found in release: 3.5,team-web,triaged-web | low | Critical |
560,981,315 | node | Improve performance of accessing globals in scripts running in `vm` modules | **Is your feature request related to a problem? Please describe.**
Accessing globals from within a `vm.Script` is guarded by interceptors which makes accessing them slow, (from my understanding, @fhinkel has an excellent comment explaining some of this here: https://github.com/facebook/jest/issues/5163#issuecomment-355509597). The suggested workaround is to evaluate and cache the global you want access to (`Math` in my example case below) and inject that into the `vm.Script`. This is an acceptable workaround for `vm.Script` and `vm.compileFunction` but it cannot be done when using ESM `vm.SourceTextModule` as there is no wrapping function call.
I've put together this script you can run to see the numbers:
```js
// test.js
const {
compileFunction,
createContext,
runInContext,
SourceTextModule
} = require("vm");
const context = createContext({ console });
const sourceCode = `
const runTimes = 10000000;
let sum = 0;
console.time();
for (let i = 0; i < runTimes; i++) {
sum += Math.abs(0);
}
console.timeEnd();
`;
(async () => {
const compiledFunction = compileFunction(sourceCode, [], {
parsingContext: context
});
const vmModule = new SourceTextModule(sourceCode, { context });
await vmModule.link(() => {
throw new Error("should never be called");
});
// These are both super slow (1.6 seconds on my machine)
compiledFunction();
await vmModule.evaluate();
// however, this is quick (less than 10 milliseconds on my machine)
const compiledWithMathFunction = compileFunction(sourceCode, ["Math"], {
parsingContext: context
});
compiledWithMathFunction(runInContext("Math", context));
})().catch(error => {
console.error(error);
process.exitCode = 1;
});
```
Running this gives the following results on my machine:
```sh-session
$ node --experimental-vm-modules test.js
(node:19958) ExperimentalWarning: VM Modules is an experimental feature. This feature could change at any time
default: 1.698s
default: 1.620s
default: 7.306ms
```
So ~1600ms if not using the workaround, which reduces it to ~7ms.
**Describe the solution you'd like**
I'd like for accessing globals to be as fast, or as close to as possible, as fast as accessing globals outside of `vm`. Would it be possible for Node's `vm` implementation to cache the global property lookups, so that the price is only paid once instead of on every single access?
**Describe alternatives you've considered**
As mentioned, caching and injecting the global manually works when there is a function wrapper, but this doesn't work with ESM. I've tried assigning the `Math` global to the `context`, and while that halves the time spent (~800ms), it's still 2 orders of magnitude slower than injecting a reference. | vm,v8 engine,performance | medium | Critical |
561,072,029 | node | net: setKeepAlive options ignored before connect | <!--
Thank you for reporting a possible bug in Node.js.
Please fill in as much of the template below as you can.
Version: output of `node -v`
Platform: output of `uname -a` (UNIX), or version and 32 or 64-bit (Windows)
Subsystem: if known, please specify the affected core module name
If possible, please provide code that demonstrates the problem, keeping it as
simple and free of external dependencies as you can.
-->
* **Version**: v13.7.0
* **Platform**: linux 5.5.2
* **Subsystem**: net
<!-- Please provide more details below this comment. -->
If `setKeepAlive()` is called on a tcp socket before the `connect` event, then the `initialDelay` parameter is ignored.
This bug was alread reported [here](https://github.com/nodejs/node-v0.x-archive/issues/8572) for version 0.12, but it still seems to be there in 13.7.
If, for instance, I call `setKeepAlive(true, 10000)` before the `connect` event, the default initialDelay of 7200s is used.
`netstat -ntp --timers` gives:
```
tcp 0 0 10.10.15.220:56866 10.10.15.148:6666 ESTABLISHED 132820/node keepalive (7132.09/0/0)
```
| net | low | Critical |
561,091,706 | pytorch | torch.nn.functional import grid_sample | If you have a question or would like help and support, please ask at our
[forums](https://discuss.pytorch.org/).
If you are submitting a feature request, please preface the title with [feature request].
If you are submitting a bug report, please fill in the following details.
## Issue description
torch.nn.functional.grid_sample() behavior is differ from pytorch 1.2.0
## Code example
pytorch 1.2.0:
grid_sample(input=inp, grid=grid, mode='bilinear', padding_mode='border')
pytorch 1.4.0:
the correct way to reproduce behavior of 1.2.0 is:
grid_sample(input=inp, grid=grid, mode='bilinear', padding_mode='border', align_corners=True)
but the default value of align_corners is None, so that produces some errors when update to pytorch 1.4 from 1.2
## System Info
PyTorch 1.2.0:
Collecting environment information...
PyTorch version: 1.2.0
Is debug build: No
CUDA used to build PyTorch: 10.0
OS: Microsoft Windows 10 Enterprise
GCC version: (x86_64-posix-seh-rev0, Built by MinGW-W64 project) 8.1.0
CMake version: version 3.15.5
Python version: 3.6
Is CUDA available: Yes
CUDA runtime version: 10.1.105
GPU models and configuration: GPU 0: GeForce GTX 1080 Ti
Nvidia driver version: 418.96
cuDNN version: Could not collect
Versions of relevant libraries:
[pip] numpy==1.18.1
[pip] pytorch-memlab==0.0.2
[pip] torch==1.2.0
[pip] torchvision==0.4.0
[conda] blas 1.0 mkl
[conda] libmklml 2019.0.3 0
[conda] mkl 2019.5 281 anaconda
[conda] mkl-service 2.3.0 py36hb782905_0 anaconda
[conda] mkl_fft 1.0.12 py36h14836fe_0
[conda] mkl_random 1.0.2 py36h343c172_0
[conda] pytorch 1.2.0 py3.6_cuda100_cudnn7_1 pytorch
[conda] pytorch-memlab 0.0.2 pypi_0 pypi
[conda] torchvision 0.4.0 py36_cu100 pytorch
PyTorch 1.4.0:
Collecting environment information...
PyTorch version: 1.4.0
Is debug build: No
CUDA used to build PyTorch: 10.1
OS: Microsoft Windows 10 Enterprise
GCC version: (x86_64-posix-seh-rev0, Built by MinGW-W64 project) 8.1.0
CMake version: version 3.15.5
Python version: 3.6
Is CUDA available: Yes
CUDA runtime version: 10.1.105
GPU models and configuration: GPU 0: GeForce GTX 1080 Ti
Nvidia driver version: 418.96
cuDNN version: Could not collect
Versions of relevant libraries:
[pip] numpy==1.18.1
[pip] pytorch-memlab==0.0.2
[pip] torch==1.4.0
[pip] torchvision==0.5.0
[conda] blas 1.0 mkl
[conda] libmklml 2019.0.3 0
[conda] mkl 2019.5 281 anaconda
[conda] mkl-service 2.3.0 py36hb782905_0 anaconda
[conda] mkl_fft 1.0.12 py36h14836fe_0
[conda] mkl_random 1.0.2 py36h343c172_0
[conda] pytorch 1.4.0 py3.6_cuda101_cudnn7_0 pytorch
[conda] pytorch-memlab 0.0.2 pypi_0 pypi
[conda] torchvision 0.5.0 py36_cu101 pytorch | needs reproduction,triaged | low | Critical |
561,117,256 | go | x/tools/gopls: add support for OnTypeFormatting | ### What version of Go are you using (`go version`)?
<pre>
$ go version
go version devel +b7689f5aa3 Fri Jan 31 06:02:00 2020 +0000 linux/amd64
$ go list -m golang.org/x/tools
golang.org/x/tools v0.0.0-20200205190317-6e8b36d2c76b
$ go list -m golang.org/x/tools/gopls
golang.org/x/tools/gopls v0.1.8-0.20200205190317-6e8b36d2c76b
</pre>
### Does this issue reproduce with the latest release?
Per https://microsoft.github.io/language-server-protocol/specifications/specification-current/#textDocument_onTypeFormatting, add support for `OnTypeFormatting`.
@stamblerre I'll leave you to milestone.
cc @findleyr
FYI @leitzler
| help wanted,FeatureRequest,gopls,Tools | low | Minor |
561,121,403 | pytorch | Warning when link libtorch and opencv4.2.0 together | ## To Reproduce
Steps to reproduce the behavior:
1. Download official opencv4.2.0
2. Download libtorch1.4.0 of windows 64 bits, cpu version
3. Create CMakeLists.txt
```
cmake_minimum_required(VERSION 3.0 FATAL_ERROR)
project(person_reid)
set(LIBTORCH_PATH ${CMAKE_CURRENT_LIST_DIR}/../../3rdLibs/libtorch/libtorch_cpu_1_4_0/)
set(OPENCV_CV_PATH ${CMAKE_CURRENT_SOURCE_DIR}/../../3rdLibs/opencv/opencv_4_2_0/opencv/build/)
set(CMAKE_PREFIX_PATH "${OPENCV_CV_PATH};${LIBTORCH_PATH};${CMAKE_PREFIX_PATH};")
find_package(OpenCV REQUIRED)
find_package(Torch REQUIRED)
file(GLOB UTILS_SRC
"*.hpp"
"*.cpp"
)
add_executable(person_reid ${UTILS_SRC})
target_link_libraries(person_reid "${TORCH_LIBRARIES};${OpenCV_LIBS}")
set_property(TARGET person_reid PROPERTY CXX_STANDARD 11)
```
4. Codes
```
#include <opencv2/core.hpp>
#include <opencv2/highgui.hpp>
#include <opencv2/imgproc.hpp>
#include <iostream>
#include <memory>
//#undef slots
#include <torch/script.h>
//#def slots Q_SLOTS
int main()try
{
cv::Mat img = cv::imread("some.jpg");
// Deserialize the ScriptModule from a file using torch::jit::load().
torch::jit::script::Module module = torch::jit::load("traced_osnet_ain_x1_0.pt");
}catch(std::exception const &ex){
std::cerr << ex.what() << "\n";
return -1;
}
```
5. Build with release version
## Expected behavior
Compile without any warnings
## 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).
- PyTorch Version: 1.4.0
- OS (e.g., Linux): windows 10 64 bits
- How you installed PyTorch (`conda`, `pip`, source): conda
- Python version: 3.6.7
- CUDA/cuDNN version: none, cpu only
- GPU models and configuration: use libtorch with cpu only
- Compiler vc2017 64bits
## Warnings
```
cl : Command line warning D9025 : overriding '/EHs' with '/EHa'
C:\Users\yyyy\programs\Qt\3rdLibs\libtorch\libtorch_cpu_1_4_0\include\torch/csrc/jit/tracer.h(289): warning C4273: 'torch::jit::tracer::addInputs': inconsistent dll linkage
C:\Users\yyyy\programs\Qt\3rdLibs\libtorch\libtorch_cpu_1_4_0\include\torch/csrc/jit/tracer.h(277): note: see previous definition of 'addInputs'
C:\Users\yyyy\programs\Qt\3rdLibs\libtorch\libtorch_cpu_1_4_0\include\torch/csrc/jit/tracer.h(296): warning C4273: 'torch::jit::tracer::addInputs': inconsistent dll linkage
C:\Users\yyyy\programs\Qt\3rdLibs\libtorch\libtorch_cpu_1_4_0\include\torch/csrc/jit/tracer.h(283): note: see previous definition of 'addInputs'
``` | module: build,triaged,better-engineering | low | Critical |
561,150,502 | opencv | matchTemplate returns different scores for same input with IPP | <!--
If you have a question rather than reporting a bug please go to http://answers.opencv.org where you get much faster responses.
If you need further assistance please read [How To Contribute](https://github.com/opencv/opencv/wiki/How_to_contribute).
Please:
* Read the documentation to test with the latest developer build.
* Check if other person has already created the same issue to avoid duplicates. You can comment on it if there already is an issue.
* Try to be as detailed as possible in your report.
* Report only one problem per created issue.
This is a template helping you to create an issue which can be processed as quickly as possible. This is the bug reporting section for the OpenCV library.
-->
##### System information (version)
- OpenCV => 4.2.0-dev (also observed on 4.1.2 release)
- Operating System / Platform => Windows 64 Bit
- Compiler => Visual Studio 2015
##### Detailed description
If IPP is active (`WITH_IPP=ON`), `matchTemplate` scores are non-deterministic. I.e. the function returns different scores for the same input. The problem can be observed for all [TemplateMatchModes](https://docs.opencv.org/3.4.0/df/dfb/group__imgproc__object.html#ga3a7850640f1fe1f58fe91a2d7583695d).
Note that `ipp_matchTemplate` is will *not* be used in the sample below, as the images are the same size. So the issue is probably not caused by a bug in `templmatch.cpp`, but comes from somewhere deeper down.
##### Steps to reproduce
<!-- to add code example fence it with triple backticks and optional file extension
```.cpp
// C++ code example
```
or attach as .txt or .zip file
-->
```python
import cv2
import numpy as np
image = np.full([64, 64], 0, dtype=np.uint8)
cv2.circle(image, (32, 32), 30, 255, cv2.FILLED)
template = np.full([64, 64], 0, dtype=np.uint8)
cv2.circle(template, (45, 45), 30, 128, cv2.FILLED)
method_names = ['SQDIFF', 'SQDIFF_NORMED', 'CCORR', 'CCORR_NORMED', 'CCOEFF', 'CCOEFF_NORMED']
# just testing non-normalized methods, as the normed versions rely on their results anyway
methods = [cv2.TM_SQDIFF, cv2.TM_CCORR, cv2.TM_CCOEFF]
for method in methods:
print('======= {} ======='.format(method_names[method]))
scores = []
run_count = 100000
for i in range(run_count):
match_res = cv2.matchTemplate(image, template, method)
_, score, _, pos = cv2.minMaxLoc(match_res)
scores.append(score)
scores = np.array(scores)
mx = np.max(scores)
mn = np.min(scores)
max_diff = mx - mn
rel_diff = max_diff / mx if mx != 0 else 0
print('max: {} | min: {} | diff: {} ({:.3f}% of max)'.format(mx, mn, max_diff, rel_diff * 100))
```
Sample output:
```
======= SQDIFF =======
max: 104318944.0 | min: 104318488.0 | diff: 456.0 (0.000% of max)
======= CCORR =======
max: 56597880.0 | min: 56597652.0 | diff: 228.0 (0.000% of max)
======= CCOEFF =======
max: 9839805.0 | min: 9839601.0 | diff: 204.0 (0.002% of max)
``` | bug,category: imgproc,category: 3rdparty | low | Critical |
561,163,315 | pytorch | Recover from CUDA runtime error | ## ❓ Questions and Help
I wanted to recover from
RuntimeError: inverse_cuda: For batch 0: U(2344,2344) is zero, singular U.
so I tried to catch the error and do
torch.cuda.empty_cache()
which leads to the another error
RuntimeError: CUDA error: misaligned address
without empty cache I got the illegal memory access error below.
Any suggestions?
I use CUDA 10 pytorch 1.1.0~ TitanXP *4
Also I use dataparallel (not distributed one) during training.
Thanks a lot.
CUDA runtime error: an illegal memory access was encountered (77) in magma_sgetrf_batched at /opt/conda/conda-bld/magma-cuda100_1549065924616/work/src/sgetrf_batched.cpp:213
CUDA runtime error: an illegal memory access was encountered (77) in magma_sgetrf_batched at /opt/conda/conda-bld/magma-cuda100_1549065924616/work/src/sgetrf_batched.cpp:214
CUDA runtime error: an illegal memory access was encountered (77) in magma_sgetrf_batched at /opt/conda/conda-bld/magma-cuda100_1549065924616/work/src/sgetrf_batched.cpp:215
CUDA runtime error: an illegal memory access was encountered (77) in magma_sgetri_outofplace_batched at /opt/conda/conda-bld/magma-cuda100_1549065924616/work/src/sgetri_outofplace_batched.cpp:137
CUDA runtime error: an illegal memory access was encountered (77) in magma_queue_destroy_internal at /opt/conda/conda-bld/magma-cuda100_1549065924616/work/interface_cuda/interface.cpp:944
CUDA runtime error: an illegal memory access was encountered (77) in magma_queue_destroy_internal at /opt/conda/conda-bld/magma-cuda100_1549065924616/work/interface_cuda/interface.cpp:945
CUDA runtime error: an illegal memory access was encountered (77) in magma_queue_destroy_internal at /opt/conda/conda-bld/magma-cuda100_1549065924616/work/interface_cuda/interface.cpp:946
### Please note that this issue tracker is not a help form and this issue will be closed.
We have a set of [listed resources available on the website](https://pytorch.org/resources). Our primary means of support is our discussion forum:
- [Discussion Forum](https://discuss.pytorch.org/)
cc @ngimel | module: cuda,triaged,module: third_party | low | Critical |
561,209,478 | go | misc/android: go_android_exec fails to copy other package sources in the main module when executing tests | In order to ensure that a test binary can find any associated test data, `go test` runs the resulting test binary in its own source directory.
That implies that, if a `go` binary is present, the test can `go build` both files in its own directory and any associated packages in the same module.
Unfortunately, `adbCopyTree` function in `go_android_exec.go` does not ensure that the source code for the complete module is present. Instead, it copies only the `go.mod` and `go.sum` files and `testdata` directories going up the tree.
That implies that `go_android_exec` cannot successfully execute an external test that runs an associated `main` package as a subprocess if that `main` package imports any other package outside of the standard library, such as in [CL 218277](https://golang.org/cl/218277).
CC @hajimehoshi @hyangah @steeve @eliasnaur | help wanted,NeedsFix,mobile | low | Minor |
561,230,361 | pytorch | [docs] Strange order of items in docs contents in left pane | ```
torch.optim
Quantization
Distributed RPC Framework
torch.random
```
why this order? not even alphabetical. I'd suggest that big special topics such as Quantization are presented on top in a separate section, and the rest to be in alphabetical order.
 | module: docs,triaged | low | Minor |
561,235,303 | flutter | Dynamically load ICU | I will fill this out more shortly, but want to brain dump some things before I forget:
- Move ICU to a storage bucket
- See if we can split ICU into a "core" package and language specific packages (for Burmese, Japanese, Thai, Lao, Khmer, and Finish).
- Always grab the core package on first load
- Grab language specific packages only if needed.
This would reduce the core distribution size of a Flutter app by about 500kb, and reduce the on disk footprint for users who never see or enter burmese text for example.
Need to determine:
- Can ICU support this, or would we need to customize the ICU build?
- Does this have a highly adverse impact on startup times/early frame performance?
- Is supporting the infra for it worthwhile (a storage bucket that would be hit by users of Flutter apps rather than developers using the framework).
/cc @goderbauer @xster
/cc @blasten | engine,c: performance,a: internationalization,a: size,perf: app size,P3,team-engine,triaged-engine | low | Major |
561,239,305 | rust | Command causes subsequent ANSI codes to render incorrectly | I'm experiencing some odd behavior with ANSI codes and `std::process::Command`. The ANSI codes render differently after having executed a command.
1. In Git Bash using `\u{1b}[2m` (gray/dim) does not render at all, which is understanble as it might not be supported. However, after executing some command, then it renders correctly.
2. On the other hand, in CMD and PowerShell, then after running some command, then the ANSI codes appear *literally* in the output. (Which is the actual issue.)
The command being executed seem to be important, as changing it either results in the issue being triggered or no change at all.
Both `spawn() + wait()` and `status()` behave the same. Whereas `output()` does not trigger the issue. So I'm assuming the command is somehow able to affect the inherited stdin/stdout/stderr, causing the issue to persist after the command has finished.
## Git Bash
It's understandable that some ANSI code might not be supported by my terminal. But it's quite odd, that executing some command using `std::process::Command`, results in it rendering correctly after.
## CMD
However, the real issue is in CMD (and PowerShell), where every ANSI code renders *literally* in the output, after having executed some command.
## Minimal, Reproducible Example
I also tried using the [clicolors-control](https://crates.io/crates/clicolors-control) crate to enable/disable ANSI colors. However the issue still persists, and I originally submitted the issue there earlier mitsuhiko/clicolors-control#15.
```rust
use std::process::Command;
fn main() {
println!("Foo \u{1b}[36mBar\u{1b}[0m Baz"); // Cyan
println!("Foo \u{1b}[2mBar\u{1b}[0m Baz <- Not colored"); // Dim
println!("Foo \u{1b}[31mBar\u{1b}[0m Baz"); // Red
println!();
// ANSI codes render incorrectly if any of these commands are executed
Command::new("git").arg("log").status().unwrap();
// Command::new("git").arg("diff").status().unwrap();
// However, ANSI codes are not affected if any of these commands are executed
// Command::new("git").arg("--version").status().unwrap();
// Command::new("cargo").arg("--version").status().unwrap();
// Command::new("cargo").arg("build").status().unwrap();
println!();
println!("Foo \u{1b}[36mBar\u{1b}[0m Baz"); // Cyan
println!("Foo \u{1b}[2mBar\u{1b}[0m Baz"); // Dim
println!("Foo \u{1b}[31mBar\u{1b}[0m Baz"); // Red
println!();
}
```
## Python
I thought it might be my terminal being weird. So I test the equivalent in Python, and it rendered correctly.
```python
from subprocess import run
print("Foo \u001b[36mBar\u001b[0m Baz") # Cyan
print("Foo \u001b[2mBar\u001b[0m Baz") # Dim
print("Foo \u001b[31mBar\u001b[0m Baz") # Red
print()
run(["git", "log"])
print()
print("Foo \u001b[36mBar\u001b[0m Baz") # Cyan
print("Foo \u001b[2mBar\u001b[0m Baz") # Dim
print("Foo \u001b[31mBar\u001b[0m Baz") # Red
```
-----
I've tested in both debug and release mode, as well as stable and nightly. The issue remains the same.
```
$ rustc --version --verbose
rustc 1.41.0 (5e1a79984 2020-01-27)
binary: rustc
commit-hash: 5e1a799842ba6ed4a57e91f7ab9435947482f7d8
commit-date: 2020-01-27
host: x86_64-pc-windows-msvc
release: 1.41.0
LLVM version: 9.0
$ cargo --version --verbose
cargo 1.41.0 (626f0f40e 2019-12-03)
release: 1.41.0
commit-hash: 626f0f40efd32e6b3dbade50cd53fdfaa08446ba
commit-date: 2019-12-03
``` | O-windows,T-libs-api,A-docs,C-bug | low | Critical |
561,268,822 | go | text/template: Permit user of Reader objects in templates | In the case where we need to use a reader to avoid buffering large amounts of data in memory, there appears no way to use this reader in a go template.
This issue was raised before here:
https://github.com/golang/go/issues/25160
In summary, it was dismissed on the basis of "most people don't need this". I would like to contest this. There are a lot of things "most people don't need" but what if the reason for the need is an important edge case (i.e. applications processing large data sets). Surely there are some people processing applications that operate over large data sets? | NeedsInvestigation,FeatureRequest | low | Major |
561,269,929 | three.js | Seams with normal map and mirrored uvs | ##### Description of the problem
I guess it may not be just a threejs issue but I ran into a problem when using a normal map with a mesh having mirrored uvs.
With the current code, my understanding is that the RGB value (128, 128, 255) which is supposed to represent a non perturbed normal by convention is read as (0.50196, 0.50196, 1.0) - in linear color space. When converted to [-1, 1] using the function `map * 2.0 - 1.0` the vector encoded in the texture becomes (0.00392, 0.00392, 1.0) when it shall be (0.0, 0.0, 1.0). This is the reason why when using a mesh with mirrored uvs and a normal texture, the two normals at a seam have different directions creating the artifact.
It can be easily proven by modifying the **perturbNormal2Arb** function.
Replace `vec3 mapN = map * 2.0 - 1.0;` by `vec3 mapN = map * 2.0 - 1.00392;` and the seams disappear. I attached two screenshots made using a fully metallic material and no roughness to make it very obvious:
Not too sure how this problem is generally solved as this is not my area of expertise.
##### Three.js version
- [x] Dev
- [x] r113
##### Browser
- [x] All of them
##### OS
- [x] All of them
| Needs Investigation | medium | Critical |
561,271,836 | godot | Outlines go around shadows on (non-RichText) Labels | **Godot version:**
3.2
**OS/device including version:**
Windows 10
AMD FX-8320
GTX 1060 6GB
**Issue description:**
3.2 caused (non-RichText) Label nodes with an outline and a shadow to have the outline to go around the shadow, instead of the outline going around the text and the shadow going underneath the outline.
RichTextLabel nodes still have expected functionality. If you open the project in 3.1 then regular Labels also work as expected.
**Steps to reproduce:**
Create a (regular, non-RichText) Label with a shadow and outline.
**Minimal reproduction project:**
[outline_text_bug.zip](https://github.com/godotengine/godot/files/4167682/outline_text_bug.zip)
| enhancement,discussion,topic:gui | low | Critical |
561,296,086 | flutter | Make ShellTestPlatformView fully runtime configurable | Right now there is #ifdef soup in ShellTestPlatformView.cc in order to determine which backends are available.
We'd like to move to a system where all backends are built for `ShellTest` and at runtime we do proc resolution for GL or Vulkan symbols then determine which backends can be run based off that.
This is currently not possible because the code doesn't compile on all platforms (e.g. `//testing/opengl` won't compile on Fuchsia, `//flutter/vulkan` won't compile on Windows). | engine,platform-fuchsia,P3,team-engine,triaged-engine | low | Minor |
561,311,734 | terminal | TestMouseWheelReadConsoleInputHelper tests are flaky | In the InputTests, there are a handful of them that inject mouse events into the window handle.
They center around the method `TestMouseWheelReadConsoleInputHelper`.
They tend to fail when there are two input events in the queue instead of just the one they expect.
My theory is that the extra event in the queue is a focus or unfocus event.
We should make them all more robust to the point where they don't care/skip excess focus/unfocus events. | Product-Conhost,Help Wanted,Area-Input,Issue-Task,Area-CodeHealth | low | Minor |
561,314,815 | vscode | [Feature-request] Remove cursor | I would like the multi cursor to behave like this plugin - https://atom.io/packages/multi-cursor
This plugin allows to remove last created cursor by adding new cursor in oposite direction.
Exaple.
```
some text|
some text 2
some text 3
```
add cusor down
```
some text|
some text| 2
some text 3
```
add cursor dows
```
some text|
some text| 2
some text| 3
```
add cursor up
```
some text|
some text| 2
some text 3
```
add cursor up
```
some text|
some text 2
some text 3
```
this also works with ctrl - d command.
| feature-request,editor-multicursor | medium | Major |
561,335,529 | go | cmd/compile: ARM's MOVWnop causes regalloc to insert unnecessary copies | MOVWnop is a nop/copy designed to ensure type safety. It generates no code. However, it is not free; regalloc sometimes makes a copy of a value to provide to MOVWnop, since it doesn't know that MOVWnop doesn't modify its argument.
I noticed this while tracking down regressions due to changes in rewrite rule order application.
I don't know what the best approach to fixing it is. Suggestions?
cc @cherrymui
| Performance,NeedsInvestigation,compiler/runtime | low | Minor |
561,335,763 | TypeScript | Support closed-file diagnostics in VS Code | For https://github.com/microsoft/vscode/issues/13953
## Background
Project wide diagnostics are the most requested VS Code JS/TS feature. However, we've been blocked on implementing them by the fact that TS server is single threaded. This means that computing these diagnostics could interrupt normal user operations like completions.
We currently recommend that people use tasks to get around this limitation (since tasks always run as separate process) however few people actually use tasks.
With https://github.com/microsoft/vscode/commit/f0942786b4ee338f02449e2e5751091e028eb898, I've added a new setting that spins up a new TS Server just for diagnostics. This accomplishes much the same thing as using a task.
## Problem
In the prototype, I'm using the `geterrForProject` command. This works ok for smaller projects but triggers a huge number of events for larger projects like the VS Code source. Instead, we need an API that: given a file, returns a list of inverse dependencies (i.e. the files that need to be rechecked after the file changes)
I believe that this API more or less already exists with `compileOnSaveAffectedFileList`? But that command only works if `compileOnSave` is enabled. We'd like to enable project wide error reporting even when users have not enabled compileOnSave.
/cc @sheetalkamat For background on `compileOnSaveAffectedFileList`
/cc @amcasey As you suggested using a separate diagnostics only server
| Suggestion,Committed | medium | Critical |
561,392,624 | flutter | Scaffold body is hidden behind bottomSheet | <!-- 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
-->
## Steps to Reproduce
<!-- You must include full steps to reproduce so that we can reproduce the problem. -->
1. Clone https://github.com/yringler/reproduce-bottom-sheet
2. Run it
**Expected results:**
The body should be scroll-able to see all content.
**Actual results:** <!-- what did you see? -->
Part of the body content is cut off, and it can't be scrolled to | framework,f: material design,has reproducible steps,found in release: 3.3,found in release: 3.7,team-design,triaged-design | medium | Critical |
561,417,589 | godot | Path (3D) is not selectable/editable near some objects | **Godot version:**
3.2
**OS/device including version:**
Windows 10
**Issue description:**
The path circles are hard to select whenever an object is near the point you select, selecting the object insterad of creating/selecting the circle you want to handle.
**Steps to reproduce:**
Place lot of objects and then a path, try to make non-uniformed scaled objects too
**Minimal reproduction project:**
will provide if needed | bug,topic:editor,usability | low | Minor |
561,432,520 | pytorch | DataParallel gives different gradients when using LSTMs | ## 🐛 Bug
When using DataParallel on a model with LSTMs the losses obtained compared to the same model run on a single GPU are different.
## To Reproduce
Here is a sample code block that seeds everything to ensure the weights are the same.
Model:
``` python
class Model(nn.Module):
def __init__(self):
super(Model, self).__init__()
self.input_transform = nn.Linear(100, 512)
self.lstm = nn.LSTM(512,512,batch_first=True)
self.output_transform = nn.Linear(512, 10)
def forward(self, x):
self.lstm.flatten_parameters()
input_t = self.input_transform(x)
output, hidden = self.lstm(input_t)
return self.output_transform(hidden[0][0])
```
Running on a single GPU
```python
torch.manual_seed(1234)
np.random.seed(1234)
device = torch.device("cuda:0")
model = Model().to(device)
optimizer = torch.optim.Adam(model.parameters())
for i in range(5):
input = torch.rand(128,10,100).to(device)
output = torch.from_numpy(np.random.randint(10, size=(128,))).to(device)
pred = model(input)
loss = nn.functional.cross_entropy(pred, output)
print (loss)
optimizer.zero_grad()
loss.backward()
optimizer.step()
```
These are the losses:
```
tensor(2.3043, device='cuda:0', grad_fn=<NllLossBackward>)
tensor(2.3364, device='cuda:0', grad_fn=<NllLossBackward>)
tensor(2.3563, device='cuda:0', grad_fn=<NllLossBackward>)
tensor(2.2963, device='cuda:0', grad_fn=<NllLossBackward>)
tensor(2.3315, device='cuda:0', grad_fn=<NllLossBackward>)
```
Running on 4 P100s
```python
torch.manual_seed(1234)
np.random.seed(1234)
device = torch.device("cuda:0")
model = Model().to(device)
model = DataParallel(model)
optimizer = torch.optim.Adam(model.parameters())
for i in range(5):
input = torch.rand(128,10,100).to(device)
output = torch.from_numpy(np.random.randint(10, size=(128,))).to(device)
pred = model(input)
loss = nn.functional.cross_entropy(pred, output)
print (loss)
optimizer.zero_grad()
loss.backward()
optimizer.step()
```
These are the results:
```
tensor(2.3043, device='cuda:0', grad_fn=<NllLossBackward>)
tensor(2.3068, device='cuda:0', grad_fn=<NllLossBackward>)
tensor(2.3137, device='cuda:0', grad_fn=<NllLossBackward>)
tensor(2.2964, device='cuda:0', grad_fn=<NllLossBackward>)
tensor(2.3243, device='cuda:0', grad_fn=<NllLossBackward>)
```
This behaviour is on the stable version of PyTorch 1.4. I face convergence issues on my actual code that uses multiple LSTMs when using dataparallel.
Things work fine when I downgrade to PyTorch 1.2.
## Environment
```
PyTorch version: 1.4.0
Is debug build: No
CUDA used to build PyTorch: 10.0
OS: Ubuntu 18.04.2 LTS
GCC version: (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0
CMake version: Could not collect
Python version: 3.7
Is CUDA available: Yes
CUDA runtime version: Could not collect
GPU models and configuration:
GPU 0: Tesla P100-PCIE-12GB
GPU 1: Tesla P100-PCIE-12GB
GPU 2: Tesla P100-PCIE-12GB
GPU 3: Tesla P100-PCIE-12GB
Nvidia driver version: 418.40.04
cuDNN version: Could not collect
Versions of relevant libraries:
[pip] numpy==1.18.1
[pip] torch==1.4.0
[pip] torchvision==0.5.0
[conda] blas 1.0 mkl
[conda] mkl 2020.0 166
[conda] mkl-service 2.3.0 py37he904b0f_0
[conda] mkl_fft 1.0.15 py37ha843d7b_0
[conda] mkl_random 1.1.0 py37hd6b4f25_0
[conda] pytorch 1.4.0 py3.7_cuda10.0.130_cudnn7.6.3_0 pytorch
[conda] torchvision 0.5.0 py37_cu100 pytorch
``` | triaged,module: determinism,module: data parallel | low | Critical |
561,464,057 | go | syscall/js: add support for fork/exec syscalls? | <!-- Please answer these questions before submitting your issue. Thanks! -->
### What version of Go are you using (`go version`)?
<pre>
$ go version
go version devel +b7689f5aa3 Fri Jan 31 06:02:00 2020 +0000 linux/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
GO111MODULE="on"
GOARCH="amd64"
GOBIN=""
GOCACHE="/home/myitcv/.cache/go-build"
GOENV="/home/myitcv/.config/go/env"
GOEXE=""
GOFLAGS=""
GOHOSTARCH="amd64"
GOHOSTOS="linux"
GOINSECURE=""
GONOPROXY=""
GONOSUMDB=""
GOOS="linux"
GOPATH="/home/myitcv/gostuff"
GOPRIVATE=""
GOPROXY="https://proxy.golang.org,direct"
GOROOT="/home/myitcv/gos"
GOSUMDB="sum.golang.org"
GOTMPDIR=""
GOTOOLDIR="/home/myitcv/gos/pkg/tool/linux_amd64"
GCCGO="gccgo"
AR="ar"
CC="gcc"
CXX="g++"
CGO_ENABLED="1"
GOMOD="/home/myitcv/gostuff/src/github.com/myitcv/playground/go.mod"
CGO_CFLAGS="-g -O2"
CGO_CPPFLAGS=""
CGO_CXXFLAGS="-g -O2"
CGO_FFLAGS="-g -O2"
CGO_LDFLAGS="-g -O2"
PKG_CONFIG="pkg-config"
GOGCCFLAGS="-fPIC -m64 -pthread -fmessage-length=0 -fdebug-prefix-map=/tmp/go-build703123889=/tmp/go-build -gno-record-gcc-switches"
</pre></details>
### What did you do?
```
-- go.mod --
module mod
-- main.go --
package main
import (
"fmt"
"os"
"os/exec"
)
func main() {
cmd := exec.Command("echo", "hello")
cmd.Stdout = os.Stdout
cmd.Stderr = os.Stderr
if err := cmd.Run(); err != nil {
fmt.Fprintln(os.Stderr, err)
os.Exit(1)
}
}
```
```
GOOS=js GOARCH=wasm go build -o main.wasm
$(go env GOROOT)/misc/wasm/go_js_wasm_exec main.wasm
```
### What did you expect to see?
```
hello
```
### What did you see instead?
```
exec: "echo": executable file not found in $PATH
```
As some folks on Slack kindly pointed out, this is because `GOOS=js ARCH=wasm` does not support fork/exec syscalls. Those same folks were not clear that it made sense for this to be supported (especially given the browser focus of this port), but when executing through NodeJS (i.e. the `go_js_wasm_exec` wrapper) it does have meaning.
Thoughts?
cc @neelance | NeedsInvestigation,FeatureRequest,arch-wasm | low | Critical |
561,489,820 | angular | @HostBinding('attr.id') with dynamic attribute during testing conflicting Karma logic | # 🐞 bug report
### Affected Package
The issue is caused by package @angular/core, @angular/core/testing
### Is this a regression?
No
### Description
Creating elements that bind to the id attribute of their host from an input causes Karma to display all previously tested elements instead of clearing after each spec.
It seems the issue occurs when the property that is bound is set programmatically during the tests.
## 🔬 Minimal Reproduction
https://github.com/hodossy/ng-hostbinding-id-test-issue
<pre><code>
npm install
ng test
</code></pre>
## 🔥 Exception or Error
As it can be seen, multiple test elements are left behind, probably because Karma clears the elements by ID.
My expactation would be that Karma removes the previous tested element and only the last one is visible after the test suite is run.
## 🌍 Your Environment
**Angular Version:**
<pre><code>
Angular CLI: 8.3.25
Node: 10.16.0
OS: win32 x64
Angular: 8.2.14
... animations, common, compiler, compiler-cli, core, forms
... language-service, platform-browser, platform-browser-dynamic
... router
Package Version
-----------------------------------------------------------
@angular-devkit/architect 0.803.25
@angular-devkit/build-angular 0.803.25
@angular-devkit/build-optimizer 0.803.25
@angular-devkit/build-webpack 0.803.25
@angular-devkit/core 8.3.25
@angular-devkit/schematics 8.3.25
@angular/cli 8.3.25
@ngtools/webpack 8.3.25
@schematics/angular 8.3.25
@schematics/update 0.803.25
rxjs 6.4.0
typescript 3.5.3
webpack 4.39.2
</code></pre>
**Anything else relevant?**
I had no time to test with other browsers than Chrome, but I think it should not be specific to Chrome
| type: bug/fix,area: testing,freq1: low,P3 | low | Critical |
561,507,967 | three.js | Too dark edges due to new IBL multi-scattering approach | To continue the discussion of #16409, it still seems to me that three.js's new IBL multi-scattering code is yielding too strong dark edges on light fully glossy material.
Maybe this is now something that is acted as acceptable or even an improvement (as the author of the [new approach's paper](http://www.jcgt.org/published/0008/01/03/) states), but I thought I'd report examples and then have an official decision or start an improvement process if need here.
Here is a demo using the [current latest (r113)](https://s3.amazonaws.com/static.sayduck.com/2020.02.06.rendering_tests/three_r113_test.html)
Here is the [same scene before the IBL multi-scattering update (r104)](https://s3.amazonaws.com/static.sayduck.com/2020.02.06.rendering_tests/three_r104_test.html)
And as a reference, here is the [same scene (using the same environment map converted to DDS) in BabylonJS](https://www.babylonjs-playground.com/#MITBWZ#1)
 | Needs Investigation | low | Major |
561,537,539 | pytorch | not able to import * from fastai.vision in Google collab | ## 🐛 Bug
<!-- A clear and concise description of what the bug is. -->
## To Reproduce
run in Google collab
from fastai.vision import *
Error:
---------------------------------------------------------------------------
ImportError Traceback (most recent call last)
<ipython-input-3-c0e76450f370> in <module>()
----> 1 from fastai.vision import *
5 frames
/usr/local/lib/python3.6/dist-packages/torchvision/models/inception.py in <module>()
6 import torch.nn as nn
7 import torch.nn.functional as F
----> 8 from torch.jit.annotations import Optional
9 from torch import Tensor
10 from .utils import load_state_dict_from_url
ImportError: cannot import name 'Optional'
| triaged | low | Critical |
561,551,233 | TypeScript | Optional chaining should always include `undefined` in its resolved type | **TypeScript Version:** 3.7.5
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:** optional chaining return type
**Code**
I have a formatting function, where the input should always be a string, but in certain cases, from the backend, it's `undefined` for some reason. So to make sure the front-end at least doesn't crash when that happens, I added optional chaining, which fixed the issue.
But the types are kind of off now, and I didn't realize until now because Typescript didn't complain about anything.
```ts
export const formatAccountNumber = (account: string): string =>
account?.replace(/^(\d{4})(\d{2})(\d{5})/, '$1 $2 $3');
```
**Expected behavior:** Should get an error/warning about _something_ here. Preferably, I guess optional chaining should always expand the type to include `undefined` to whatever it is added to? Then, in this example, Typescript could say that the return type is `string`, but I'm trying to return `string | undefined`? Either that, or maybe it shouldn't be possible to use optional chaining on a value that does not include either the type `undefined` and/or `null`?
**Actual behavior:** No warnings or errors.
**Related Issues:** no | Suggestion,In Discussion | low | Critical |
561,552,358 | flutter | Can't bring custom selection into view from Scrollable when showing keyboard | I have a forgot password screen that contains a button at the bottom of the ListView.
The default behaviour is the image below. But we want to make sure that the button is visible as well.
This is the desired behaviour:
This code sample below is what I have at this point. The animate to does not work because the keyboard animates up. and at startup the the listview can not scroll so there is no need to scroll to the bottom. How can I fix this?
A small screen that could be used for testing:
```
class ForgotPasswordScreen extends StatefulWidget {
const ForgotPasswordScreen();
@override
State<StatefulWidget> createState() => ForgotPasswordScreenState();
}
class ForgotPasswordScreenState extends State<ForgotPasswordScreen> {
ScrollController _scrollController;
@override
void initState() {
super.initState();
_scrollController = ScrollController(keepScrollOffset: true);
_scrollToBottom();
}
@override
Widget build(BuildContext context) {
return Scaffold(
appBar: AppBar(),
body: SafeArea(
child: ListView(
controller: _scrollController,
children: [
Center(
child: Padding(
padding: const EdgeInsets.all(34),
child: Column(
children: [
Text(
'long textlong textlong textlong textlong textlong textlong textlong textlong textlong textlong textlong textlong textlong textlong text',
style: ThemeTextStyles.inputTextTextStyle,
),
Container(
height: 47,
),
TextField(
autofocus: true,
decoration: InputDecoration(
border: const OutlineInputBorder(),
),
),
Container(
height: 47,
),
MaterialButton(
color: Colors.deepOrange,
child: Text('button'),
onPressed: () {},
)
],
),
),
),
],
),
),
);
}
Future<void> _scrollToBottom() async {
await _scrollController.animateTo(
_scrollController.position.maxScrollExtent,
duration: Duration(milliseconds: 250),
curve: Curves.easeIn,
);
}
}
```
Flutter Doctor:
```
[✓] Flutter (Channel unknown, v1.12.13+hotfix.5, on Mac OS X 10.15.1 19B88, locale en-BE)
• Flutter version 1.12.13+hotfix.5 at /Users/vanlooverenkoen/flutter
• Framework revision 27321ebbad (8 weeks ago), 2019-12-10 18:15:01 -0800
• Engine revision 2994f7e1e6
• Dart version 2.7.0
[✓] Android toolchain - develop for Android devices (Android SDK version 29.0.2)
• Android SDK at /Users/vanlooverenkoen/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: /Applications/Android Studio.app/Contents/jre/jdk/Contents/Home/bin/java
• Java version OpenJDK Runtime Environment (build 1.8.0_202-release-1483-b49-5587405)
• All Android licenses accepted.
[✓] Xcode - develop for iOS and macOS (Xcode 11.3)
• Xcode at /Applications/Xcode.app/Contents/Developer
• Xcode 11.3, Build version 11C29
• CocoaPods version 1.8.4
[✓] Chrome - develop for the web
• Chrome at /Applications/Google Chrome.app/Contents/MacOS/Google Chrome
[✓] Android Studio (version 3.5)
• Android Studio at /Applications/Android Studio.app/Contents
• Flutter plugin version 43.0.1
• Dart plugin version 191.8593
• Java version OpenJDK Runtime Environment (build 1.8.0_202-release-1483-b49-5587405)
[!] VS Code (version 1.42.0)
• VS Code at /Applications/Visual Studio Code.app/Contents
✗ Flutter extension not installed; install from
https://marketplace.visualstudio.com/items?itemName=Dart-Code.flutter
[✓] Connected device (4 available)
• iPhone 5s • 2A34F270-D88D-445A-B723-E6C9C439E662 • ios • com.apple.CoreSimulator.SimRuntime.iOS-12-4 (simulator)
• macOS • macOS • darwin-x64 • Mac OS X 10.15.1 19B88
• Chrome • chrome • web-javascript • Google Chrome 78.0.3904.108
• Web Server • web-server • web-javascript • Flutter Tools
```
| a: text input,framework,f: scrolling,f: focus,has reproducible steps,P3,team-framework,triaged-framework,found in release: 3.13,found in release: 3.17 | low | Major |
561,554,398 | youtube-dl | zen.yandex.ru/media | <!--
######################################################################
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 2020.01.24. 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 **2020.01.24**
- [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://zen.yandex.ru/media/id/5beae6995521b800ab4e3e50/5e1c33f0aad43600ae4164d6
## Description
<!--
Provide any additional information.
If work on your issue requires account credentials please provide them or explain how one can obtain them.
-->
| site-support-request | low | Critical |
561,562,678 | godot | Borderless Windows Crash Desktop Environment | **Godot version: 3.2**
**OS/device including version:** Linux Mint 19.3 Tricia x86_64
**Issue description:** Project > Project Settings... > Display > Window > Size: Borderless when set to True will cause Cinnamon Desktop to crash upon launching the game. When Cinnamon is restarted, game seems to work fine and not crash again.
**Steps to reproduce:** After setting Borderless "True", Play the project and click on anything inside the window.
**Minimal reproduction project:** it was tried with a scene composed of just a top Control node with a Panel child
| bug,platform:linuxbsd,topic:porting | low | Critical |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.