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 |
|---|---|---|---|---|---|---|
653,467,754 |
flutter
|
Can't test navigation between Flutter Web pages using Flutter Driver
|
Flutter Driver has only very basic documentation for simple tests like finding a button and pressing the button. Anyone know if I can do other operations like navigate to a specific page from driver (e.g. using a # url fragment), test a link which leads to an external site, check visual setup of the page against an image, and other such tests which would be standard in Silenium and the like?
Or alterantively a strategy for using Silenium/Ghost Inspector with Flutter Web
Thanks!
|
framework,t: flutter driver,f: routes,customer: webeap,P3,team-framework,triaged-framework
|
low
|
Minor
|
653,468,291 |
pytorch
|
Memory bug for backward on torch.sparse.mm?
|
## 🐛 Bug
<!-- A clear and concise description of what the bug is. -->
## To Reproduce
Steps to reproduce the behavior:
<!-- If you have a code sample, error messages, stack traces, please provide it here as well -->
> import torch
import numpy as np
print(f'GPU memory usage: {torch.cuda.memory_allocated(device=None)/10\**9}')
n = 110000
nnz = 860000
rows = np.random.randint(0, n, nnz)
cols = np.random.randint(0, n, nnz)
values = torch.randn(nnz)
X_sparse = torch.sparse_coo_tensor([rows,cols], values, size=(n,n)).cuda().requires_grad_(True)
Y_dense = torch.randn((n,200)).cuda().requires_grad_(True)
print(f'GPU memory usage: {torch.cuda.memory_allocated(device=None)/10\**9}')
input("Press")
t = torch.sparse.mm(X_sparse, Y_dense).sum()
t.backward()
print(f'SPMM, x.grad: {X_sparse.grad}, y.grad: {Y_dense.grad}')
print(f'GPU memory usage: {torch.cuda.memory_allocated(device=None)/10\**9}')
## Expected behavior
<!-- A clear and concise description of what you expected to happen. -->
I got an error about OOM only in the backward step. `RuntimeError: CUDA out of memory. Tried to allocate 45.08 GiB (GPU 0; 10.73 GiB total capacity; 202.00 MiB already allocated; 9.68 GiB free; 204.00 MiB reserved in total by PyTorch)`.
I think the memory usage of backward should be similar to that of forward. Before backward, there is still ~9GB free. I expect it would not try to allocate 45GB. I don't know whether there is a bug in backward support for torch.sparse.mm? Feel free to correct me if I am wrong.
## Environment
Please copy and paste the output from our
[environment collection script](https://raw.githubusercontent.com/pytorch/pytorch/master/torch/utils/collect_env.py)
(or fill out the checklist below manually).
You can get the script and run it with:
```
wget https://raw.githubusercontent.com/pytorch/pytorch/master/torch/utils/collect_env.py
# For security purposes, please check the contents of collect_env.py before running it.
python collect_env.py
```
- PyTorch Version (e.g., 1.0): 1.4.0
- OS (e.g., Linux): Ubuntu 18.04.2 LTS
- How you installed PyTorch (`conda`, `pip`, source): pip
- Build command you used (if compiling from source):
- Python version: 3.7
- CUDA/cuDNN version: 10.1
- GPU models and configuration: GPU 0: GeForce RTX 2080 Ti
- Any other relevant information:
## Additional context
<!-- Add any other context about the problem here. -->
I am not sure if it is a bug. But if so and anyone can point out the problem, I am willing to help and dig further to see how to fix it.
cc @ezyang @SsnL @albanD @zou3519 @gqchen @vincentqb @aocsa
|
module: sparse,module: autograd,triaged
|
low
|
Critical
|
653,506,524 |
pytorch
|
libtorch cmake should PUBLIC ally advertise it as requiring C++14
|
In a bunch of our sample CMakeLists.txt, for example https://pytorch.org/tutorials/advanced/torch_script_custom_ops.html#building-with-cmake we explicitly force our end users to specify that the extension is C++14. Modern cmake lets us define a target that will automatically enforce c++14 on downstream users: https://stackoverflow.com/a/46809634/23845 We should use this, and remove those lines.
cc @ezyang @seemethere @malfet
|
module: build,triaged
|
low
|
Minor
|
653,551,810 |
pytorch
|
Sum of two tuples becomes a list in torchscript
|
## 🐛 Bug
Sum of two tuples becomes a list in torchscript
## To Reproduce
```python
import torch
from torch import Tensor
from torch.jit.annotations import Tuple
def foo4() -> Tuple[Tensor, Tensor, Tensor, Tensor]:
out1 = (torch.tensor(0), torch.tensor(0),)
out2 = (torch.tensor(0), torch.tensor(0),)
return out1 + out2
scripted_fn = torch.jit.script(foo4)
scripted_fn()
>
RuntimeError:
Return value was annotated as having type Tuple[Tensor, Tensor, Tensor, Tensor] but is actually of type List[Tensor]:
File "<ipython-input-34-1cd92d739981>", line 11
out2 = (torch.tensor(0), torch.tensor(0),)
return out1 + out2
~~~~~~~~~~~~~~~~~~ <--- HERE
```
## Expected behavior
Output type should be a tuple
## Environment
- PyTorch Version (e.g., 1.0): '1.7.0.dev20200701'
- OS (e.g., Linux): linux
- How you installed PyTorch (`conda`, `pip`, source): conda
- Build command you used (if compiling from source):
- Python version: 3.7
- CUDA/cuDNN version:
- GPU models and configuration:
- Any other relevant information:
## Additional context
<!-- Add any other context about the problem here. -->
cc @suo @gmagogsfm
|
oncall: jit
|
low
|
Critical
|
653,552,663 |
flutter
|
Show a warning when building APK/AAB with unsupported architecture
|
See https://github.com/flutter/flutter/issues/37566 and https://github.com/flutter/flutter/issues/59501.
It's possible for users to end up adding their own native libraries in their Flutter Android project's gradle. If the user ends up adding an x86 .so to their project, the APK will be installable on x86 physical devices, cause it to skip the ARM emulation and crash when trying to load Flutter since there are no flutter .so.
In flutter.gradle, we should look for the presence of native libraries in the APK with unsupported architectures and show a warning to prevent this subtle crasher.
cc @jmagman @blasten @chinmaygarde
|
c: new feature,c: crash,platform-android,tool,t: gradle,P3,team-android,triaged-android
|
low
|
Critical
|
653,574,290 |
opencv
|
in opencv.js findTransformECC, argument inputMask is required (vs. documented optional) and does not seem to accept any input
|
##### System information (version)
- OpenCV => 4.3.0 (opencv.js)
- Operating System / Platform => Windows 64 Bit
- Compiler => using Javascript
##### Detailed description
Using opencv.js (4.3.0 and tested todays build as well) findTransformECC has 2 optional parameter (inputMask and gaussFiltSize) but if you don't include them you get an error that they are required.
I have not found a way to represent ("inputMask=None") without getting an unhandled exception. "null" also does not work and a mask does not work.
##### Steps to reproduce
```
function Align_img(){
let image_baseline = cv.imread(imgElement_Baseline);
let image = cv.imread('imageChangeup');
let im1_gray = new cv.Mat();
let im2_gray = new cv.Mat();
let im2_aligned = new cv.Mat();
//get size of baseline image
width1 = image_baseline.cols;
height1 = image_baseline.rows;
//resize image to baseline image
let dim1 = new cv.Size(width1, height1);
cv.resize(image, image, dim1, cv.INTER_AREA);
// Convert images to grayscale
cv.cvtColor(image_baseline, im1_gray, cv.COLOR_BGR2GRAY);
cv.cvtColor(image, im2_gray, cv.COLOR_BGR2GRAY);
// Find size of image1
let dsize = new cv.Size(image_baseline.rows, image_baseline.cols);
// Define the motion model
warp_mode = cv.MOTION_HOMOGRAPHY;
// Define 3x3 matrix and initialize the matrix to identity
let warp_matrix = cv.Mat.eye(3, 3, cv.CV_8U);
// Specify the number of iterations.
number_of_iterations = 5000;
// Specify the threshold of the increment in the correlation coefficient between two iterations
termination_eps = 0.0000000001; //1e-10;
// Define termination criteria
criteria = (cv.TERM_CRITERIA_EP | cv.TERM_CRITERIA_COUNT, number_of_iterations, termination_eps);
//Run the ECC algorithm. The results are stored in warp_matrix.
cv.findTransformECC(im1_gray, im2_gray, warp_matrix, warp_mode, criteria, null, 1);
// Use warpPerspective for Homography
cv.warpPerspective (image, im2_aligned, warp_matrix, dsize, cv.INTER_LINEAR + cv.WARP_INVERSE_MAP);
cv.imshow('imageChangeup', im2_aligned);
im1_gray.delete();
im2_gray.delete();
im2_aligned.delete();
};
```
##### Issue submission checklist
- [x] I report the issue, it's not a question
<!--
OpenCV team works with answers.opencv.org, Stack Overflow and other communities
to discuss problems. Tickets with question without real issue statement will be
closed.
-->
- [x] I checked the problem with documentation, FAQ, open issues,
answers.opencv.org, Stack Overflow, etc and have not found solution
<!--
Places to check:
* OpenCV documentation: https://docs.opencv.org
* FAQ page: https://github.com/opencv/opencv/wiki/FAQ
* OpenCV forum: https://answers.opencv.org
* OpenCV issue tracker: https://github.com/opencv/opencv/issues?q=is%3Aissue
* Stack Overflow branch: https://stackoverflow.com/questions/tagged/opencv
-->
- [x] I updated to latest OpenCV version and the issue is still there
<!--
master branch for OpenCV 4.x and 3.4 branch for OpenCV 3.x releases.
OpenCV team supports only latest release for each branch.
The ticket is closed, if the problem is not reproduced with modern version.
-->
- [x] There is reproducer code and related data files: videos, images, onnx, etc
<!--
The best reproducer -- test case for OpenCV that we can add to the library.
Recommendations for media files and binary files:
* Try to reproduce the issue with images and videos in opencv_extra repository
to reduce attachment size
* Use PNG for images, if you report some CV related bug, but not image reader
issue
* Attach the image as archite to the ticket, if you report some reader issue.
Image hosting services compress images and it breaks the repro code.
* Provide ONNX file for some public model or ONNX file with with random weights,
if you report ONNX parsing or handling issue. Architecture details diagram
from netron tool can be very useful too. See https://lutzroeder.github.io/netron/
-->
|
category: javascript (js)
|
low
|
Critical
|
653,585,930 |
three.js
|
PropertyBinding: Unit test for custom interpolants
|
In https://github.com/mrdoob/three.js/issues/19239 we discussed wanting to have a test covering use of custom interpolants with the animation system, which GLTFLoader relies on. I haven't had the chance yet to do so, but don't want to lose track of that thread, so filing a new issue here.
|
Suggestion
|
low
|
Minor
|
653,586,280 |
pytorch
|
Support for More Loss Functions
|
## 🚀 Feature
Supporting more loss functions like `FocalLoss` or `DICE` natively will be a nice addition, especially as their already implementations to them out there.
## Motivation
Native support for more loss functions will reduce boilerplate code.
## Pitch
Review and addition of loss functions found [Here](https://github.com/JunMa11/SegLoss)
## Alternatives
There is implementations on PyTorch Forums for some of the loss functions mentioned above but because they are not reviewed they sometimes have big flaws.
|
feature,module: loss,triaged
|
low
|
Major
|
653,612,196 |
go
|
x/pkgsite: in cmd/frontend's --proxy_url argument, exclude localhost from https requirement
|
### What did you do?
I tried to start `pkgsite/cmd/frontend` using this command line:
```
./frontend --direct_proxy --proxy_url http://localhost:3000
```
I want to use `http://localhost:3000` with the `--proxy_url` argument, because I'm trying to create a small kubernetes deployment with the athens module proxy and pkgsite in the same pod.
### What did you expect to see?
pkgsite/cmd/frontend starts without errors using `http://localhost:3000` as the proxy url.
### What did you see instead?
The below fatal error:
```
Error: proxy.New("http://localhost:3000"): scheme must be https (got http)
```
---
In https://github.com/golang/pkgsite/blob/master/internal/proxy/client.go#L53 if we exclude localhost from the https requirement, that would solve my immediate problem.
But later on, if we want to move athens and pkgsite into their own separate deployments, I'll have to point pkgsite to `http://athens.develop:3000`, and that won't work (we don't have https for communication within the cluster)
If we want to allow all such internal communications, another option would be to resolve the hostname, and if it is in a private IP address range, then don't expect https. But I'm not sure if that would have implications in other private IP situations.
|
NeedsInvestigation,pkgsite
|
low
|
Critical
|
653,615,697 |
PowerToys
|
[KBM] Add a hotkey key to turn it on/off
|
A way in which to easily toggle on and off the keyboard manager feature without having to leave fullscreen apps, specifically games.
A shortcut keybind that is resignable that would allow for the keyboard manager to be toggled on and off without the need to alt+tab out of games.
A resign-able shortcut key to toggle keyboard manager. This is for those of us out there who game, having an easy to access way to toggle the keyboard from alt setups to default qwerty for WASD would be great. At the moment I need to alt+tab and manually click the feature off and on and it is very cumbersome. I had a previous setup with a custom keyboard and portable keyboard layout, but there was always the odd app that would choose the wrong keyboard, so powertoys is better, but the toggle would be a huge help.
|
Idea-Enhancement,Help Wanted,Product-Keyboard Shortcut Manager,Product-Settings
|
medium
|
Critical
|
653,618,228 |
pytorch
|
Source code for some /_modules/.../.../ not displaying on https://pytorch.org/docs/master/_modules/
|
## 📚 Documentation
From the module code overview page: https://pytorch.org/docs/master/_modules/ there are some modules whose source code documentation page isn't properly displaying and instead displays the "*Oops! You've reached a dead end*" page.
e.g.
* https://pytorch.org/docs/master/_modules/torch/__config__.html
* https://pytorch.org/docs/master/_modules/torch/_jit_internal.html
* https://pytorch.org/docs/master/_modules/torch/_lobpcg.html
* https://pytorch.org/docs/master/_modules/torch/_lowrank.html
* https://pytorch.org/docs/master/_modules/torch/jit/_script.html
* https://pytorch.org/docs/master/_modules/torch/jit/_trace.html
When I build the docs using a local copy of PyTorch however, the HTML files properly build for these pages.
e.g. for `torch/__config__`
**Locally built:**
<img width="960" alt="locally-built-pytorch-docs" src="https://user-images.githubusercontent.com/54918401/86973415-85804600-c142-11ea-9358-2ebf95152bf8.png">
**Public page on PyTorch website:**
<img width="1044" alt="public-pytorch-docs" src="https://user-images.githubusercontent.com/54918401/86973452-9335cb80-c142-11ea-8c88-0d39b1ba636c.png">
<!-- A clear and concise description of what content in https://pytorch.org/docs is an issue. If this has to do with the general https://pytorch.org website, please file an issue at https://github.com/pytorch/pytorch.github.io/issues/new/choose instead. If this has to do with https://pytorch.org/tutorials, please file an issue at https://github.com/pytorch/tutorials/issues/new -->
cc @jlin27
|
module: docs,triaged
|
low
|
Minor
|
653,619,638 |
pytorch
|
eager mode quantization should remove qconfig in the (non-leaf module of) quantized model
|
After eager mode quantization, we find that qconfig are not removed for non-leaf modules, this will affect future function calls to numeric suite etc.
cc @jianyuh @raghuramank100 @jamesr66a @vkuzo @jgong5 @Xia-Weiwen @leslie-fang-intel @jerryzh168 @dzhulgakov
|
oncall: quantization,low priority,triaged
|
low
|
Minor
|
653,621,540 |
go
|
x/tools/cmd/goyacc: Please add "version" command or flag
|
<!--
Please answer these questions before submitting your issue. Thanks!
For questions please use one of our forums: https://github.com/golang/go/wiki/Questions
-->
### What version of Go are you using (`go version`)?
```
bash-4.2$ go version
go version go1.12.15 linux/amd64
```
### Does this issue reproduce with the latest release?
Can't tell, see below for details.
### What operating system and processor architecture are you using (`go env`)?
<details><summary><code>go env</code> Output</summary><br><pre>
bash-4.2$ go env
GOARCH="amd64"
GOBIN=""
GOCACHE="/home/prow/.cache/go-build"
GOEXE=""
GOFLAGS=""
GOHOSTARCH="amd64"
GOHOSTOS="linux"
GOOS="linux"
GOPATH="/home/prow/go"
GOPROXY=""
GORACE=""
GOROOT="/usr/local/go"
GOTMPDIR=""
GOTOOLDIR="/usr/local/go/pkg/tool/linux_amd64"
GCCGO="gccgo"
CC="gcc"
CXX="g++"
CGO_ENABLED="1"
GOMOD=""
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-build991634038=/tmp/go-build -gno-record-gcc-switches"
</pre></details>
### What did you do?
Ran `goyacc -version`:
```
bash-4.2$ goyacc -version
flag provided but not defined: -version
```
### What did you expect to see?
I expected a version to be output.
### What did you see instead?
I received an error message.
I'm working on a project that uses `goyacc` to generate files. There are differences in the generated files between versions. Experentially, I can tell that I am using a different version of `goyacc` in my local environment than is being used in my CI container because the output is different, but I have no way to tell what version is being used where because this flag is not defined.
I spent hours today trying to pin down the source of the differences, and which version of `goyacc` I needed to be using in order to get a clean build that matched CI. I ended up needing to use `git bisect` on the `goyacc` tool to determine a known good version. I even tried using `readelf -x` and `strings` on the `goyacc` binary to determine version information, but I could not find anything there.
Please consider adding a `-version` flag to the `goyacc` tool so it is clear and easy to determine differences in output between different versions of the tool. Thank you!
|
help wanted,NeedsFix,FeatureRequest,Tools
|
low
|
Critical
|
653,624,781 |
TypeScript
|
Difference in overload pick between JS and TS with default generic type parameter
|
<!-- 🚨 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:** 4.0.0-dev.20200703, 3.9.2 (playground)
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:** js ts overload difference, javascript different overload, javascript wrong overload, javascript overload generic empty object
**Code**
`definition.d.ts`
```ts
export class EmptyObjectOverloads<T = {}> {
constructor(a: Readonly<T>);
/** @deprecated */
constructor(a: T, b?: any);
}
export class RecordOverloads<T = Record<string, any>> {
constructor(a: Readonly<T>);
/** @deprecated */
constructor(a: T, b?: any);
}
```
`ts-example.ts`
```ts
import { EmptyObjectOverloads, RecordOverloads } from './declaration';
class TSExampleEmpty extends EmptyObjectOverloads {
constructor(a: any) {
// hover over super: constructor EmptyObjectOverloads<{}>(a: Readonly<{}>): EmptyObjectOverloads<{}> (+1 overload)
super(a);
}
}
class TSExampleRecord extends RecordOverloads {
constructor(a: any) {
// hover over super: constructor RecordOverloads<Record<string, any>>(a: Readonly<Record<string, any>>): RecordOverloads<Record<string, any>> (+1 overload)
super(a);
}
}
```
`js-example.js`
```js
class JSExampleEmpty extends EmptyObjectOverloads {
/** @param {*} a */
constructor(a) {
// hover over super: constructor EmptyObjectOverloads<any>(a: any, b?: any): EmptyObjectOverloads<any> (+1 overload)
super(a); // deprecation warning '(a: any, b?: any): EmptyObjectOverloads<any>' is deprecated ts(6385)
}
}
class JSExampleRecord extends RecordOverloads {
/** @param {*} a */
constructor(a) {
// hover over super: constructor RecordOverloads<Record<string, any>>(a: Readonly<Record<string, any>>): RecordOverloads<Record<string, any>> (+1 overload)
super(a);
}
}
```
**Expected behavior:**
Both `TSExampleEmpty` and `JSExampleEmpty` pick same overload similar to `TSExampleRecord` and `JSExampleRecord`.
**Actual behavior:**
`TSExampleEmpty` uses default generic value `{}` and picks first overload.
For `JSExampleEmpty` generic value to is `any` and second (deprecated) overload picked, while `JSExampleRecord` works as expected.
In real world examples this behavior marks all `super(props);` in React class components in JS files as deprecated.
Of course it's impossible to report TS issue without mentioning React 😆 related [constructor declaration](https://github.com/DefinitelyTyped/DefinitelyTyped/blob/b5b627cf71bebf6dc7fcfdc7284a44065d378e20/types/react/index.d.ts#L478-L483)
**Playground Link:**
[Example TS](https://www.staging-typescript.org/play?ts=4.0.0-dev.20200703#code/JYWwDg9gTgLgBAJQKYEMDGMA0cDecDCE4EAdkifAL5wBmURcA5FKhowNwBQnaANigGcBcALIBPAEIBXGDFJwkADxjkAJsMLEyFXJwCQaUgJhQpGaAAow9MAIBccFCTEBKXXr0B6T3EMljpuZQiKwwAHSakNowADw4lNjx2E5iAHxWNvYhKKqkvGJxlKkuDsjo4ZGk5LFJuAmOzqlwFgDUAIxwEABuSFC8EDku+noCUmC9GRC2Luxw+pScC0A) [Example JS](https://www.staging-typescript.org/play?ts=4.0.0-dev.20200703&useJavaScript=true#code/JYWwDg9gTgLgBAJQKYEMDGMA0cDecDCE4EAdkifAL5wBmURcA5FKhowNwBQnaANigGcBcALIBPAEIBXGDFJwkADxjkAJsMLEyFXJwCQAegBURuAAEwKKChC4j1MPTDCjB-WlICYUKRmgAKRwhnAEpdPUMDOA8SLx8-KERWGAA6TUhtGAAeFBIxbFz8uEKAPkCnAQAuYrzsGJVlAH5qwpDq5HRU9NJybMKC2pqxErh-AGoARjgIADckKF4IFFUQ-T0BKTB58uCBEPY4AyjGHecWwfqlGGahtoIiDN6cwf6hksY4YGFVJEckNBQKlU-gAbABmAAcAFZVnpKJx4UA)
**Related Issues:** Haven't found any
|
Needs Investigation
|
low
|
Critical
|
653,630,893 |
rust
|
`unused_lifetimes` warning misses unused lifetime
|
I discovered an unused lifetime in my code that the compiler didn't tell me about. I was able to reduce the problem to the following snippet.
```rust
#![warn(unused_lifetimes)]
struct Foo<'a>(for<'r> fn(&'r mut Foo<'a>));
impl<'a> Foo<'a> {
pub fn new() -> Self {
Self(Self::a)
}
pub fn a(&mut self) {}
}
fn main() {
let f = Foo::new();
drop(f);
}
```
The lifetime 'a is unused, but the compiler doesn't complain at all. I confirmed that it still exists on stable 1.44.1 and on nightly 1.46.0.
|
A-lints,A-lifetimes,T-compiler,C-bug
|
low
|
Minor
|
653,652,935 |
node
|
Add a `stream` event for client-side HTTP2 sessions
|
**Is your feature request related to a problem? Please describe.**
```js
const http2 = require('http2');
const session = http2.connect('https://google.com');
// Unfortunately there is no API for this.
// It would be very useful for example to catch how many streams are pending in this session.
session.once('stream', stream => {
console.log('Got stream!');
});
const request = session.request();
session.destroy();
```
**Describe the solution you'd like** Introduce a new `stream` event to HTTP2 client-side sessions, which is emitted on **new requests**. Push streams could be emitted e.g. via `pushStream`.
**Describe alternatives you've considered** None.
|
http2
|
low
|
Minor
|
653,655,182 |
node
|
Graceful close for HTTP2 sessions
|
**Is your feature request related to a problem? Please describe.**
IMO it would be very nice to have `session.gracefullyClose()` so pending streams can be finished but no more streams can be created on this session ~(or closing is canceled when `session.request()` is called)~. I'm developing some sort of a pool (agent?) for `http2`, and I need to gracefully close a session when it becomes a subset of another session.
I do **NOT** want to notify the server because they can reject current streams with `REFUSED_STREAM` which means it hasn't processed the request yet and it is safe to retry. But the request body could have been uploaded in that case.
**Describe the solution you'd like**
See the paragraph above.
**Describe alternatives you've considered** I can implement this on my own, eg. `gracefullyClose(session)`.
|
help wanted,http2
|
medium
|
Major
|
653,666,082 |
pytorch
|
Advanced indexing gradient is extremely slow when there are many duplicate indices
|
## 🐛 Bug
We observed a ~2x slowdown in training performance of a detection network when we did broadcast on a tensor.
This broadcasting operation launches the following kernel in the backward pass which completely eats away performance:
https://github.com/pytorch/pytorch/blob/master/aten/src/ATen/native/cuda/Indexing.cu#L25
The index of the broadcast has a lot of duplicates and the broadcast is done on the last dimension.
There is no performance loss when the index doesn't contain duplicates.
Also, it doesn't matter if the index is sorted or unsorted. The code path which calls `indexing_backward_kernel` function sorts the index before computation.
## To Reproduce
Steps to reproduce the behavior:
1. Here is a python script to see the difference in performance:
```
import torch
from time import time
input_tensor = torch.randn(3, 224, 224, requires_grad=True, device='cuda')
bad_index = torch.randint(19999, 20000, (1, 224, 512), dtype=torch.long, device='cuda')
index = torch.randint(0, 20000, (1, 224, 512), dtype=torch.long, device='cuda')
x = torch.randint(0, 20, [3, 40000], dtype=torch.float32, device='cuda', requires_grad=True)
##### Good Indexing ######
#warming up cache
for i in range(10):
y = x[:, index]
torch.cuda.synchronize()
t1 = time()
#broadcasting
y = x[:, index]
z = y.sum()
z.backward()
torch.cuda.synchronize()
t1 = time() - t1
##### Bad Indexing ######
#warming up cache
for i in range(10):
y = x[:, bad_index]
torch.cuda.synchronize()
t2 = time()
#broadcasting
y = x[:, bad_index]
z = y.sum()
z.backward()
torch.cuda.synchronize()
t2 = time() - t2
print("Bad Indexing time: {:.3f}ms".format(t2*1000))
print("Good Indexing time: {:.3f}ms".format(t1*1000))
print("Ratio of Bad Indexing vs Good Indexing: {:.4f}x".format(t2/t1))
```
**Output:**
```
> python indexing_perf_bug.py
Bad Indexing time: 41.887ms
Good Indexing time: 2.191ms
Ratio of Bad Indexing vs Good Indexing: 19.1172x
```
Clearly with duplicate indexing @performance is ~20x worse than the non-duplicate indexing.
## Environment
Please copy and paste the output from our
[environment collection script](https://raw.githubusercontent.com/pytorch/pytorch/master/torch/utils/collect_env.py)
(or fill out the checklist below manually).
You can get the script and run it with:
```
Collecting environment information...
PyTorch version: 1.5.0
Is debug build: No
CUDA used to build PyTorch: 10.2
OS: Ubuntu 18.04.4 LTS
GCC version: (Ubuntu 8.4.0-1ubuntu1~18.04) 8.4.0
CMake version: version 3.17.3
Python version: 3.7
Is CUDA available: Yes
CUDA runtime version: 10.1.243
GPU models and configuration:
GPU 0: TITAN RTX
GPU 1: TITAN RTX
Nvidia driver version: 440.100
cuDNN version: /usr/lib/x86_64-linux-gnu/libcudnn.so.7.6.5
Versions of relevant libraries:
[pip3] numpy==1.16.6
[conda] blas 1.0 mkl
[conda] cudatoolkit 10.2.89 hfd86e86_1
[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] numpy 1.18.1 py37h4f9e942_0
[conda] numpy-base 1.18.1 py37hde5b4d6_1
[conda] numpydoc 0.9.2 py_0
[conda] pytorch 1.5.0 py3.7_cuda10.2.89_cudnn7.6.5_0 pytorch
[conda] torchvision 0.6.0 py37_cu102 pytorch
[conda] torchviz 0.0.1 pypi_0 pypi
```
@ngimel @mcarilli @csarofeen @ptrblck
cc @ezyang @gchanan @zou3519 @bdhirsh @jbschlosser @anjali411 @VitalyFedyunin @ngimel @heitorschueroff
|
module: performance,triaged,has workaround
|
low
|
Critical
|
653,667,071 |
pytorch
|
Pytorch.js would be nice!
|
## 🚀 Feature
Tensorflow.js allows tensorflow users to run models in a browser. They can share models with their friends, and run realtime inference easily. While ONNX.js exists, I'm finding it doesn't really have a community or any learning resources really except for one youtube video. The official recommendation is to convert pytorch to onnx to tensorflow to tensorflow.js
It would be nice if we could have just a pytorch.js to convert to.
## Motivation
I would love to create ML applications that people can use just by navigating to a URL, instead of having to install dependencies.
## Pitch
I want to be able to use Pytorch models in the browser
## Alternatives
ONNX.js, I found it to be too confusing, and I couldn't figure out how to make it work with object detection.
|
feature,triaged
|
low
|
Minor
|
653,748,449 |
TypeScript
|
IntelliSense for generic Promise function shows truncated return type
|
*TS Template added by @mjbvz*
**TypeScript Version**: 4.0.0-dev.20200702
**Search Terms**
- hover / quick info
- truncation
---
<!-- ⚠️⚠️ Do Not Delete This! bug_report_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: https://code.visualstudio.com/insiders/ -->
<!-- Use Help > Report Issue to prefill these. -->
- VSCode Version: 1.46.1 (user setup)
- OS Version: Windows_NT x64 10.0.18363
Steps to Reproduce:
**App.ts:**
```typescript
import { Repro} from "./Repro";
function test()
{
Repro.repro<number>("test");
}
```
**Repro.ts:**
```typescript
export namespace Repro
{
export interface Arg { argName: string, argValue: any }
// export async function repro<T>(AbcdefghijklmnopqrstuvwxyzAbcdefghijklmnopqrstuvwxyzAbcdefghijklmnopqrstuvwxyzAbcdefghijklmnopqrstuvwxyzAbcdefghijkl: string, ...methodArgs: Arg[]): Promise<T> // Working
export async function repro<T>(AbcdefghijklmnopqrstuvwxyzAbcdefghijklmnopqrstuvwxyzAbcdefghijklmnopqrstuvwxyzAbcdefghijklmnopqrstuvwxyzAbcdefghijklX: string, ...methodArgs: Arg[]): Promise<T> // Broken
{
return (new Promise<T>(null));
}
}
```
**Problem:**
In App.ts, hover over the `.repro` function. The IntelliSense shows `repro` as having a return type of `Promise<...>`. This is incorrect: it should be `Promise<number>`.
Now, in Repro.ts edit the `repro` function by remove the trailing `X` in the first parameter name. Return to App.ts and note that the IntelliSense for `repro` is now correct (the function return type is shown as `Promise<number>`).
While at first glance this may appear to be an edge case due to the long identifier name, it is not. I ran into this problem with "normal length" parameters, but the example above was the simplest repro I could find. Here's the signature of the function where I originally saw the problem:
```typescript
export async function postForkAsync<T>(destinationInstance: string, methodName: string, methodVersion: number, errorTemplate: string = null, resultTimeoutInMs: number = -1, ...methodArgs: PostMethodArg[]): Promise<T>
```
<!-- Launch with `code --disable-extensions` to check. -->
Does this issue occur when all extensions are disabled?: Yes
|
Suggestion,Awaiting More Feedback
|
low
|
Critical
|
653,750,933 |
excalidraw
|
Add ability to add a point on either side of a linear element
|
I thought this was already in place, but I can't make it work:
I expect that if I select a "starting" point, then press "alt", I should be able to "prepend" points to the line. But it seems that I can only add points to the end.
<img width="401" alt="Screen Shot 2020-07-08 at 21 19 07" src="https://user-images.githubusercontent.com/1473433/86996547-d0a95180-c160-11ea-9342-365defb3c57f.png">
|
enhancement
|
low
|
Major
|
653,755,531 |
excalidraw
|
Double click should pass through selected forefront objects for editing
|
Not 100% sure but it feels like this should be changed.
Steps to repro:
1. Create a text
2. Create a rectangle over it
3. Deselect all
4. Double click on the text
5. You will enter edit mode for the text (good)
6. Now select the rectangle
7. Again double click on the text
8. This time you will create a new text element
Step 4 and what I would expect at step 7:
<img width="337" alt="Screen Shot 2020-07-08 at 21 30 28" src="https://user-images.githubusercontent.com/1473433/86997149-419d3900-c162-11ea-9a1d-c3bb3033737c.png">
Current step 8:
<img width="362" alt="Screen Shot 2020-07-08 at 21 31 57" src="https://user-images.githubusercontent.com/1473433/86997234-76a98b80-c162-11ea-9d4c-75d4e520fc76.png">
A different solution that would also work "for me" (in the situation I was running into) is to consider "transparent" shapes as really only occupying their border (and therefore their z index wouldn't matter).
|
bug
|
low
|
Minor
|
653,763,912 |
TypeScript
|
Consider adding `primitive` as a union of all primitive types
|
## Search Terms
- primitive type
- primitive type union
## Suggestion
`primitive` would be a union of all primitive types, e.g.:
```ts
type primitive = string | number | bigint | boolean | symbol | null | undefined; // unknown & not object
```
The advantage of doing this as a **TypeScript** built‑in rather than defining it in user space is that when a new primitive type gets added to **ECMAScript** (e.g.: `bigdecimal`), it will be supported automatically without needing to update the type definitions of the library.
## Use Cases
When describing an API that takes all‑but‑one primitive type, it’s useful to be able to do `Exclude<primitive, symbol>` or `Exclude<primitive, number | bigint>` without needing <https://github.com/microsoft/TypeScript/pull/29317>.
## Examples
<dl>
<dt>From <a href="https://github.com/DefinitelyTyped/DefinitelyTyped/pull/44805">DefinitelyTyped/DefinitelyTyped#44805</a>:</dt>
<dd>
```ts
/**
* @see https://tc39.es/ecma262/#sec-createhtml
*/
export function CreateHTML(
string: object | Exclude<primitive, string>, // unknown & not string
tag: string,
attribute: string,
value?: object | Exclude<primitive, string>, // unknown & not string
): string;
/**
* @throws {TypeError} If `x` or `y` is a `number` or `bigint` or they're different types.
* @see https://tc39.es/ecma262/#sec-samevaluenonnumeric
*/
export function SameValueNonNumeric(
x: object | Exclude<primitive, number | bigint>, // unknown & not (number | bigint)
y: object | Exclude<primitive, number | bigint>, // unknown & not (number | bigint)
): boolean;
```
</dd>
</dl>
## 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).
## Related
- <https://github.com/microsoft/TypeScript/issues/28131>
- <https://github.com/microsoft/TypeScript/pull/29317>
- <https://github.com/microsoft/TypeScript/issues/39498>
|
Suggestion,Awaiting More Feedback
|
low
|
Critical
|
653,768,809 |
scrcpy
|
[feature request] Add option to clamp events coordinates inside the device screen (swipe from left/right edge)
|
- [x] I have checked that a similar [feature request](https://github.com/Genymobile/scrcpy/issues?q=is%3Aopen+is%3Aissue+label%3A%22feature+request%22) does not already exist.
**Is your feature request related to a problem? Please describe.**
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
I am using Samsung Note 10 and I setup my phone to remove Navigation Bar and using One Hand Operation+ for swipe gesture from left/right edge of the screen for navigation.
Like swipe left from right edge to back, swipe left + down from right edge for home.
For that setup, currently I increase horizontal of the window to add black padding zone
But the mouse drag from black padding zone is not recognized.
**Describe the solution you'd like**
A clear and concise description of what you want to happen.
- Accept swipe started from black padding zone that ended inside screen zone
I am using follow coordinate:
^
|
y
x ->
So when a swipe start from padding zone, we would wait only until the mouse click and being dragged into the screen as filter condition of mistake.
When condition meet, the gesture would start from same Y of the coordinate but we would rewrite the start X coordinator to be inside of screen
The end gesture would be where mouse ended the drag
**Describe alternatives you've considered**
A clear and concise description of any alternative solutions or features you've considered.
**Additional context**
Add any other context or screenshots about the feature request here.
|
feature request
|
low
|
Major
|
653,799,333 |
PowerToys
|
[FancyZones] Dragging to edge of screen could bring up the FancyZones overlay, instead of default Windows snap overlay
|
# Summary of the new feature/enhancement
Currently when dragging a window to an edge of a monitor, the window will snap to Windows' default snap positions.
It would be pretty cool if FancyZones could override this interaction to snap to a FancyZone
|
Idea-Enhancement,Product-FancyZones
|
low
|
Major
|
653,827,948 |
pytorch
|
The new version of the libtorch become slow
|
My previous program based on libtorch 1.0.0 Slow down 100ms after the upgrade. And then I find that the same code ran libtorch1.0.0 would be faster than 1.3.0-1.5.1. That's why? 1.1.0-1.2.0 I have not test.
cc @suo @gmagogsfm @yf225 @glaringlee @VitalyFedyunin @ngimel
|
module: performance,oncall: jit,module: cpp,triaged
|
medium
|
Major
|
653,836,248 |
pytorch
|
Failure to build on Power9 due to FXDIV_SOURCE_DIR not being set
|
## 🐛 Bug
The build (actually the cmake configure) on Power9 architectures fails with a strange error:
```
CMake Error at cmake/Codegen.cmake:207 (add_custom_target):
add_custom_target cannot create target "ATEN_CPU_FILES_GEN_TARGET" because
another target with the same name already exists. The existing target is a
custom target created in source directory
"/dev/shm/s3248973-EasyBuild/PyTorch/1.5.1/fosscuda-2019b-Python-3.7.4/pytorch-1.5.1/caffe2".
See documentation for policy CMP0002 for more details.
Call Stack (most recent call first):
caffe2/CMakeLists.txt:3 (include)
CMake Error at cmake/Codegen.cmake:208 (add_custom_target):
add_custom_target cannot create target "ATEN_CUDA_FILES_GEN_TARGET" because
another target with the same name already exists. The existing target is a
custom target created in source directory
"/dev/shm/s3248973-EasyBuild/PyTorch/1.5.1/fosscuda-2019b-Python-3.7.4/pytorch-1.5.1/caffe2".
See documentation for policy CMP0002 for more details.
Call Stack (most recent call first):
caffe2/CMakeLists.txt:3 (include)
```
## To Reproduce
Steps to reproduce the behavior:
1. `python setup.py install` on Power9 with the 1.5.* releases and current master branch
## Expected behavior
Build works
## Additional context
I used `cmake --trace` to narrow this down:
- `pytorch-1.5.1/CMakeLists.txt(564): add_subdirectory(caffe2 )` starts running the caffe2 CMakeLists
- `pytorch-1.5.1/caffe2/CMakeLists.txt(61): ADD_SUBDIRECTORY(${FXDIV_SOURCE_DIR} ${CMAKE_BINARY_DIR}/FXdiv )` starts running the same CMakeLists again, see https://github.com/pytorch/pytorch/blob/a8bc7545d5998c5e3e0c07f01f8c816cd381904f/cmake/Dependencies.cmake#L256 and https://github.com/pytorch/pytorch/blob/a6b703cc89e57273ab943e0ac38eb09b7e6ecd14/caffe2/CMakeLists.txt#L94-L96
Having that I found that in `cmake/Dependencies.cmake` the variable `FXDIV_SOURCE_DIR` is set only if *NPACK is build, which is not the case on Power. Using this unset variable seems to include the current folder again leading to the problem
cc @malfet
|
module: build,triaged,module: POWER
|
low
|
Critical
|
653,846,063 |
go
|
encoding/json: add Encoder.EncodeToken method
|
Currently there is a way to decode JSON token-by-token, but the JSON package does not support encoding tokens that way. It has been stated that just writing the tokens yourself is straightforward enough to do (citation needed), but it's not actually _that_ easy to do.
Here's some code that streams JSON from input to output: https://play.golang.org/p/H6Xl_twRIyC. It's at least 50 lines of code to do the basic streaming functionality, it's easy to get wrong by forgetting to put a colon or a comma in the right place, and if you want indentation or HTML escaping, you have to do it yourself.
I propose that we add an `EncodeToken` method to `json.Encoder`:
```
// EncodeToken encodes a single JSON token to the encoder. The order of
// tokens must result in valid JSON syntax (as returned by the
// Decoder.Token method). That is, delimiters [ ] { } must be properly
// nested and matched, and object keys must be strings (but see the note
// on using Encode below).
//
// EncodeToken returns an error if the token cannot be encoded or is
// misplaced, for example because there's a delimiter mismatch, a
// non-string token is passed where an object key is expected, a string
// contains invalid UTF-8 or a Number isn’t formatted correctly.
//
// Note that it’s OK to mix calls to EncodeToken with calls to Encode.
// In particular, it’s also valid to call Encode when an object key is
// expected - in this case, the value is subject to the same conversion
// rules as for map keys.
func (enc *Encoder) EncodeToken(tok Token) error
```
There would be no way to produce syntactically invalid output (other than truncating the output by not completing the encoded value). The Encoder would keep track of the current state in a stack internally.
An example: if we wanted to stream a large array of JSON objects, we could do:
```
enc.EncodeToken(json.Delim('['))
for {
enc.Encode(arrayItem)
}
enc.EncodeToken(json.Delim(']'))
```
A slightly more comprehensive example is [here](https://play.golang.org/p/QF65gjFapux)
The code to stream a set of JSON values would become considerably simpler: https://play.golang.org/p/Wec5wepCYbE
## Completeness validation
It might be useful to provide callers with a final check that their encoded value is in fact complete (that is, no final closing tokens have been omitted). To do that, the following method could be provided:
```
// ClosingToken returns the token necessary to close the current encoding object or array.
// It returns either Delim('}'), Delim(']'), or nil if there is no current object or array.
func (enc *Encoder) ClosingToken() Token
```
This method makes it straightforward to check that the object is complete (`if env.ClosingToken == nil`), but also could be used for sanity checks when developing, better errors, or to implement a Flush method that automatically closes all open objects and arrays:
```
for enc.ClosingToken() != nil {
enc.EncodeToken(enc.ClosingToken())
}
```
## Discussion
As with the current `Decoder.Token` implementation, `Encoder.EncodeToken` will not be particularly efficient, as it requires passing strings and numbers as strings inside an interface value. I think that this can be considered an orthogonal issue: a solution to the `Decoder.Token` garbage issue may also be applied to `Decoder.Token`. The symmetry between `Decoder.Token` and `Encoder.EncodeToken` is a useful property which makes this API easier to explain and use, and shouldn’t be discarded lightly, in my view.
Once `Encoder.Encode` is invoked, there’s no way to return to streaming mode, for example to encode a large array within a `MarshalJSON` method. This is a limitation of the `json.Marshaler` interface, and could potentially be addressed by a fix in that area.
It would _almost_ be possible to implement this without adding a method at all by changing the `Encode` method to recognize `Delim`. However, it’s currently valid to pass a `Delim` value to encode (it encodes as a number), so this would break backward compatibility. Also, it’s arguably cleaner to have a separation between the low level token API and the higher level Encode API.
## Related proposals
https://github.com/golang/go/issues/33714
|
Proposal,Proposal-Accepted
|
medium
|
Critical
|
653,849,793 |
go
|
proposal: encoding/json: garbage-free reading of tokens
|
As @bradfitz noted in the reviews of the original API, the `Decoder.ReadToken` API [is a garbage factory](https://go-review.googlesource.com/c/go/+/11651/2/src/encoding/json/stream.go#279). Although, as @rsc noted at the time, "a clean API ... is more important here. I expect people to use it to get to the position they want in the stream and then call Decode", the inefficiency is a problem in practice for anyone that wishes to use the encoding/json tokenizer as a basis for some other kind of decoder.
Dave Cheney's ["Building a high performance JSON parser"](https://dave.cheney.net/high-performance-json.html) details some of the issues involved. He comes to the conclusion that the interface-based nature of [json.Token](https://golang.org/pkg/encoding/json/#Token) is a fundamental obstacle. I like the current interface-based API, but it does indeed make it impossible to return arbitrary tokens without creating garbage. Dave suggests [a new `Scanner` API](https://dave.cheney.net/high-performance-json.html#_scanning), somewhat more complex, that is also not backwardly compatible with the current API in encoding/json.
I propose instead that the following method be added to the `encoding/json` package:
```
// TokenBytes is like Token, except that for strings and numbers, it returns
// a static Token value with the actual data payload in the []byte parameter,
// which is only valid until the next call to Token or TokenBytes or Decode.
// For strings, the returned Token will be ""; for a number, it will be
// Number("0"); for all other kinds of token, the Token will be returned as by
// Token method and the []byte value will be nil.
//
// This is more efficient than using Token because it avoids the
// allocations required by that API.
func (dec *Decoder) TokenBytes() (Token, []byte, error)
```
`Token` can be implemented in terms of `TokenBytes` as follows:
```
func (dec *Decoder) Token() (Token, error) {
tok, data, err := dec.TokenBytes()
if err != nil || data == nil {
return tok, err
}
switch tok {
case "":
return string(data)
case Number(0):
return Number(data)
}
panic("unreachable")
}
```
## Discussion
This proposal relies on the observation that the `Decoder.Token` API only generates garbage for two kinds of tokens: numbers and strings. For all other token types, no garbage need be generated, as small numbers (`json.Delim` and bool) do not incur an allocation when boxed in an interface.
It maintains the current API as-is. Users can opt-in to the new API if they require efficiency at some risk of incorrectness (the caller could hold onto the data slice after the next call to Decode). The cognitive overhead of `TokenBytes` is arguably low because of its similarity to the existing API.
If [this proposal](https://github.com/golang/go/issues/40127) is accepted, an `Encoder.EncodeTokenBytes` could easily be added to provide garbage-free streaming JSON generation too.
|
Proposal,Proposal-Hold
|
medium
|
Critical
|
653,856,217 |
youtube-dl
|
Reverse playlist and then start from particular number
|
<!--
######################################################################
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:
- Look through the README (http://yt-dl.org/readme) and FAQ (http://yt-dl.org/faq) for similar questions
- Search the bugtracker for similar questions: http://yt-dl.org/search-issues
- Finally, put x into all relevant boxes (like this [x])
-->
- [x] I'm asking a question
- [x] I've looked through the README and FAQ for similar questions
- [x] I've searched the bugtracker for similar questions including closed ones
## Question
<!--
Ask your question in an arbitrary form. Please make sure it's worded well enough to be understood, see https://github.com/ytdl-org/youtube-dl#is-the-description-of-the-issue-itself-sufficient.
-->
WRITE QUESTION HERE
Want to download all uploaded videos of a channel in reverse order with the renaming of the filename adding a prefix number associated video number, i.e., first video uploaded should have 01 in front of the filename. I used the following command.
`youtube-dl --playlist-reverse -o "SubFolder/%(playlist_index)s-%(title)s.%(ext)s" https://www.youtube.com/playlist?list=AbcdEf1GhIjK2lmn3OpqRs4t -f "bestvideo[height<=720][ext=mp4]+bestaudio/[height <=? 720]"`
This works perfectly.
But I had to stop, unfortunately in the middle, say 6 videos downloaded. Thus, want to start from 7th. I used the following command.
`youtube-dl --playlist-reverse -o "SubFolder/%(playlist_index)s-%(title)s.%(ext)s" https://www.youtube.com/playlist?list=AbcdEf1GhIjK2lmn3OpqRs4t -f "bestvideo[height<=720][ext=mp4]+bestaudio/[height <=? 720]" --playlist-start 7`
Now, what it does is, 1st it removes the top 6 videos without reversing and then reverse the playlist (probably) and starts downloading from 1st video of the channel and the file name starts with 07 along with video title.
What I actually expected/wanted: a command that reverses the playlist first and then starts from --playlist-start.
Probably similar to [this issue](https://github.com/ytdl-org/youtube-dl/issues/6845) but that is opened from Sept 2015 :(.
|
question
|
low
|
Critical
|
653,863,747 |
ant-design
|
Add swipe support to `Tabs` to switch between `TabPane`s
|
- [x] I have searched the [issues](https://github.com/ant-design/ant-design/issues) of this repository and believe that this is not a duplicate.
### What problem does this feature solve?
When using the component `Tabs`, it would be great for users on mobile to be able to switch between `TabPane`s by a swipe gesture like on `ant-design-mobile`.
I am using ant-design for a web application which should work equally good on desktop and on mobile (which is why we chose `ant-design` instead of `ant-design-mobile`).
### What does the proposed API look like?
Add a `swipeable` prop to `Tabs` to enable switching tabs with a swipe gesture.
<!-- generated by ant-design-issue-helper. DO NOT REMOVE -->
|
💡 Feature Request,Inactive
|
low
|
Major
|
653,868,293 |
TypeScript
|
Head comments are removed in some cases for next `import` line
|
---
name: Bug
about: Create a report to help us improve TypeScript
title: 'Head comments are removed in some cases'
labels: ''
assignees: ''
---
**TypeScript Version:** 3.7.x-dev.201xxxxx
There is a repo, the same result for 3.6 and latest next 4.0...
**Search Terms:**
**Code**
```ts
// @some-comment
import { A } from "./mod";
const a: A = {};
```
Repro repo: https://github.com/whitecolor/ts-remove-comment-import
**Expected behavior:**
Comments is not removed
```js
"use strict";
// @some-comment
Object.defineProperty(exports, "__esModule", { value: true });
var a = {};
```
**Actual behavior:**
Head comments are removed in some cases for next `import` line. Produced output:
```js
"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
var a = {};
```
**Playground Link:** <!-- A link to a TypeScript Playground "Share" link which demonstrates this behavior -->
No.
**Related Issues:** <!-- Did you find other bugs that looked similar? -->
Yes. There were a few alike issues before.
|
Bug
|
low
|
Critical
|
653,926,734 |
go
|
runtime: darwin using -buildmode=c-shared so got runtime: unexpected return pc for runtime.sigpanic called from 0x41e745a
|
<!--
Please answer these questions before submitting your issue. Thanks!
For questions please use one of our forums: https://github.com/golang/go/wiki/Questions
-->
### What version of Go are you using (`go version`)?
<pre>
$ go version
go version go1.14.4 darwin/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=""
GOARCH="amd64"
GOBIN=""
GOCACHE="/Users/wangzhizhao/Library/Caches/go-build"
GOENV="/Users/wangzhizhao/Library/Application Support/go/env"
GOEXE=""
GOFLAGS=""
GOHOSTARCH="amd64"
GOHOSTOS="darwin"
GOINSECURE=""
GONOPROXY="*.byted.org,*.everphoto.cn"
GONOSUMDB="*.byted.org,*.everphoto.cn"
GOOS="darwin"
GOPATH="/Users/wangzhizhao/go"
GOPRIVATE="*.byted.org,*.everphoto.cn"
GOPROXY="https://go-mod-proxy.byted.org,https://goproxy.cn,https://proxy.golang.org,direct"
GOROOT="/usr/local/go"
GOSUMDB="sum.golang.google.cn"
GOTMPDIR=""
GOTOOLDIR="/usr/local/go/pkg/tool/darwin_amd64"
GCCGO="gccgo"
AR="ar"
CC="clang"
CXX="clang++"
CGO_ENABLED="1"
GOMOD="/Users/wangzhizhao/go/src/goscripts/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 -fno-caret-diagnostics -Qunused-arguments -fmessage-length=0 -fdebug-prefix-map=/var/folders/c5/gc9h916j1578f82zvg1hgxf40000gp/T/go-build852002820=/tmp/go-build -gno-record-gcc-switches -fno-common"
</pre></details>
<details><summary><code>uname -a</code> Output</summary><br><pre>
$uname -a
Darwin wangzhizhaodeMacBook-Pro.local 18.7.0 Darwin Kernel Version 18.7.0: Mon Apr 27 20:09:39 PDT 2020; root:xnu-4903.278.35~1/RELEASE_X86_64 x86_64
</pre></details>
### What did you do?
<!--
If possible, provide a recipe for reproducing the error.
A complete runnable program is good.
A link on play.golang.org is best.
-->
1. source code is the following two files:
verify.go
```
package main
// #include <stdlib.h>
import "C"
import (
"encoding/hex"
"strings"
"verify/sec"
)
//export Verify
func Verify(serviceNameC *C.char, keyC *C.char) *C.char {
serviceName := C.GoString(serviceNameC)
key := []byte(C.GoString(keyC))
sec.InitWithKey(key)
res := "res"
return C.CString(strings.Join([]string{serviceName, res, hex.EncodeToString(key)}, "-"))
}
func main() {
}
```
sec/key.go
```
package sec
import (
"encoding/hex"
"fmt"
)
var CryptoKey []byte
func InitWithKey(key []byte) {
CryptoKey = key
fmt.Println(hex.EncodeToString(key))
}
```
2. go build genrate c-shared .so
`go build -buildmode=c-shared -a -o libverify.so verify.go`
3. using libverify.so
```
package main
// #cgo LDFLAGS: -lverify
// #include </usr/local/lib/verify.h>
import "C"
import (
"fmt"
"unsafe"
)
func freeCString(cString *C.char) {
C.free(unsafe.Pointer(cString))
}
func Init(serviceName string) {
cServiceName := C.CString(serviceName)
defer freeCString(cServiceName)
key := C.CString("test")
defer freeCString(key)
cRes := C.Verify(cServiceName, key)
defer freeCString(cRes)
fmt.Println(C.GoString(cRes))
}
func main() {
Init("my-service")
}
```
### What did you expect to see?
no panic, and print out a string.
### What did you see instead?
```
fatal error: unexpected signal during runtime execution
[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x41e745a]
goroutine 1 [running, locked to thread]:
runtime.throw(0x40d653d, 0x2a)
/usr/local/go/src/runtime/panic.go:1116 +0x72 fp=0xc0000439b0 sp=0xc000043980 pc=0x4030dd2
runtime: unexpected return pc for runtime.sigpanic called from 0x41e745a
stack: frame={sp:0xc0000439b0, fp:0xc0000439e0} stack=[0xc000043000,0xc000044000)
000000c0000438b0: 0000000000000000 0000000000000000
000000c0000438c0: 0000000000000000 0000000000000000
000000c0000438d0: 0000000000000000 0000000000000000
000000c0000438e0: 0000000000000000 0000000000000000
000000c0000438f0: 0000000000000000 0000000000000000
000000c000043900: 0000000000000000 0000000000000000
000000c000043910: 0000000000000000 0000000000000000
000000c000043920: 0000000000000000 0000000000000000
000000c000043930: 0000000000000000 0000000000000000
000000c000043940: 0000000004030f97 <runtime.fatalthrow+87> 000000c000043950
000000c000043950: 000000000405a860 <runtime.fatalthrow.func1+0> 000000c000000180
000000c000043960: 0000000004030dd2 <runtime.throw+114> 000000c000043980
000000c000043970: 000000c0000439a0 0000000004030dd2 <runtime.throw+114>
000000c000043980: 000000c000043988 000000000405a7e0 <runtime.throw.func1+0>
000000c000043990: 00000000040d653d 000000000000002a
000000c0000439a0: 000000c0000439d0 0000000004044f2a <runtime.sigpanic+1130>
000000c0000439b0: <00000000040d653d 000000000000002a
000000c0000439c0: 0000000000000000 0000000000000000
000000c0000439d0: 000000c000043aa8 !00000000041e745a
000000c0000439e0: >0000000000000000 0000000000000000
000000c0000439f0: 0000000000000000 0000000000000000
000000c000043a00: 0000000000000000 0000000000000000
000000c000043a10: 0000000000000000 0000000000000000
000000c000043a20: 000802aaaa00aaaa 0000000000000034
000000c000043a30: 0000000000000020 0000000000000000
000000c000043a40: 0000000000000000 0000000000000000
000000c000043a50: 0000000000000000 0000000000203000
000000c000043a60: 0000000000203000 0000000000203000
000000c000043a70: 0000000000000000 0000000000000000
000000c000043a80: 0000000000000008 0000000000000000
000000c000043a90: 0000000000000000 000000c000043b38
000000c000043aa0: 000000000400d726 <runtime.mallocgc+790> 000000c000043b48
000000c000043ab0: 00000000041df53e 000000c000012200
000000c000043ac0: 0000000000000010 0000000000000010
000000c000043ad0: 0000000004281340 0000000000000680
runtime.sigpanic()
/usr/local/go/src/runtime/signal_unix.go:679 +0x46a fp=0xc0000439e0 sp=0xc0000439b0 pc=0x4044f2a
goroutine 2 [force gc (idle)]:
runtime.gopark(0x40d7b98, 0x417f670, 0x1411, 0x1)
/usr/local/go/src/runtime/proc.go:304 +0xe0 fp=0xc000030fb0 sp=0xc000030f90 pc=0x40337f0
runtime.goparkunlock(...)
/usr/local/go/src/runtime/proc.go:310
runtime.forcegchelper()
/usr/local/go/src/runtime/proc.go:253 +0xb7 fp=0xc000030fe0 sp=0xc000030fb0 pc=0x40336a7
runtime.goexit()
/usr/local/go/src/runtime/asm_amd64.s:1373 +0x1 fp=0xc000030fe8 sp=0xc000030fe0 pc=0x405da91
created by runtime.init.6
/usr/local/go/src/runtime/proc.go:242 +0x35
goroutine 18 [GC sweep wait]:
runtime.gopark(0x40d7b98, 0x417f7a0, 0x140c, 0x1)
/usr/local/go/src/runtime/proc.go:304 +0xe0 fp=0xc00002c7a8 sp=0xc00002c788 pc=0x40337f0
runtime.goparkunlock(...)
/usr/local/go/src/runtime/proc.go:310
runtime.bgsweep(0xc00008e000)
/usr/local/go/src/runtime/mgcsweep.go:70 +0x9c fp=0xc00002c7d8 sp=0xc00002c7a8 pc=0x4021a4c
runtime.goexit()
/usr/local/go/src/runtime/asm_amd64.s:1373 +0x1 fp=0xc00002c7e0 sp=0xc00002c7d8 pc=0x405da91
created by runtime.gcenable
/usr/local/go/src/runtime/mgc.go:214 +0x5c
goroutine 19 [GC scavenge wait]:
runtime.gopark(0x40d7b98, 0x417f760, 0x140d, 0x1)
/usr/local/go/src/runtime/proc.go:304 +0xe0 fp=0xc00002cf78 sp=0xc00002cf58 pc=0x40337f0
runtime.goparkunlock(...)
/usr/local/go/src/runtime/proc.go:310
runtime.bgscavenge(0xc00008e000)
/usr/local/go/src/runtime/mgcscavenge.go:237 +0xd0 fp=0xc00002cfd8 sp=0xc00002cf78 pc=0x401ffa0
runtime.goexit()
/usr/local/go/src/runtime/asm_amd64.s:1373 +0x1 fp=0xc00002cfe0 sp=0xc00002cfd8 pc=0x405da91
created by runtime.gcenable
/usr/local/go/src/runtime/mgc.go:215 +0x7e
goroutine 3 [finalizer wait]:
runtime.gopark(0x40d7b98, 0x41a8ca8, 0x1410, 0x1)
/usr/local/go/src/runtime/proc.go:304 +0xe0 fp=0xc000030758 sp=0xc000030738 pc=0x40337f0
runtime.goparkunlock(...)
/usr/local/go/src/runtime/proc.go:310
runtime.runfinq()
/usr/local/go/src/runtime/mfinal.go:175 +0xa3 fp=0xc0000307e0 sp=0xc000030758 pc=0x4017773
runtime.goexit()
/usr/local/go/src/runtime/asm_amd64.s:1373 +0x1 fp=0xc0000307e8 sp=0xc0000307e0 pc=0x405da91
created by runtime.createfing
/usr/local/go/src/runtime/mfinal.go:156 +0x61
exit status 2
```
|
OS-Darwin,NeedsInvestigation,compiler/runtime
|
low
|
Critical
|
653,977,305 |
ant-design
|
分页组件（Pagination）的 pageSize 选择器中能否添加一个“显示全部”的选项
|
- [x] I have searched the [issues](https://github.com/ant-design/ant-design/issues) of this repository and believe that this is not a duplicate.
### What problem does this feature solve?
现在虽然可以设置一个例如 1000 条/页 这样比较大的页数，但是实际情况下需要用户自己去理解一下当总条目数小于 1000 时，选这个“1000 条/页”就是展示全部的效果，而且有些业务场景我们也无法预知这个总条目会不会超过 1000 条。能不能做到当这个 Pagination 能够获取到总条目数（total）的情况下，允许提供一个 API 使 pageSize 的下拉框中出现一个“展示全部”的选项。
### What does the proposed API look like?
```tsx
<Pagination showSizeChanger showTotalSizeOption total={865} />
```
<!-- generated by ant-design-issue-helper. DO NOT REMOVE -->
|
Inactive
|
low
|
Minor
|
653,983,826 |
material-ui
|
[material-ui][Autocomplete] Type issue once wrapped with styled()
|
I'm trying to use `Autocomplete` with `styled-components` in typescript environment. But when using the component, linter start to complain about styled autocomplete component. I checked for this issue but not sure if it is duplicate. Could not find something related?
<!-- 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 😯
```
No overload matches this call.
Overload 2 of 2, '(props: StyledComponentPropsWithAs<(<T, Multiple extends boolean | undefined = undefined, DisableClearable extends boolean | undefined = undefined, FreeSolo extends boolean | undefined = undefined>(props: AutocompleteProps<T, Multiple, DisableClearable, FreeSolo>) => Element), DefaultTheme, {}, never>): ReactElement<...>', gave the following error.
Type '(_event: React.ChangeEvent<{}>, value: OptionData | null) => void | undefined' is not assignable to type '(event: ChangeEvent<{}>, value: unknown, reason: AutocompleteChangeReason, details?: AutocompleteChangeDetails<unknown> | undefined) => void'.
Overload 2 of 2, '(props: StyledComponentPropsWithAs<(<T, Multiple extends boolean | undefined = undefined, DisableClearable extends boolean | undefined = undefined, FreeSolo extends boolean | undefined = undefined>(props: AutocompleteProps<T, Multiple, DisableClearable, FreeSolo>) => Element), DefaultTheme, {}, never>): ReactElement<...>', gave the following error.
Type '(data: OptionData) => string' is not assignable to type '(option: unknown) => string'.
Overload 2 of 2, '(props: StyledComponentPropsWithAs<(<T, Multiple extends boolean | undefined = undefined, DisableClearable extends boolean | undefined = undefined, FreeSolo extends boolean | undefined = undefined>(props: AutocompleteProps<T, Multiple, DisableClearable, FreeSolo>) => Element), DefaultTheme, {}, never>): ReactElement<...>', gave the following error.
Type '(option: OptionData) => boolean' is not assignable to type '(option: unknown, value: unknown) => boolean'. TS2769
```
```
62 | return (
63 | <StyledFormControl>
> 64 | <StyledAutocomplete
| ^
65 | id={id}
66 | options={options}
67 | // classes={classes}
```
I started to add one prop by another then i realised the following:
```tsx
<StyledAutocomplete
id={id}
options={options} // type of OptionData[]
onChange={(_event: React.ChangeEvent<{}>, value) => // <<== if i set value: OptionData | null linter will complain
onChange && onChange(value as OptionData | null)
}
value={currentValue}
getOptionLabel={data => (data as OptionData).label} // <<== if i set data: OptionData | null linter will complain
getOptionSelected={option => // <<== if i set option: OptionData | null linter will complain
(option as OptionData).label === currentValue?.label
}
autoHighlight
renderInput={params => (
<TextField
{...params}
error={!!error}
name={name}
label={label}
variant="standard"
/>
)}
/>
```
## Expected Behavior 🤔
Would it be possible to take `T` type of the `options` property and pass it throw all function that working with the same type ?
## Steps to Reproduce 🕹
Simply styled `Autocomplete` with `styled-components`.
```tsx
[...]
const StyledAutocomplete = styled(Autocomplete)`
& .MuiInputBase-root {
height: 4.1rem;
}
& label: {
fontsize: 2rem;
}
`;
[....]
<StyledAutocomplete
id={id}
options={options}
// onChange={(_event: React.ChangeEvent<{}>, value: OptionData | null) => // <<== Linter error
// onChange && onChange(value)
// }
value={currentValue}
// getOptionLabel={(data: OptionData) => data.label} // <<== Linter error
// getOptionSelected={(option: OptionData) =>. // <<== Linter error
// option.label === currentValue?.label
// }
autoHighlight
renderInput={params => (
<TextField
{...params}
error={!!error}
name={name}
label={label}
variant="standard"
/>
)}
/>
```
## Context 🔦
As consistency we use `styled-components` as a generic way of styling. We need to use the same here. The styling suggestion works fine using `makeStyles`
## Your Environment 🌎
| Tech | Version |
| ----------- | ------- |
| Material-UI | v4.9.12 |
| React | 16.13.1 |
| Browser | Chrome 83.0.4103.116 |
| TypeScript | 3.8.0 |
| OS | MacOS. |
|
docs,typescript,component: autocomplete,ready to take
|
medium
|
Critical
|
653,999,436 |
react-native
|
iOS can't dismiss Modal!
|
## Description
On iOS the `<Modal>` component cannot be dismissed by dragging down. (gesture is a native behaviour on iOS)
The reason for that being is that the `onDismiss` or `onRequestClose` function is never called. It is only being called on tvOS.
## React Native version:
```
info Fetching system and libraries information...
(node:10129) Warning: Accessing non-existent property 'padLevels' of module exports inside circular dependency
(Use `node --trace-warnings ...` to show where the warning was created)
System:
OS: macOS 10.15.5
CPU: (16) x64 Intel(R) Core(TM) i9-9980HK CPU @ 2.40GHz
Memory: 2.78 GB / 32.00 GB
Shell: 5.7.1 - /bin/zsh
Binaries:
Node: 14.4.0 - /usr/local/bin/node
Yarn: 1.22.4 - /usr/local/bin/yarn
npm: 6.14.5 - /usr/local/bin/npm
Watchman: 4.9.0 - /usr/local/bin/watchman
Managers:
CocoaPods: 1.9.3 - /usr/local/bin/pod
SDKs:
iOS SDK:
Platforms: iOS 13.5, DriverKit 19.0, macOS 10.15, tvOS 13.4, watchOS 6.2
Android SDK:
API Levels: 25, 28, 29
Build Tools: 28.0.3, 29.0.2, 29.0.3, 30.0.0, 30.0.0, 30.0.0
System Images: android-29 | Google APIs Intel x86 Atom, android-29 | Google Play Intel x86 Atom
Android NDK: 21.2.6472646
IDEs:
Android Studio: 4.0 AI-193.6911.18.40.6514223
Xcode: 11.5/11E608c - /usr/bin/xcodebuild
Languages:
Java: 1.8.0_252 - /usr/bin/javac
Python: 2.7.16 - /usr/bin/python
npmPackages:
@react-native-community/cli: Not Found
react: 16.13.1 => 16.13.1
react-native: 0.63.0 => 0.63.0
npmGlobalPackages:
*react-native*: Not Found
```
## Steps To Reproduce
1. Create `<Modal>` component and set `visible` and `onRequestClose` properties
2. Open Modal by setting `visible` to `true`
3. Try to close Modal by dragging down
## Expected Results
Modal should be dragged down and `onDismiss` or `onRequestClose` should be called
## Snack, code example, screenshot, or link to a repository:
```tsx
<Modal
presentationStyle="pageSheet"
visible={true}
onRequestClose={() => console.log("will never be called.")} />
```
|
Platform: iOS,Resolution: PR Submitted,Component: Modal,Never gets stale
|
medium
|
Critical
|
654,039,003 |
godot
|
`drop_data` doesnt fire if `visible = false ; visible = true`was called before
|
<!-- Please search existing issues for potential duplicates before filing yours:
https://github.com/godotengine/godot/issues?q=is%3Aissue
-->
**Godot version:** 3.2.1.stable
<!-- Specify commit hash if using non-official build. -->
**OS/device including version:** Arch Linux
<!-- Specify GPU model, drivers, and the backend (GLES2, GLES3, Vulkan) if graphics-related. -->
**Issue description:**
<!-- What happened, and what was expected. -->
Drag&Drop on a panel with mouse_filter 'Pass' calls the `drop_data` function on the panel's parent as expected. But if the panel's `visible` property was set false and then to true immediately before that, it prevents the `drop_data` call, as if it's mouse_filter was set to 'Stop'. When it's mouse_filter is set to 'Ignore' it works as expected.
**Steps to reproduce:**
Open the minimal reproduction projection and drag the Label with the text 'drag me!' on the red panel with the mouse_filter property set to 'Pass' and its visibility double-toggled on each mouse motion event. The drop event will not fire although it should.
**Minimal reproduction project:**
[drop_data_visibility_changed.zip](https://github.com/godotengine/godot/files/4896741/drop_data_visibility_changed.zip)
<!-- A small Godot project which reproduces the issue. Drag and drop a zip archive to upload it. -->
**workaround:** My workaround is to buffer the visibility in a special variable and set the visible property later.
|
bug,topic:gui
|
low
|
Minor
|
654,057,248 |
pytorch
|
DLPack cannot accept empty CUDA tensor
|
## 🐛 Bug
<!-- A clear and concise description of what the bug is. -->
When converting a DLPack object representing an empty CUDA tensor to a PyTorch tensor, it crashes with "CUDA error: invalid argument".
## To Reproduce
Steps to reproduce the behavior:
```python
# bug.py
import torch as th
from torch.utils import dlpack
y = th.tensor([]).cuda()
y_dlpack = dlpack.to_dlpack(y)
dlpack.from_dlpack(y_dlpack)
```
1. Run the above `bug.py`.
<!-- If you have a code sample, error messages, stack traces, please provide it here as well -->
Error message:
```
Traceback (most recent call last):
File "bug.py", line 6, in <module>
dlpack.from_dlpack(y_dlpack)
RuntimeError: CUDA error: invalid argument
```
## Expected behavior
<!-- A clear and concise description of what you expected to happen. -->
Shall not crash and get a tensor of shape `torch.Size([0])`.
## Environment
```
Collecting environment information...
PyTorch version: 1.5.1
Is debug build: No
CUDA used to build PyTorch: 10.1
OS: Ubuntu 18.04.3 LTS
GCC version: (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
CMake version: version 3.13.3
Python version: 3.7
Is CUDA available: Yes
CUDA runtime version: 10.0.130
GPU models and configuration: GPU 0: Tesla V100-SXM2-16GB
Nvidia driver version: 418.87.01
cuDNN version: /usr/local/cuda-10.1/targets/x86_64-linux/lib/libcudnn.so.7.6.2
Versions of relevant libraries:
[pip3] numpy==1.18.5
[pip3] torch==1.5.1
[pip3] torch-cluster==1.4.5
[pip3] torch-geometric==1.4.2
[pip3] torch-scatter==2.0.4
[pip3] torch-sparse==0.6.0
[pip3] torchtext==0.5.0
[conda] blas 1.0 mkl
[conda] cudatoolkit 10.1.243 h6bb024c_0
[conda] mkl 2020.1 217
[conda] mkl-service 2.3.0 py37he904b0f_0
[conda] mkl_fft 1.1.0 py37h23d657b_0
[conda] mkl_random 1.1.1 py37h0573a6f_0
[conda] numpy 1.18.5 py37ha1c710e_0
[conda] numpy-base 1.18.5 py37hde5b4d6_0
[conda] pytorch 1.5.1 py3.7_cuda10.1.243_cudnn7.6.3_0 pytorch
[conda] torch-cluster 1.4.5 pypi_0 pypi
[conda] torch-geometric 1.4.2 pypi_0 pypi
[conda] torch-scatter 2.0.4 pypi_0 pypi
[conda] torch-sparse 0.6.0 pypi_0 pypi
[conda] torchtext 0.5.0 pypi_0 pypi
```
## Additional context
<!-- Add any other context about the problem here. -->
CPU tensor works fine.
cc @ezyang @gchanan @zou3519 @ngimel
|
module: cuda,triaged
|
low
|
Critical
|
654,058,851 |
flutter
|
CupertinoAlertDialog inconsistent dividers across IOS and Android
|
## Steps to Reproduce
```
showDialog(
context: context,
builder: (BuildContext context) {
return CupertinoAlertDialog(
title: Text('title'),
actions: [
CupertinoDialogAction(child: Text('button')),
CupertinoDialogAction(child: Text('button')),
],
);
},
);
showDialog(
context: context,
builder: (BuildContext context) {
return CupertinoActionSheet(
title: Text('title'),
actions: [
CupertinoDialogAction(child: Text('button')),
CupertinoDialogAction(child: Text('button')),
],
);
},
);
```
**Expected results:** Dividers should look the same. They should be similar brightness as in the example image used here: [Cupertino (iOS-style) widgets](https://flutter.dev/docs/development/ui/widgets/cupertino)
<img src="https://flutter.dev/images/widget-catalog/cupertino-alert-dialog.png" width="300" />
**Actual results:** On iOS CupertinoAlertDialog divider is darker and thicker. In overall the divider is darker than it should be.
| Platform | CupertinoAlertDialog | CupertinoActionSheet |
|:---------------------------------------:|:---------------------------------------:|:---------------------------------------:|
| iOS | <img src="https://i.imgur.com/TErLjZA.jpg" width="300" /> | <img src="https://i.imgur.com/o5zAGjG.jpg" width="300" /> |
| Android | <img src="https://i.imgur.com/twJT7cQ.png" width="300" /> | <img src="https://i.imgur.com/SmC4kTn.png" width="300" /> |
<details>
<summary>flutter doctor -v</summary>
```
[✓] Flutter (Channel stable, v1.17.5, on Mac OS X 10.15.5 19F101, locale en-LV)
• Flutter version 1.17.5 at /Users/martinsandersons/Development/Flutter
• Framework revision 8af6b2f038 (9 days ago), 2020-06-30 12:53:55 -0700
• Engine revision ee76268252
• Dart version 2.8.4
[✓] Android toolchain - develop for Android devices (Android SDK version 29.0.2)
• Android SDK at /Users/martinsandersons/Library/Android/sdk
• Platform android-29, build-tools 29.0.2
• Java binary at: /Users/martinsandersons/Library/Application Support/JetBrains/Toolbox/apps/AndroidStudio/ch-0/193.6514223/Android Studio.app/Contents/jre/jdk/Contents/Home/bin/java
• Java version OpenJDK Runtime Environment (build 1.8.0_242-release-1644-b3-6222593)
• All Android licenses accepted.
[✓] Xcode - develop for iOS and macOS (Xcode 11.5)
• Xcode at /Applications/Xcode.app/Contents/Developer
• Xcode 11.5, Build version 11E608c
• CocoaPods version 1.9.3
[✓] VS Code (version 1.46.1)
• VS Code at /Applications/Visual Studio Code.app/Contents
• Flutter extension version 3.12.1
[✓] Connected device (2 available)
• SM G920F • android-arm64 • Android 7.0 (API 24)
• Martins’s iPhone SE • ios • iOS 13.5.1
```
</details>
|
framework,f: cupertino,a: quality,has reproducible steps,P2,found in release: 3.3,found in release: 3.7,team-design,triaged-design
|
low
|
Major
|
654,099,587 |
rust
|
Add support for `#[doc(since="version")]` attribute
|
To the present day, the _internal_ crates like `core` and `std` can annotate things with `#[stable(since="version")]`. This is extremely useful for users, especially because `rustdoc` is able to write this information in the rendered documentation.
Unfortunately, this kind of feature is not available for _normal_ crates. Take, for instance, [`tokio::runtime::Handle::current`](https://docs.rs/tokio/0.2.21/tokio/runtime/struct.Handle.html): this function is available since version `0.2.8` of the `tokio` crate (not sure about `0.2.7`, that's not the point), but this information is not available at a first glimpse.
IMHO, adding the possibility of annotating functions, structs and traits (and maybe something I missed) with `#[doc(since="version")]` would give more _doc_ power to crates' authors. Moreover, it should be possible to leverage the current implementation of `stable(since="version")` in order to introduce the feature without reinventing the wheel.
|
T-rustdoc,C-feature-request
|
medium
|
Major
|
654,129,012 |
rust
|
Missed optimization: Useless branches are kept when matching pattern
|
EDIT: this is now reasonably optimized: https://rust.godbolt.org/z/nP599MaKM
So it just needs tests: @rustbot label +E-needs-test
---
I tried this code: https://rust.godbolt.org/z/hP3YKe
The code below does not compile because I strip some code, try the godbolt link above:
```rust
const fn doc_comment_kind(s: &str) -> Option<(DocCommentKind, AttrStyle)> {
match s.as_bytes() {
[b'/', b'/', b'/', b'/', ..] => None,
[b'/', b'/', b'/', ..] => Some((DocCommentKind::Line, AttrStyle::Inner)),
[b'/', b'/', b'!', ..] => Some((DocCommentKind::Line, AttrStyle::Outer)),
[b'/', b'*', b'*', b'*', _, ..] => None,
[b'/', b'*', b'*', _, _, ..] => Some((DocCommentKind::Block, AttrStyle::Inner)),
[b'/', b'*', b'!', _, _, ..] => Some((DocCommentKind::Block, AttrStyle::Outer)),
_ => None,
}
}
pub const fn is_line_doc_comment(s: &str) -> bool {
match doc_comment_kind(s) {
Some((DocCommentKind::Line, _)) => true,
_ => false
}
}
```
**I expected to see this happen:**
The generated code of `is_line_doc_comment` should be optimized out useless branches.
And it should be equivalent to this function:
```rust
pub const fn is_line_doc_comment_2(s: &str) -> bool {
match s.as_bytes() {
[b'/', b'/', b'/', b'/', ..] => false,
[b'/', b'/', b'/', ..] => true,
[b'/', b'/', b'!', ..] => true,
_ => false,
}
}
```
*How would the two `is_line_doc_comment` functions be able to generate the same code?*
When `doc_comment_kind` inlined in `is_line_doc_comment`,
the optimizer could see that
* `Some((DocCommentKind::Block, _))` branches are useless and remove it
```rust
const fn doc_comment_kind(s: &str) -> Option<(DocCommentKind, AttrStyle)> {
match s.as_bytes() {
[b'/', b'/', b'/', b'/', ..] => None,
[b'/', b'/', b'/', ..] => Some((DocCommentKind::Line, AttrStyle::Inner)),
[b'/', b'/', b'!', ..] => Some((DocCommentKind::Line, AttrStyle::Outer)),
[b'/', b'*', b'*', b'*', _, ..] => None,
_ => None,
}
}
```
* `is_line_doc_comment` doesn't care about `AttrStyle`,
```rust
const fn doc_comment_kind(s: &str) -> Option<DocCommentKind> {
match s.as_bytes() {
[b'/', b'/', b'/', b'/', ..] => None,
[b'/', b'/', b'/', ..] => Some(DocCommentKind::Line),
[b'/', b'/', b'!', ..] => Some(DocCommentKind::Line),
[b'/', b'*', b'*', b'*', _, ..] => None,
_ => None,
}
}
```
* The first branch overlaps with second branch of `DocCommentKind::Line` branches,
keep this branch.
* Other `None` branches don't overlap, just make them as fallback:
```rust
const fn doc_comment_kind(s: &str) -> Option<DocCommentKind> {
match s.as_bytes() {
[b'/', b'/', b'/', b'/', ..] => None,
[b'/', b'/', b'/', ..] => Some(DocCommentKind::Line),
[b'/', b'/', b'!', ..] => Some(DocCommentKind::Line),
_ => None,
}
}
```
Final result: https://rust.godbolt.org/z/jrPGr7
**Instead, this happened:**
The generated code still keeps some useless branches.
Maybe I am just asking too much from optimizer.
### Meta
`rustc --version --verbose`:
```
rustc 1.46.0-nightly (8aa18cbdc 2020-07-08)
binary: rustc
commit-hash: 8aa18cbdc5d4bc33bd61e2d9a4b643d87f5d21de
commit-date: 2020-07-08
host: x86_64-unknown-linux-gnu
release: 1.46.0-nightly
LLVM version: 10.0
```
|
I-slow,E-needs-test,T-compiler,C-bug
|
low
|
Minor
|
654,168,581 |
pytorch
|
test_autograd failures on Power
|
## 🐛 Bug
I have some tests failing on a Power system. As MKL is not (easily?) available there I'm using the automatic fallback to Eigen/OpenBlas.
The failing tests are:
- test_solve_batched_broadcast_A_cpu
- test_solve_batched_broadcast_b_cpu
- test_solve_batched_cpu
- test_solve_batched_dims_cpu
Those are only 4 out of 1961 which fail
## To Reproduce
Steps to reproduce the behavior:
1. `python setup.py install`
1. `cd tests`
1. `python run_tests.py`
<!-- If you have a code sample, error messages, stack traces, please provide it here as well -->
## Expected behavior
Tests pass
## Environment
PyTorch Version: 1.5.1 and 1.6rc3
OS: Red Hat Enterprise Linux Server release 7.6 (Maipo)
GCC version: (GCC) 8.3.0
CMake version: version 3.15.3
Python version: 3.7
Is CUDA available: N/A
CUDA runtime version: 10.1.243
GPU models and configuration:
GPU 0: Tesla V100-SXM2-32GB
GPU 1: Tesla V100-SXM2-32GB
GPU 2: Tesla V100-SXM2-32GB
GPU 3: Tesla V100-SXM2-32GB
GPU 4: Tesla V100-SXM2-32GB
GPU 5: Tesla V100-SXM2-32GB
Nvidia driver version: 440.64.00
cuDNN version: 7.6.4.38
Versions of relevant libraries:
[pip3] numpy==1.17.3
## Additional context
The failed tests output look like this:
```
RuntimeError: Jacobian mismatch for output 0 with respect to input 0,
numerical:tensor([[ 1.3259e+00, 0.0000e+00, 0.0000e+00, ..., 3.9608e+06,
0.0000e+00, 0.0000e+00],
[ 0.0000e+00, 1.3259e+00, 0.0000e+00, ..., 0.0000e+00,
0.0000e+00, 0.0000e+00],
[ 0.0000e+00, 0.0000e+00, 1.3259e+00, ..., 0.0000e+00,
0.0000e+00, 0.0000e+00],
...,
[ 0.0000e+00, 0.0000e+00, 0.0000e+00, ..., -2.3575e-02,
0.0000e+00, 0.0000e+00],
[ 0.0000e+00, 0.0000e+00, 0.0000e+00, ..., 0.0000e+00,
-2.3575e-02, 0.0000e+00],
[ 0.0000e+00, 0.0000e+00, 0.0000e+00, ..., 0.0000e+00,
0.0000e+00, -2.3575e-02]])
analytical:tensor([[ 1.3259, 0.0000, 0.0000, ..., 0.0000, 0.0000, 0.0000],
[ 0.0000, 1.3259, 0.0000, ..., 0.0000, 0.0000, 0.0000],
[ 0.0000, 0.0000, 1.3259, ..., 0.0000, 0.0000, 0.0000],
...,
[ 0.0000, 0.0000, 0.0000, ..., -0.0236, 0.0000, 0.0000],
[ 0.0000, 0.0000, 0.0000, ..., 0.0000, -0.0236, 0.0000],
[ 0.0000, 0.0000, 0.0000, ..., 0.0000, 0.0000, -0.0236]])
```
cc @ezyang @albanD @zou3519 @gqchen @pearu @nikitaved @soulitzer @ngimel @jianyuh @mruberry @heitorschueroff @walterddr @IvanYashchuk @VitalyFedyunin
|
module: autograd,module: cuda,module: tests,triaged,module: POWER,module: linear algebra
|
low
|
Critical
|
654,169,563 |
rust
|
Missed optimization: Loop with decreasing index does not elide bounds check
|
I tried code very similar to the code below. The expected produced code would perform one bounds check at the start of the function and none inside the loop. `offsets` is first asserted to be valid for the initial index. This should be enough to also find that it is valid for all smaller indices. No loop iteration can increase the index so the index correctness is a loop invariant.
```rust
fn problematic(buf: &mut [u8], offsets: &[u8], mut idx: usize) {
let offsets = &offsets[..=idx];
for b in buf {
*b = idx as u8;
idx = idx.saturating_sub(usize::from(offsets[idx]));
}
}
```
Instead we get this:
<details>
<pre>
playground::problematic:
push rax
cmp r8, -1
je .LBB0_3
mov r9, rsi
lea rsi, [r8 + 1]
cmp r8, rcx
jae .LBB0_2
test r9, r9
je .LBB0_8
xor r10d, r10d
mov rcx, r8
.LBB0_6:
mov byte ptr [rdi + r10], cl
;; The critical jump and cmp we don't want to have, costs ~5-10% loop perf
cmp rcx, r8
ja .LBB0_9
movzx eax, byte ptr [rdx + rcx]
sub rcx, rax
mov eax, 0
cmovb rcx, rax
add r10, 1
cmp r9, r10
jne .LBB0_6
.LBB0_8:
pop rax
ret
.LBB0_9:
lea rdx, [rip + .Lanon.9b054490e6ade753ffe504f874525a87.2]
mov rdi, rcx
call qword ptr [rip + core::panicking::panic_bounds_check@GOTPCREL]
ud2
.LBB0_3:
lea rdi, [rip + .Lanon.9b054490e6ade753ffe504f874525a87.1]
call qword ptr [rip + core::slice::slice_index_overflow_fail@GOTPCREL]
ud2
.LBB0_2:
lea rdx, [rip + .Lanon.9b054490e6ade753ffe504f874525a87.1]
mov rdi, rsi
mov rsi, rcx
call qword ptr [rip + core::slice::slice_index_len_fail@GOTPCREL]
ud2
.Lanon.9b054490e6ade753ffe504f874525a87.0:
.ascii "src/lib.rs"
.Lanon.9b054490e6ade753ffe504f874525a87.1:
.quad .Lanon.9b054490e6ade753ffe504f874525a87.0
.asciz "\n\000\000\000\000\000\000\000\002\000\000\000\024\000\000"
.Lanon.9b054490e6ade753ffe504f874525a87.2:
.quad .Lanon.9b054490e6ade753ffe504f874525a87.0
.asciz "\n\000\000\000\000\000\000\000\005\000\000\000.\000\000"
</pre>
</details>
|
A-LLVM,I-slow,C-enhancement,T-compiler,C-optimization
|
low
|
Minor
|
654,174,989 |
pytorch
|
test_upsampling_not_recompute_scale_factor fails with Eigen/OpenBLAS
|
## 🐛 Bug
I'm getting a test failure of test_upsampling_not_recompute_scale_factor
## To Reproduce
Steps to reproduce the behavior:
1. python setup.py install
1. cd test
1. python run_tests.py
<!-- If you have a code sample, error messages, stack traces, please provide it here as well -->
```
======================================================================
FAIL: test_upsampling_not_recompute_scale_factor (__main__.TestNN)
----------------------------------------------------------------------
Traceback (most recent call last):
File "test_nn.py", line 7199, in test_upsampling_not_recompute_scale_factor
self.assertEqual(out_t, expected_out_t)
File "/tmp/easybuild-tmp/eb-lWK_ii/tmpPLa9_o/lib/python3.7/site-packages/torch/testing/_internal/common_utils.py", line 884, in assertEqual
assertTensorsEqual(x, y)
File "/tmp/easybuild-tmp/eb-lWK_ii/tmpPLa9_o/lib/python3.7/site-packages/torch/testing/_internal/common_utils.py", line 850, in assertTensorsEqual
self.assertLessEqual(max_err, prec, message)
AssertionError: tensor(1.57356e-05, dtype=torch.float32) not less than or equal to 1e-05 :
```
## Expected behavior
Tests pass
## Environment
PyTorch version: 1.5.1
OS: Red Hat Enterprise Linux Server release 7.8 (Maipo)
GCC version: (GCC) 8.3.0
CMake version: version 3.15.3
Python version: 3.7
Is CUDA available: N/A
CUDA runtime version: 10.1.243
cuDNN version: 7.6.4.38
Versions of relevant libraries:
[pip3] numpy==1.17.3
cc @mruberry
|
module: tests,triaged
|
low
|
Critical
|
654,189,748 |
go
|
dev.boringcrypto: crypto/internal/boring: consider splitting up into smaller packages based on math/big requirement
|
As discussed in [golang.org/cl/241442](https://go-review.googlesource.com/c/go/+/241442/1#message-bb399544c64ab073e5adf50eef6a961dc798ab5e), this is a tracking issue to look into breaking `crypto/internal/boring` up into distinct packages, splitting out RSA and ECDSA—which require `math/big`—away from the rest which do not.
/cc @FiloSottile @katiehockman
|
NeedsInvestigation
|
low
|
Minor
|
654,199,429 |
TypeScript
|
Problematic Ternary-Expression Formatting Inside JS String Interpolation
|
*TS Template added by @mjbvz*
**TypeScript Version**: 4.0.0-dev.20200708
**Search Terms**
- format
- template string
---
<!-- ⚠️⚠️ Do Not Delete This! bug_report_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: https://code.visualstudio.com/insiders/ -->
<!-- Use Help > Report Issue to prefill these. -->
- VSCode Version: 1.46.1
- OS Version: Fedora 32, Linux x64 5.6.19-300.fc32.x86_64
Steps to Reproduce:
1. manual breaking of a long line of a ternary expression within template string interpolation `${ condition ? true : false }`
2. running native formatter messes up the indentation for every possible correct format
Code
```JS
var query = `
SELECT
entry.id,
entry.event_type as type,
entry.object, entry.creator,
UNIX_TIMESTAMP(entry.time_created) as timeCreated
${
isActive ?
'FROM entry INNER JOIN active ON entry.id = active.id' :
'FROM entry LEFT JOIN active ON entry.id = active.id WHERE active.id IS NULL'
}
LIMIT ${20 * (page > 1 ? page - 1 : 0)}, 20;`;
```
Screenshots
Before:
<img src="https://user-images.githubusercontent.com/34912861/87024388-536df300-c1d9-11ea-9038-fe95b415e5aa.png" width="650">
After:
<img src="https://user-images.githubusercontent.com/34912861/87024448-6bde0d80-c1d9-11ea-87c7-991250ca42b7.png" width="650">
Before:
<img src="https://user-images.githubusercontent.com/34912861/87024682-ba8ba780-c1d9-11ea-9d5c-448986ada731.png" width="650">
After:
<img src="https://user-images.githubusercontent.com/34912861/87024728-caa38700-c1d9-11ea-81af-d2120fa8301a.png" width="650">
So, the problem specifically seems to be in the way the formatter determines the indentation level to which it normalizes the rest of ternary expression body. The above screenshots show that it ignores the indentation level of the opening brace, and even the lines before it, adjusting itself to the first line indentation level.
<!-- Launch with `code --disable-extensions` to check. -->
Does this issue occur when all extensions are disabled?: Yes
|
Bug,Domain: Formatter
|
low
|
Critical
|
654,222,855 |
storybook
|
Addon-controls: Add custom user-defined controls
|
We are currently documenting our React components in Storybook, and decided to go straight to version 6.0 to make use of the new Args and Controls addons. During that work, we realised that the [choice of controls](https://github.com/storybookjs/storybook/blob/next/addons/controls/README.md#control-annotations) is very limited.
Here are some examples:
- We want users to select from a pre-defined list of colors. The `"select"` control can only display text but we want to preview the color next to it.
- We want users to select from a list of icons. The control should display a list of all icons and clicking on an icon selects it.
- One prop may allow for `string[]`, so we want to allow the user to add strings to the list, or remove existing strings.
Any chance to allow users to specify custom controls in `argTypes` that are not built into `addon-controls`?
Here's how this could look like in React code, where `ColorPicker` is a custom control we'd build:
```tsx
const ColorPicker = ({ value, setValue }) => {
return (
<select value={value} onChange={setValue}>
<option>red</option>
<option>blue</option>
</select>
);
}
export default {
title: "ExampleComponent",
component: ExampleComponent,
argTypes: {
color: {
control: ColorPicker,
},
},
};
```
|
feature request,addon: controls
|
high
|
Critical
|
654,226,270 |
TypeScript
|
Generic Type Alias Overloading
|
<!-- 🚨 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
<!-- List of keywords you searched for before creating this issue. Write them down here so that others can find this suggestion more easily -->
- variadic type alias
- Generic variadic Type Alias
- Generic Overload
- Generic Type Alias Overloading
## Suggestion
<!-- A summary of what you'd like to see added or changed -->
Ability to either:
- overload generic type definitions
_or_
- define variadic type parameters for generic type aliases. (not sure how this would work)
## Use Cases
My specific, immediate need is to create a generic variable number of input type alias for `Intersection`. I could see other use cases though, and imagine others would find it useful.
The closest work around is if I define default type parameters, like:
```
type Intersection<A, B, C = {} , D = {}, E = {}, F = {}> = A & B & C & D & E & F;
// and use like
let foo: Intersection<{ prop1: number}, {prop2: string}>;
// it evaluates to
{ prop1: number } &
{ prop2: number } &
{ } &
{ } &
{ } &
{ }
```
## Examples
With overloads, the above type could be declared like:
```
type Intersection<A , B> = A & B;
type Intersection<A , B , C> = A & B & C;
type Intersection<A , B , C , D> = A & B & C & D;
type Intersection<A , B , C , D , E> = A & B & C & D & E;
type Intersection<A , B , C , D , E , F > = A & B & C & D & E & F;
```
I am not sure of a good syntax for the variadic type, it also seems like it would have a more limited use.
## Checklist
My suggestion meets these guidelines: ( I think )
* [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
|
654,237,939 |
angular
|
[feature] Ability to clone a TemplateRef to multiple outlets at once
|
# 🚀 Ability to clone a TemplateRef to multiple outlets at once
### Relevant Package
I could see this being a useful utility in `@angular/cdk`, but would likely require an update to `@angular/core`. I'm honestly not sure.
### Description
Essentially, I would like to have the ability to take a `TemplateRef` and inject _copies_ of it into multiple template outlets. The use case that made me aware of this limitation was trying to create a `<select>`-style form control, where consumers could put any arbitrary content into an option (not just text, but icons, images, components, etc.). When an option is selected, the Select component would _re_-project that option's content into the Select's template to indicate the currently selected option.
In case that's not clear, here's a simplified/truncated version of the solution that I attempted.
```typescript
@Component({
selector: 'my-select',
template: `
<div class="selected-value">
<ng-container *ngTemplateOutlet="selectedTemplate"></ng-container>
<ng-container *ngIf="!selectedTemplate">{{ placeholder }}</ng-container>
</div>
<div class="options">
<ng-content></ng-content>
</div>
`
})
export class SelectComponent implements AfterContentInit {
@Input() placeholder = "Select an option";
@ContentChildren(OptionComponent) options: QueryList<OptionComponent>;
selectedTemplate: TemplateRef<null>;
ngAfterContentInit() {
this.options.forEach(option => {
option.selected.subscribe(() => {
this.selectedTemplate = option.contentTemplate;
});
});
}
}
@Component({
selector: 'my-option',
template: `
<ng-container *ngTemplateOutlet="contentTemplate"></ng-container>
<ng-template #contentTemplate>
<ng-content></ng-content>
</ng-template>
`
})
export class OptionComponent {
@Output() selected = new EventEmitter<void>();
@ViewChild('contentTemplate') contentTemplate: TemplateRef<null>;
@HostListener('click') onClick() {
this.selected.emit();
}
}
```
And here's an example usage that takes advantage of the template cloning:
```html
<label for="status">Task Status</label>
<my-select id="status" [(ngModel)]="statusValue">
<my-option [value]="Status.Blocked">
<my-icon icon="ban"></my-icon> Blocked
</my-option>
<my-option [value]="Status.Researching">
<my-icon icon="science"></my-icon> Researching
</my-option>
<my-option [value]="Status.InProgress">
<my-icon icon="ellipsis"></my-icon> In Progress
<my-option>
<my-option [value]="Status.Done">
<my-icon icon="check"></my-icon> Done
</my-option>
</my-select>
```
This works as intended, except it appears that the `TemplateRef` can only be embedded into one view container at any given time. So, when an option is selected, it appears in the SelectComponent's template outlet (🎉), but _disappears_ from the OptionComponent's template outlet (😭), leaving a blank spot in the options list.
### Describe the solution you'd like
It would be great if the sample code above just worked as expected out of the box, but I can understand if that's not feasible, or if it's undesirable for some reason I'm not aware of.
I dug through some of CDK's source code, because I knew that its Drag-and-Drop feature was doing some sort of element cloning for the preview and placeholder elements, so I hoped that maybe they were making use of some Angular template/view APIs that I was unfamiliar with. Unfortunately, it looks like that feature is just using some [raw, low-level DOM manipulation](https://github.com/angular/components/blob/fb769ef421f5b73a5d88163864e89e1b834086ec/src/cdk/drag-drop/drag-ref.ts#L1316) to achieve this, which would be way overkill for my use case.
I wonder if it would be possible to add a `clone` method to the `TemplateRef` class, which would return a new `TemplateRef` that could be embedded into an additional template outlet without detaching the original?
### Describe alternatives you've considered
If I specifically wanted to enable icons in my OptionComponent, of course I could add an `@Input() icon: IconName`, and replicate that to my SelectComponent. I could make the OptionComponent's label a simple `@Input label: string` while I'm at it, instead of fiddling with the content projection at all. But I like to keep these types of core UI components as flexible as possible for a wide variety of use cases, so that when a new use case comes up that I hadn't originally anticipated, it can be accommodated without needing to make major updates to the component or [endlessly tack on configuration options](https://youtu.be/0BikjL858OQ?t=324). If there's another alternative or workaround that I haven't thought of, I would love to hear it.
|
feature,area: core,core: content projection,core: dynamic view creation,type: use-case,feature: insufficient votes
|
medium
|
Critical
|
654,324,764 |
TypeScript
|
Variable "name" in VSCode editor(intelli-sense) is showing incorrect information
|
*TS Template added by @mjbvz*
**TypeScript Version**: 4.0.0-dev.20200708
**Search Terms**
- javascript
- lib.dom
- globals
---
<!-- ⚠️⚠️ Do Not Delete This! bug_report_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: https://code.visualstudio.com/insiders/ -->
<!-- Use Help > Report Issue to prefill these. -->
- VSCode Version: 1.46.1
- OS Version: Windows 10
Steps to Reproduce:
1. Add the following javascript code
var name = "Hi";
console.log(name);
name = "Hello";
console.log(name);
2. Hover over(Mouseover) the variable "name" (in any of the console log statements given above).......picture attached.
3. Expected: During mouse/hover over, variable "name" on the console statement should display "name" as a variable.
Actual: During mouse/hover over, variable "name" on the console statement is showing it as constant. (actually, it's not a constant). Because I'm able to modify the value of the variable "name" from "Hi" to "Hello". This explains "name" is implemented as a variable.
<!-- Launch with `code --disable-extensions` to check. -->
Does this issue occur when all extensions are disabled?: Yes
|
Suggestion,Needs Proposal
|
low
|
Critical
|
654,334,523 |
flutter
|
Flutter Android app “Unable to instantiate activity” (Pre-Launch crash report)
|
I build my app as a "release" build (with signing and all) to publish it to PlayStore. I get a lot of crash reports from the "Pre-launch report" within PlayStore (see "actual results"). The build process, however, worked as expected without any failure. Also, the debug build and split-by-abi build is working without any problems or crashes. The deploy to PlayStore also does not cause any problems.
## Steps to Reproduce
This is the folder structure of the Android app (Originally I used another folder structure "app/main/kotlin/com/domain/MainActivity.kt" but with the same result):
[Link to Imgur for package structure](https://i.stack.imgur.com/jTFcD.png)
This is the `gradle.build` which is on the androids folder level:
```
def localProperties = new Properties()
def localPropertiesFile = rootProject.file('local.properties')
if (localPropertiesFile.exists()) {
localPropertiesFile.withReader('UTF-8') {
reader -> localProperties.load(reader)
}
}
def flutterRoot = localProperties.getProperty('flutter.sdk')
logger.info("DEBUG - FLUTTER ROOT: " + flutterRoot)
if (flutterRoot == null) {
throw new GradleException("Flutter SDK not found. Define location with flutter.sdk in the local.properties file.")
}
def flutterVersionCode = localProperties.getProperty('flutter.versionCode')
if (flutterVersionCode == null) {
flutterVersionCode = '1'
}
def flutterVersionName = localProperties.getProperty('flutter.versionName')
if (flutterVersionName == null) {
flutterVersionName = '1.0'
}
apply plugin: 'com.android.application'
apply from: "$flutterRoot/packages/flutter_tools/gradle/flutter.gradle"
buildscript {
ext.kotlin_version = '1.3.50'
repositories {
google()
jcenter()
}
dependencies {
classpath 'com.android.tools.build:gradle:3.5.3'
classpath "org.jetbrains.kotlin:kotlin-gradle-plugin:$kotlin_version"
classpath 'com.google.gms:google-services:4.3.3'
}
}
allprojects {
repositories {
google()
jcenter()
}
}
rootProject.buildDir = '../build'
subprojects {
project.buildDir = "${rootProject.buildDir}/${project.name}"
}
subprojects {
project.evaluationDependsOn(':app')
}
/*
task clean(type: Delete) {
delete rootProject.buildDir
}
*/
def keystoreProperties = new Properties()
def keystorePropertiesFile = rootProject.file('key.properties')
if (keystorePropertiesFile.exists()) {
keystoreProperties.load(new FileInputStream(keystorePropertiesFile))
}
android {
compileSdkVersion 28
packagingOptions {
exclude ("META-INF/shared_core_release.kotlin_module")
}
defaultConfig {
applicationId "com.domain"
minSdkVersion 25
targetSdkVersion 28
compileSdkVersion 28
versionCode flutterVersionCode.toInteger()
versionName flutterVersionName
testInstrumentationRunner "android.support.test.runner.AndroidJUnitRunner"
}
signingConfigs {
release {
keyAlias keystoreProperties['keyAlias']
keyPassword keystoreProperties['keyPassword']
storeFile keystoreProperties['storeFile'] ? file(keystoreProperties['storeFile']) : null
storePassword keystoreProperties['storePassword']
}
}
buildTypes {
release {
minifyEnabled true
signingConfig signingConfigs.release
sourceSets {
main {
manifest.srcFile 'app/src/main/AndroidManifest.xml'
res.srcDirs = ['app/src/main/res']
}
}
}
}
}
flutter {
source '..'
}
dependencies {
testImplementation 'junit:junit:4.12'
androidTestImplementation 'com.android.support.test:runner:1.0.2'
androidTestImplementation 'com.android.support.test.espresso:espresso-core:3.0.2'
}
```
And this is the referenced `AndroidManifest.xml`:
```
<manifest xmlns:android="http://schemas.android.com/apk/res/android"
package="com.domain">
<application
android:name="io.flutter.app.FlutterApplication"
android:label="App"
android:icon="@mipmap/ic_launcher">
<activity
android:name=".MainActivity"
android:launchMode="singleTop"
android:theme="@style/LaunchTheme"
android:configChanges="orientation|keyboardHidden|keyboard|screenSize|smallestScreenSize|locale|layoutDirection|fontScale|screenLayout|density|uiMode"
android:hardwareAccelerated="true"
android:windowSoftInputMode="adjustResize">
<intent-filter>
<action android:name="android.intent.action.MAIN"/>
<category android:name="android.intent.category.LAUNCHER"/>
</intent-filter>
</activity>
<!-- Don't delete the meta-data below.
This is used by the Flutter tool to generate GeneratedPluginRegistrant.java -->
<meta-data
android:name="flutterEmbedding"
android:value="2" />
</application>
<!-- omitted -->
</manifest>
```
This is the content of `app/src/main/kotlin/com.domain/MainActivity.kt`:
```
package com.domain
import androidx.annotation.NonNull;
import io.flutter.embedding.android.FlutterActivity
import io.flutter.embedding.engine.FlutterEngine
import io.flutter.plugins.GeneratedPluginRegistrant
class MainActivity: FlutterActivity() {
override fun configureFlutterEngine(@NonNull flutterEngine: FlutterEngine) {
GeneratedPluginRegistrant.registerWith(flutterEngine);
}
}
```
This is the excerpt from `flutter doctor`:
```
[√] Flutter (Channel stable, v1.17.5, on Microsoft Windows [Version 10.0.18362.900], locale de-DE)
• Flutter version 1.17.5 at C:\flutter
• Framework revision 8af6b2f038 (9 days ago), 2020-06-30 12:53:55 -0700
• Engine revision ee76268252
• Dart version 2.8.4
[!] Android toolchain - develop for Android devices (Android SDK version 29.0.3)
• Android SDK at C:\Users\xxx\AppData\Local\Android\Sdk
• Platform android-29, build-tools 29.0.3
• ANDROID_HOME = C:\Users\xxx\AppData\Local\Android\Sdk
• Java binary at: C:\Program Files\Android\Android Studio\jre\bin\java
• Java version OpenJDK Runtime Environment (build 1.8.0_242-release-1644-b01)
X Android license status unknown.
Try re-installing or updating your Android SDK Manager.
See https://developer.android.com/studio/#downloads or visit visit
https://flutter.dev/docs/get-started/install/windows#android-setup for detailed instructions.
[√] Android Studio (version 4.0)
• Android Studio at C:\Program Files\Android\Android Studio
• Flutter plugin version 46.0.2
• Dart plugin version 193.7361
• Java version OpenJDK Runtime Environment (build 1.8.0_242-release-1644-b01)
[√] IntelliJ IDEA Ultimate Edition (version 2019.3)
• IntelliJ at C:\Program Files\JetBrains\IntelliJ IDEA 2019.3.4
• Flutter plugin version 44.0.3
• Dart plugin version 193.6911.31
[√] Connected device (1 available)
• Android SDK built for x86 • emulator-5554 • android-x86 • Android 10 (API 29) (emulator)
```
The deployment to the PlayStore itself works without problems. I already looked into the AAB file (and the .dex file) but it's obfuscated. I already tried to build with `--no-shrink` and `--no-obfuscate` but this does not work for some reason
Any help would be highly welcome to resolve this problem.
**Expected results:**
I like the app to run the pre-check without these errors.
**Actual results:**
```
Crash java.lang.RuntimeException: Unable to instantiate activity
ComponentInfo{com.domain/com.domain.MainActivity}: java.lang.ClassNotFoundException: Didn't find
class "com.domain.MainActivity" on path: DexPathList[[zip file "/data/app/com.domain-1/base.apk", zip
file "/data/app/com.domain-1/split_config.arm64_v8a.apk", zip file "/data/app/com.domain-
1/split_config.xxhdpi.apk"],nativeLibraryDirectories=[/data/app/com.domain-1/lib/arm64,
/data/app/com.domain-1/base.apk!/lib/arm64-v8a, /data/app/com.domain-
1/split_config.arm64_v8a.apk!/lib/arm64-v8a, /data/app/com.domain-
1/split_config.xxhdpi.apk!/lib/arm64-v8a, /system/lib64, /vendor/lib64]]
```
|
platform-android,tool,a: release,P2,c: fatal crash,team-android,triaged-android
|
low
|
Critical
|
654,338,719 |
godot
|
GridMap is visible only on the first cull_mask layer
|
<!-- Please search existing issues for potential duplicates before filing yours:
https://github.com/godotengine/godot/issues?q=is%3Aissue
-->
**Godot version:**
Godot Engine v3.2.2.stable.custom_build
**OS/device including version:**
5.7.0-3-MANJARO, OpenGL ES 3.0 Renderer: GeForce GTX 1050 Ti/PCIe/SSE2
**Issue description:**
GridMap is only visible on the first Camera's `cull_mask` layer. Once the first bit of the camera's `cull_mask` is disabled the entire grid disappears. Since it's not extending `VisualInstance` wouldn't it make more sense to not react to `cull_mask` layers?
**Steps to reproduce:**
1. Launch the project below
2. Press the checkbox to toggle camera's first `cull_mask` bit
**Minimal reproduction project:**
<!-- A small Godot project which reproduces the issue. Drag and drop a zip archive to upload it. -->
[gridmap-cull-master.zip](https://github.com/godotengine/godot/files/4899435/gridmap-cull-master.zip)
|
bug,topic:rendering,confirmed
|
medium
|
Major
|
654,341,716 |
TypeScript
|
Suggestion: Add a way to hint typescript about basic function flow (call orders)
|
<!-- 🚨 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
"control flow logic", "function call order", "static flow analysis"
## Suggestion
Add `flow` type, for statically declaring some flow logic (order of function calls)
## Use Cases
For example, we could tell typescript that the `constructor` of a React Component will be called before the `render`, and the render will be called before `componentDidMount`.
## Examples
```javascript
flow F;
let test: undefined | string;
F1~function first() { //the number of the flow indicates the order.
test = 'wow';
}
F2~function second() {
console.log(test.length); // <== ok, second() must come after first() so we know for sure that test was initialized.
}
```
example2:
```javascript
flow LifeCycle;
class Test extends React.Component {
LifeCycle1~constructor(){...}
LifeCycle3~componentDidMount(){...}
LifeCycle2~render(){...}
}
```
## 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
|
654,381,076 |
flutter
|
Scrolling jank in ListView when semantics are enabled
|
Hi i am following flutter and keep in touch with testing different apps developed in flutter.
But one major issue which is continuously under the ignorance and no thread i found on internet for now.. here is what i feel since 6 months of testing different apps.
Listview kind of scrolling jerks, specially when we have images posters like content with it. you can try this by doing create a list with image and some content list view must have content 3x of screen size.
just scroll it in medium speed, you will clearly feel jerks on screen, beside this the same test did on react native and its super smooth.
here is an example app on playstore (Flutter demo: CINEMA) you can download and test, but that i am sure its flutter bug because all similar apps or with long list views this happens.
More examples in flutter official app: Flutter Gallery
1. Flutter Gallery -> (Crane A personalized travel app) -> vertical scroll listview in bottom (Jerk is clearly visible)
2. Flutter Gallery -> (Rally A personalized finance app) -> Login -> vertical scroll it (choppy scrolling Jerks)
Above examples are not debug all are on production playstore, on stackoverflow this is under discussion but everyone is doing just tricks which is not appropriate way to develop an production grade application.
beside this i also tested horizontal listviews and they work fine.
I am expecting senior flutter teams will take this issue serious and resolve this bug on high priority.
@ianloic @anaisbetts @sgraham @gmoothart @sethladd
|
framework,engine,c: performance,a: accessibility,f: scrolling,customer: crowd,perf: speed,has reproducible steps,P2,found in release: 3.7,team-engine,triaged-engine
|
low
|
Critical
|
654,405,038 |
flutter
|
Scrollable Hint Text
|
<!-- 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
If the hint text of a textfield is larger than the space available, it cuts off the text with ellipses. In normal use-cases this is reasonable enough, but there are some where this isn't - particularly when dealing with accessibility & large text sizes.
If the hint text were to be scrollable in the same way that the non-hint-text is, that would eliminate this issue.
### With accessibility features off:
### With accessibility features on (large text + bold on iOS):
(Note that these are the exact same application, and that pretty much any flutter app using textfields w/ hintText would display the same issue).
<!--
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
For multi-line text fields this isn't as much as an issue, but for single-line ones this user experience could be improved by making the hint text scrollable horizontally.
<!--
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.
-->
|
a: text input,platform-ios,framework,f: material design,a: accessibility,c: proposal,P3,team-text-input,triaged-text-input
|
low
|
Critical
|
654,406,812 |
pytorch
|
libTorch cpp docs missing for Tensor::item()
|
## 📚 Documentation
https://pytorch.org/cppdocs/api/classat_1_1_tensor.html#_CPPv4NK2at6Tensor4itemEv
and
https://pytorch.org/cppdocs/api/classat_1_1_tensor.html#_CPPv4I0ENK2at6Tensor4itemE1Tv
are both missing descriptions.
--------
I am trying to get a `float` out of a `Tensor` (that is a single element, kFloat32) on the GPU.
e.g.
`float x = x_tensor.item()`
However, this gives a compile-time error:
`error: cannot convert ‘c10::Scalar’ to ‘double’ in assignment`
Using `float x = x_tensor<float>.item()` compiles just fine, doesn't work either (program spins forever, probably waiting for CUDA kernel to end after a bad memory access or something).
What is the proper way to get the float out of the tensor (and back into main memory)? It prints just fine, so it must be doable.
I'd prefer to do it without printing and parsing or moving the entire tensor back to the CPU.
cc @yf225 @glaringlee @jlin27
|
module: docs,module: cpp,triaged
|
low
|
Critical
|
654,414,230 |
vscode
|
[Request] Provide file-specific onDidChangeSemanticTokens event
|
In C/C++, when a modification is made to a header file, semantic tokens in files that include it could change. (i.e. when renaming a function). Because those other files are not themselves changing, VS Code does not re-request tokens for them (if also open).
There is an `onDidChangeSemanticTokens` event we can use to indicate that tokens have changed. However, this event is not specific to a file, so may cause token requests for files that don't require an update.
Could an additional event be added that allows us to indicate the specific file that needs tokens re-requested? i.e.
```
onDidChangeSemanticTokensForFile?: Event<Uri>;
```
|
feature-request,api,semantic-tokens
|
low
|
Minor
|
654,422,266 |
flutter
|
[flutter_tools] The tool should prefer the JDK vendored with Android Studio
|
As seen in issue #56572, if a version of JDK greater than 8 is set as `JAVA_HOME`, the tool will fail to detect android licenses. Since Android Studio vendors JDK 8, we should prefer that version.
|
tool,a: quality,c: proposal,P3,team-tool,triaged-tool
|
low
|
Major
|
654,426,214 |
opencv
|
Nightly(sporadic crashes): opencv_test_dnn - Test_Caffe_layers.Concat OCL_FP16
|
Observed with:
- MSVS: 2019
- 64-bit
- OpenCL: 26.20.100.7262
Builds examples: [1](http://pullrequest.opencv.org/buildbot/builders/master-win64-vc16/builds/317) [2](http://pullrequest.opencv.org/buildbot/builders/master-win64-vc16/builds/413)
```
[ RUN ] Test_Caffe_layers.Concat/1, where GetParam() = OCV/OCL_FP16
Process returned: 3221226356
ERROR: some tests have failed
program finished with exit code -1
```
|
bug,category: dnn,needs investigation
|
low
|
Critical
|
654,429,565 |
terminal
|
ConPTY: transmit DECSET/DECRST state when client application enters/exits ENABLE_VIRTUAL_TERMINAL_INPUT mode
|
Certain applications (mainly those using the Cygwin/MSYS runtime) will attempt to change VT input parameters while VT input is technically off. Because we do nothing to communicate the expected VT input parameters to a connected terminal before we _turn on VT input passthrough_, it's possible for the connected terminal to disagree with the conhost acting as its pty host.
When the application enters `ENABLE_VIRTUAL_TERMINAL_INPUT` mode, ConPTY should send a string of `DECSET`/`DECRST` sequences to bring the connected terminal in line with it to keep up the illusion of passthrough.
Notes from #6737:
> From the Terminal's perspective, `ENABLE_VIRTUAL_TERMINAL_INPUT` is always on. It can't be lost/forgotten, because the terminal only generates VT. In general, it always has conhost sitting between it and the application to do translation from VT input. There are cases when the conhost sitting in the middle enters passthrough mode and dumps input straight out of terminal into the application.
> Therefore, Terminal needs to be aware of DECCKM. Whether it receives `DECCKM` and therefore whether it _generates the right cursor keys_, however, is based on whether the application has requested `ENABLE_VIRTUAL_TERMINAL_INPUT`.
> (There are applications that ask for DECCKM or other VT input format changes but are _not in VT input mode_, and sending them that data will freak them out.)
>
> For WT 1.0, here's the matrix of behaviors.
>
> <table><tbody>
> <tr><th>Action</th><th>VT Output Mode</th><th>VT Input Mode</th><th>Notes</th></tr>
> <tr>
> <td rowspan="4">Application prints <code>\e[?1h</code></td>
> <td>❌</td>
> <td>❌</td>
> <td rowspan="2">VT output is off, request ignored</td>
> </tr>
> <tr>
> <td>❌</td>
> <td>✅</td>
> <!-- rowspanned -->
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>❌</td>
> <td>conhost (PTY) (only) enables DECCKM</td>
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>✅</td>
> <td>conhost (PTY) _and Terminal_ enable DECCKM</td>
> </tr>
> </tbody></table>
> <table><tbody>
> <tr><th>Action</th><th>VT Output Mode</th><th>VT Input Mode</th><th>PTY in DECCKM mode</th><th>Terminal in DECCKM mode</th><th>Notes</th></tr>
> <tr>
> <td rowspan="8">User presses arrow key</td>
> <td>❌</td>
> <td>❌</td>
> <td>❌</td>
> <td>❌</td>
> <td>Terminal emits <code>\e[A</code> or <code>\eOA</code><br>conhost (PTY) translates it into KEY_EVENT_RECORD</td>
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>❌</td>
> <td>✅</td>
> <td>❌</td>
> <td>Terminal emits <code>\e[A</code><br>conhost (PTY) translates it into KEY_EVENT_RECORD</td>
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>❌</td>
> <td>✅</td>
> <td>✅</td>
> <td>Terminal emits <code>\eOA</code><br>conhost (PTY) translates it into KEY_EVENT_RECORD</td>
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>✅</td>
> <td rowspan="4"><em>passthrough mode</em></td>
> <td>❌</td>
> <td rowspan="2">Terminal emits <code>\e[A</code><br>conhost (PTY) <b>passes it through unmodified</b></td>
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>❌</td>
> <td>✅</td>
> <!-- rowspanned -->
> <td>❌</td>
> <!-- rowspanned -->
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>✅</td>
> <!-- rowspanned -->
> <td>✅</td>
> <td rowspan="2">Terminal emits <code>\eOA</code><br>conhost (PTY) <b>passes it through unmodified</b></td>
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>❌</td>
> <td>✅</td>
> <!-- rowspanned -->
> <td>✅</td>
> <!-- rowspanned -->
> </tr>
> </tbody></table>
>
> <details>
> <summary>BEHAVIORS BEFORE TERMINAL 1.0</summary>
>
> <table><tbody>
> <tr><th>Action</th><th>VT Output Mode</th><th>VT Input Mode</th><th>Notes</th></tr>
> <tr>
> <td rowspan="4">Application prints <code>\e[?1h</code></td>
> <td>❌</td>
> <td>❌</td>
> <td rowspan="2">VT output is off, request ignored</td>
> </tr>
> <tr>
> <td>❌</td>
> <td>✅</td>
> <!-- rowspanned -->
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>❌</td>
> <td rowspan="2">conhost (PTY) (only) enables DECCKM</td>
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>✅</td>
> <!-- rowspanned -->
> </tr>
> </tbody></table>
> <table><tbody>
> <tr><th>Action</th><th>VT Output Mode</th><th>VT Input Mode</th><th>PTY in DECCKM mode</th><th>Notes</th></tr>
> <tr>
> <td rowspan="8">User presses arrow key</td>
> <td>❌</td>
> <td>❌</td>
> <td rowspan="2"><em>ignored</em></td>
> <td rowspan="2">Terminal emits <code>\e[A</code><br>conhost (PTY) translates it into KEY_EVENT_RECORD</td>
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>❌</td>
> <!-- rowspanned -->
> <!-- rowspanned -->
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>❌</td>
> <td>✅</td>
> <td>❌</td>
> <td rowspan="2">Terminal emits <code>\e[A</code><br>conhost (PTY) translates it into <code>\e[A</code></td>
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>✅</td>
> <td>❌</td>
> <!-- rowspanned -->
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>❌</td>
> <td>✅</td>
> <td>✅</td>
> <td rowspan="2">Terminal emits <code>\e[A</code><br>conhost (PTY) translates it into <code>\e[OA</code></td>
> </tr>
> <tr>
> <!--<td rowspan="1">Application prints `\e[?1h`</td>-->
> <td>✅</td>
> <td>✅</td>
> <td>✅</td>
> <!-- rowspanned -->
> </tr>
> </tbody></table>
>
> </details>
> _@DHowett_
> Okay, I'm literally seeing this behavior at the PTY:
>
> * Input Mode = 0x9
> * Input Mode = 0x209
> * Input Mode = 0x9
> * Print out `DECCKM`
> * Input Mode = 0x209
> * Input Mode = 0x9
>
> This happens with `printf`, this happens with less's startup (though it prints more than just DECCKM)`, this happens pretty much every time I launch a process.
>
> It's quite literally disabling VT input mode _only while it is changing the input configuration_. If we'd ever made the console more strict ("VT input mode changes only apply when VT input mode is on"), this would break horribly. I wonder why they'd do this?
> _@DHowett_
> > Whether it receives `DECCKM` and therefore whether it _generates the right cursor keys_, however, is based on whether the application has requested `ENABLE_VIRTUAL_TERMINAL_INPUT`.
>
> I totally did not expect that. If `ENABLE_VIRTUAL_TERMINAL_INPUT` is designed to affect which control sequences a pseudoconsole forwards from `WriteConsole` to the terminal, it is worth documenting.
>
> > This regressed with #6485
>
> Why doesn't Windows Terminal Preview 1.1.1812.0 have the same problem, then? [6ff8ba7](https://github.com/microsoft/terminal/commit/6ff8ba772890bf8abb8431295fe97c8e9e280dcb) was cherry-picked from [5e2c4c6](https://github.com/microsoft/terminal/commit/5e2c4c66e3dcb66c9d60769a2a028e39711168fd), which is included in that version.
> _@KalleOlaviNiemitalo_
|
Product-Conpty,Area-Input,Issue-Bug,Priority-2
|
low
|
Major
|
654,432,731 |
godot
|
Bools are implicitly cast to ints
|
<!-- Please search existing issues for potential duplicates before filing yours:
https://github.com/godotengine/godot/issues?q=is%3Aissue
-->
**Godot version:**
<!-- Specify commit hash if using non-official build. -->
3.2.2
**Issue description:**
<!-- What happened, and what was expected. -->
```GDScript
func _ready() -> void:
test(true)
func test(arg: int):
print(arg)
```
This runs without errors and prints 1.
|
bug,discussion,topic:gdscript
|
low
|
Critical
|
654,458,405 |
youtube-dl
|
Plexstorm
|
<!--
######################################################################
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 2020.06.16.1. 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 **2020.06.16.1**
- [ x] I've checked that all provided URLs are alive and playable in a browser
- [ x] I've checked that none of provided URLs violate any copyrights
- [x ] I've searched the bugtracker for similar site support requests including closed ones
## Example URLs
<!--
Provide all kinds of example URLs support for which should be included. Replace following example URLs by yours.
-->
- Single video: https://www.plexstorm.com/stream/rileyoh
- Single video: https://www.plexstorm.com/stream/velrala
URLs rely upon the streamer being live at the time. https://www.plexstorm.com/ has a live list of current streamers.
## Description
Plexstorm is a gamer streaming site but very much NSFW, like a combination between twitch and pornhub.
<!--
Provide any additional information.
If work on your issue requires account credentials please provide them or explain how one can obtain them.
-->
I have a stub that I wrote in the web console that gives me a link to copy to play on the command line (mostly what I use youtube-dl for):
var a=document.createElement('a'); document.body.appendChild(a); a.setAttribute('style', 'position:absolute; top:0; left:0; z-index:1000;'); a.innerHTML='Video link'; a.href=document.getElementsByTagName('video')[0].firstElementChild.src;
|
site-support-request,nsfw
|
low
|
Critical
|
654,475,575 |
rust
|
Add item suggestions for broken intra doc links
|
The diagnostics for intra-doc links could be improve to match the diagnostics available in rust.
Current error message
```
warning: `[extend]` cannot be resolved, ignoring it.
--> src/vec.rs:704:45
|
704 | /// Note that this function is same as [`extend`] except that it is specialized to work with
| ^^^^^^^^ cannot be resolved, ignoring
|
= note: `#[warn(intra_doc_link_resolution_failure)]` on by default
= help: to escape `[` and `]` characters, just add '\' before them like `\[` or `\]`
```
The diagnostics should suggest `Self::extend()`. Note that it should also handle path ambiguity unlike the current diagnostics with `struct@`, `enum@` and others in the original RFC.
Original RFC: https://github.com/rust-lang/rfcs/blob/master/text/1946-intra-rustdoc-links.md
Tracking issue for intra-doc links: https://github.com/rust-lang/rust/issues/43466
CC ~~@QuietMisdreavus~~ @Manishearth @jyn514
|
T-rustdoc,C-enhancement,A-diagnostics,A-intra-doc-links,A-suggestion-diagnostics
|
low
|
Critical
|
654,487,347 |
deno
|
Runtime features to support Quokka.js
|
Hi Deno team – firstly, I wanted to say thanks for all your work creating deno. It’s impressive and I am personally looking forward to seeing more community adoption.
Our company creates a couple of popular JavaScript/TypeScript developer tools ([Wallaby.js](https://wallabyjs.com/) and [Quokka.js](https://quokkajs.com/)).
We’ve had a number of our Quokka users asking us to [add support for deno](https://github.com/wallabyjs/quokka/issues/456). If you are not familiar with Quokka, Quokka provides an in-editor scratchpad experience for JavaScript/TypeScript. Runtime values and results are displayed in the IDE right next to your code. Quokka re-executes code as soon as you start typing and provides real-time/immediate feedback. For VS Code alone, Quokka has [733,000+ installations](https://marketplace.visualstudio.com/items?itemName=WallabyJs.quokka-vscode) so we’re keen to add support for deno if we can; this may also help drive adoption of deno, or at the very least would help developers explore the deno runtime.
We spent the last week exploring deno and have identified some blockers to being able to support deno (without us forking the deno code base, which we don’t want to do). We were hoping to work with you so we can get to the point of supporting deno.
The issues we have are:
1. Need to be able to cancel Program unload. (as per your [docs](https://deno.land/manual/runtime/program_lifecycle): […unload event] cannot be cancelled)
Providing real-time feedback while developers are typing means that Quokka has to be fast. To achieve that, we recycle the runner process. The way we do this in node.js is we have a network socket that we unref() before we execute user code and then we ref() it again in process.once('beforeExit'); this stops node.js from terminating between runs. For small projects with no dependencies, this probably doesn’t matter too much, but for larger projects, this will stop the effective use of Quokka.
2. Need to be able to invalidate local files on re-execution
This requirement falls out of (1). If we are recycling the runner process, we need to be able to invalidate any cached local files that may have changed when they are subsequently imported again on a fresh execution.
3. Ability to modify input files prior to execution
This is really the “no public API for compilers” issue (https://github.com/denoland/deno/issues/1739) but we can/have solved lack of API for compilers in other ways in the past that may be lower cost to implementing a full API for compilers with exposed AST, etc. In some other scenarios we have achieved similar functionality by duck-punching fs.readFileSync that is used by node.js when modules are required.
Obviously we appreciate that the paradigm for deno is different to node.js and with the focus on providing a secure by default runtime, we can’t do things the same way so may need first class support for some of our requirements vs. doing things in a hacky way.
---
Finally, we do know that we could get Quokka working with deno today without any of these features, but to do this, we’ll need to make some pretty big internal architectural changes and more importantly, we don’t think our approach will scale for larger projects, at least not until we have a solution to (1) and (2).
|
cli,suggestion
|
medium
|
Critical
|
654,496,182 |
TypeScript
|
mixin constructor wrongly typed
|
<!-- 🚨 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:** 3.9.2
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:** mixin constructor signature
**Code**
```ts
class A
{
constructor(public x:number){}
}
function f<T extends { new(...args: any[]): object }>(cls:T) {
return class DynamicProxy extends cls {
constructor(...args: any[]) {
super(1);
}
}
}
const B = f(A);
var b = new B();
b.x;
```
**Expected behavior:** Either make it a warning or just allow any constructor signature.
**Actual behavior:** Errors with `Expected 1 arguments, but got 0.(2554)`
**Playground Link:** https://www.typescriptlang.org/play/index.html?ssl=1&ssc=1&pln=16&pc=5#code/MYGwhgzhAECCBQBve1XWAewHYQC4CcBXYXDfACgAdCAjEAS2GgA8AuLQgWxoFN8BKRAF94I+ADNCWEvWzRxAHgAq0Hs1w8sAExiJoWHgHdyAOjNh8AcwitoYLAE8A2gF1+tjDQBWPEtCEAfOSgNkr80Mho0Pg8uIT4WOjgUNAAIg5YYJyMAAr4GMwOquqaOkm6KFFomDgExKQUZiYW1rb2zm4RlVVVEISUfOQAjPwA3N1VIlEiYjV40ABC0AC88uSwY-AAbhbQNCv6RovkmzQmzKNAA
**Related Issues:** <!-- Did you find other bugs that looked similar? -->
|
Suggestion,Awaiting More Feedback
|
low
|
Critical
|
654,497,679 |
TypeScript
|
tsconfig extend field should be more user friendly
|
## Suggestion
Probably some functionality may refer to editor (i.e VSCode)
`extends` field in `tsconfig.json`:
1) propose auto suggestions (propose existing parent configs in the project)
2) check if a path exists, and expose an error if the path doesn't exist.
My suggestion meets these guidelines:
* [ ] This wouldn't be a breaking change in existing TypeScript/JavaScript code
* [ ] 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
* [ ] 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
|
654,504,978 |
pytorch
|
The name "Contiguous" in `torch::MemoryFormat::Contiguous` can mislead users into assuming contiguous memory
|
When looking at `torch::MemoryFormat::Contiguous`, users can relate it to `tensor.contiguous()`, and assume that by passing it as the memory_format argument to operators/functions, the output tensor will be with contiguous memory. However this is not true e.g:
In [`at::Tensor::suggest_memory_format`](https://github.com/pytorch/pytorch/blob/16f276c/aten/src/ATen/templates/TensorBody.h#L236), it returns `MemoryFormat::Contiguous` for all non NHWC tensors.
In [`tensor.to(memory_format=torch::MemoryFormat::Contiguous)`](https://github.com/pytorch/pytorch/blob/16f276cef94559cb11bc95a4fb3bffc0334f93e5/aten/src/ATen/native/TensorConversions.cpp#L25-L29), non contiguous input tensor `self` can be returned and so there is no guarantee on the output being contiguous.
I would suggest renaming `torch::MemoryFormat::Contiguous` to something else.
cc @ezyang @gchanan @zou3519 @VitalyFedyunin @jamesr66a
|
high priority,triaged,module: memory format
|
low
|
Minor
|
654,584,300 |
pytorch
|
Cleanup git branches
|
## 🚀 Feature
I'd like to propose cleaning up the branches of the pytorch git repo removing all merged or outdated ones
## Motivation
There are currently almost 3800 branches. Lot's of them refer to closed (rejected) PRs or are outdated or merged (e.g. gh/suo/8/base).
Having this many branches makes it harder to work with the repo as the size increases and every new fork will get all those branches making it harder for contributors to create and manage own branches as the amount of existing branches are overwhelming and name clashes are possible.
## Pitch
In a first step with low effort and high gain all branches which are strictly behind current master can be deleted. Those are either merged branches or branches that were started but with no work in them. Removing them doesn't loose anything.
In a second step branches with closed PRs can be deleted. Those are likely rejected features or superseded by other PRs. Loss is possible but very unlikely.
The remaining branches would need to be checked for and deleted as appropriate. For example branches older than a couple years are likely abandoned and no longer relevant. Furthermore the older a branch gets the harder it will be to use it as merge conflicts will make that almost impossible if the code changed significantly in the meantime.
|
triaged,small,better-engineering,actionable
|
medium
|
Major
|
654,624,637 |
create-react-app
|
[docs] I translated Create React App docs into Chinese, can i add a link to readme or send a PR?
|
I translated Create React App docs into Chinese, including docs under docusaurus directory, i want to contribute the works to community.
You could view the docs here: http://docs.breword.com/facebook-create-react-app
Can i add the above link to repo README.md or send a PR with my translated docs content ?
The CRA version i translated is v3.4.1
|
issue: proposal,needs triage
|
low
|
Minor
|
654,647,398 |
flutter
|
[pigeon] Clarify in example that the command depends on the plugin details
|
I am learning how to use pigeon from https://s0pub0dev.icopy.site/packages/pigeon#-example-tab-
But when I call the following command,it prompts me:
```
maixiangMacBook-Pro:vision_ward wyj$ flutter pub run pigeon --input pigeons/message.dart --dart_out lib/pigeon.dart --objc_header_out ios/Runner/pigeon.h --objc_source_out ios/Runner/pigeon.m --java_out ./android/app/src/main/java/com/yj/vision_ward/Pigeon.java --java_package "com.yj.vision_ward"
Unhandled exception:
ProcessException: No such file or directory
Command: dart _pigeon_temp_.dart --input pigeons/message.dart --dart_out lib/pigeon.dart --objc_header_out ios/Runner/pigeon.h --objc_source_out ios/Runner/pigeon.m --java_out ./android/app/src/main/java/com/yj/vision_ward/Pigeon.java --java_package com.yj.vision_ward
pub finished with exit code 255
```
This is my project directory
<img width="391" alt="android@2x" src="https://user-images.githubusercontent.com/14922509/87137322-47fcf380-c2cf-11ea-9fa6-36c76ce2764d.png">
<img width="773" alt="WX20200710-165929@2x" src="https://user-images.githubusercontent.com/14922509/87137018-d91f9a80-c2ce-11ea-95ed-cc69a11cbbbd.png">
And Flutter docrot:
```
..../flutter/bin/flutter doctor --verbose
[✓] Flutter (Channel stable, v1.17.5, on Mac OS X 10.15.4 19E266, locale zh-Hans-CN)
• Flutter version 1.17.5 at /Users/wyj/Library/Flutter/stable/flutter
• Framework revision 8af6b2f038 (10 days ago), 2020-06-30 12:53:55 -0700
• Engine revision ee76268252
• Dart version 2.8.4
[✓] Android toolchain - develop for Android devices (Android SDK version 29.0.3)
• Android SDK at /Users/wyj/Library/Android/sdk
• Platform android-29, build-tools 29.0.3
• Java binary at: /Applications/Android Studio.app/Contents/jre/jdk/Contents/Home/bin/java
• Java version OpenJDK Runtime Environment (build 1.8.0_242-release-1644-b3-6222593)
• All Android licenses accepted.
[✓] Xcode - develop for iOS and macOS (Xcode 11.5)
• Xcode at /Applications/Xcode.app/Contents/Developer
• Xcode 11.5, Build version 11E608c
• CocoaPods version 1.9.1
[✓] Android Studio (version 4.0)
• Android Studio at /Applications/Android Studio.app/Contents
• Flutter plugin version 47.1.2
• Dart plugin version 193.7361
• Java version OpenJDK Runtime Environment (build 1.8.0_242-release-1644-b3-6222593)
[✓] Connected device (1 available)
• JSN AL00a • SJQNW19429000887 • android-arm64 • Android 10 (API 29)
• No issues found!
Process finished with exit code 0
```
pigeon version:
```
pigeon: ^0.1.2+1
```
I have tried to use the immobile version of pigeon.flutter clear ,modify directory.....
but did not find an effective way.
So what is the problem and how can I solve it , thanks in advance.
|
d: api docs,package,team-ecosystem,p: pigeon,P2,triaged-ecosystem
|
low
|
Major
|
654,647,676 |
pytorch
|
"out of scope" variable `cmd` used
|
https://github.com/pytorch/pytorch/blob/df252c059c3ea4c04f8c39c3b4c25a3f4572b0b7/torch/distributed/launch.py#L257
should be `cmd=process.args` ?
cc @pietern @mrshenli @pritamdamania87 @zhaojuanmao @satgera @rohan-varma @gqchen @aazzolini @xush6528 @osalpekar @jiayisuse @agolynski
|
oncall: distributed,triaged,better-engineering
|
low
|
Minor
|
654,663,194 |
vscode
|
Better API to associate a notebook cell document with its notebook
|
A notebook document contains cell documents, which are `vscode.TextDocument` types. There is no easy/nice way to get to a notebook from a document. Ideas
* add a util like `notebook.findContainerNotebook(document: TextDocument): NotebookDocument | undefined`
* add a pointer from the document, e.g `TextDocument#notebook: NotebookDocument | undefined`
|
feature-request,api-finalization,notebook-api
|
low
|
Major
|
654,707,301 |
pytorch
|
test_backward_deadlock fails with "Directory not empty"
|
## 🐛 Bug
I'm running a build & test on a node without a GPU (installation for a cluster)
One test (`TestCppExtensionJIT`) seems to fail with what looks like a race condition. The error is "OSError: [Errno 39] Directory not empty: 'test_backward_deadlock'"
See output below
## To Reproduce
Steps to reproduce the behavior:
1. python setup.py install
1. python tests/run_tests.py
Part of the output before the failure:
```
test_jit_cuda_archflags (__main__.TestCppExtensionJIT) ... skipped 'CUDA not found'
test_jit_cuda_extension (__main__.TestCppExtensionJIT) ... skipped 'CUDA not found'
test_jit_cudnn_extension (__main__.TestCppExtensionJIT) ... skipped 'CuDNN not found'
test_lenient_flag_handling_in_jit_extensions (__main__.TestCppExtensionJIT) ... Loading extension module jit_extension...
Using /home/h3/s3248973/.cache/torch_extensions as PyTorch extensions root...
Creating extension directory /home/h3/s3248973/.cache/torch_extensions/lenient_flag_handling_extension...
Emitting ninja build file /home/h3/s3248973/.cache/torch_extensions/lenient_flag_handling_extension/build.ninja...
Building extension module lenient_flag_handling_extension...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
[1/2] g++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=lenient_flag_handling_extension -DTORCH_API_INCLUDE_EXTENSION_H -I/dev/shm/s3248973-EasyBuild/PyTorch/1.6.0/fosscuda-2019b-Python-3.7.4/pytorch-1.6.0-rc2/test/cpp_extensions -isystem /tmp
/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/
lib/python3.7/site-packages/torch/include/TH -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include/THC -isystem /sw/installed/Python/3.7.4-GCCcore-8.3.0/include/python3.7m -D_GLIBCXX_USE_CXX11_ABI=1 -fP
IC -std=c++14 -g -O0 -Wall -c /home/h3/s3248973/.cache/torch_extensions/lenient_flag_handling_extension/main.cpp -o main.o
[2/2] g++ main.o -shared -L/tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/lib -lc10 -ltorch_cpu -ltorch -ltorch_python -o lenient_flag_handling_extension.so
ok
test_reload_jit_extension (__main__.TestCppExtensionJIT) ... Loading extension module lenient_flag_handling_extension...
Using /home/h3/s3248973/.cache/torch_extensions as PyTorch extensions root...
Creating extension directory /home/h3/s3248973/.cache/torch_extensions/reloaded_jit_extension...
Emitting ninja build file /home/h3/s3248973/.cache/torch_extensions/reloaded_jit_extension/build.ninja...
Building extension module reloaded_jit_extension...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
[1/2] g++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=reloaded_jit_extension -DTORCH_API_INCLUDE_EXTENSION_H -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpN
peQu4/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include/TH -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-package
s/torch/include/THC -isystem /sw/installed/Python/3.7.4-GCCcore-8.3.0/include/python3.7m -D_GLIBCXX_USE_CXX11_ABI=1 -fPIC -std=c++14 -c /home/h3/s3248973/.cache/torch_extensions/reloaded_jit_extension/main.cpp -o main.o
[2/2] g++ main.o -shared -L/tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/lib -lc10 -ltorch_cpu -ltorch -ltorch_python -o reloaded_jit_extension.so
Loading extension module reloaded_jit_extension...
Using /home/h3/s3248973/.cache/torch_extensions as PyTorch extensions root...
The input conditions for extension module reloaded_jit_extension have changed. Bumping to version 1 and re-building as reloaded_jit_extension_v1...
Emitting ninja build file /home/h3/s3248973/.cache/torch_extensions/reloaded_jit_extension/build.ninja...
Building extension module reloaded_jit_extension_v1...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
[1/2] g++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=reloaded_jit_extension_v1 -DTORCH_API_INCLUDE_EXTENSION_H -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include -isystem /tmp/easybuild-tmp/eb-RgtlFs/t
mpNpeQu4/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include/TH -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-pack
ages/torch/include/THC -isystem /sw/installed/Python/3.7.4-GCCcore-8.3.0/include/python3.7m -D_GLIBCXX_USE_CXX11_ABI=1 -fPIC -std=c++14 -c /home/h3/s3248973/.cache/torch_extensions/reloaded_jit_extension/main.cpp -o main.o
[2/2] g++ main.o -shared -L/tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/lib -lc10 -ltorch_cpu -ltorch -ltorch_python -o reloaded_jit_extension_v1.so
Loading extension module reloaded_jit_extension_v1...
Using /home/h3/s3248973/.cache/torch_extensions as PyTorch extensions root...
No modifications detected for re-loaded extension module reloaded_jit_extension_v1, skipping build step...
Loading extension module reloaded_jit_extension_v1...
Using /home/h3/s3248973/.cache/torch_extensions as PyTorch extensions root...
The input conditions for extension module reloaded_jit_extension have changed. Bumping to version 2 and re-building as reloaded_jit_extension_v2...
Emitting ninja build file /home/h3/s3248973/.cache/torch_extensions/reloaded_jit_extension/build.ninja...
Building extension module reloaded_jit_extension_v2...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
[...]
test_set_default_type_also_changes_aten_default_type (__main__.TestCppExtensionJIT) ... Loading extension module is_python_module_v1...
Using /home/h3/s3248973/.cache/torch_extensions as PyTorch extensions root...
Creating extension directory /home/h3/s3248973/.cache/torch_extensions/test_set_default_type...
Emitting ninja build file /home/h3/s3248973/.cache/torch_extensions/test_set_default_type/build.ninja...
Building extension module test_set_default_type...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
[1/2] g++ -MMD -MF main.o.d -DTORCH_EXTENSION_NAME=test_set_default_type -DTORCH_API_INCLUDE_EXTENSION_H -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include/TH -isystem /tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/include/THC -isystem /sw/installed/Python/3.7.4-GCCcore-8.3.0/include/python3.7m -D_GLIBCXX_USE_CXX11_ABI=1 -fPIC -std=c++14 -c /home/h3/s3248973/.cache/torch_extensions/test_set_default_type/main.cpp -o main.o
[2/2] g++ main.o -shared -L/tmp/easybuild-tmp/eb-RgtlFs/tmpNpeQu4/lib/python3.7/site-packages/torch/lib -lc10 -ltorch_cpu -ltorch -ltorch_python -o test_set_default_type.so
ok
test_warning (__main__.TestCppExtensionJIT) ... Loading extension module test_set_default_type...
WARNING: Logging before InitGoogleLogging() is written to STDERR
W0710 12:47:03.169983 24274 main.cpp:12] Warning: Error with CPUDoubleType (function foo)
W0710 12:47:03.170961 24274 main.cpp:12] Warning: Error with CPUDoubleType (function foo)
W0710 12:47:03.171689 24274 main.cpp:12] Warning: Error with CPUDoubleType (function foo)
W0710 12:47:03.172047 24274 main.cpp:12] Warning: Error with CPUDoubleType (function foo)
W0710 12:47:16.688925 24274 main.cpp:12] Warning: Error with torch.DoubleTensor (function foo)
W0710 12:47:16.689333 24274 main.cpp:12] Warning: Error with torch.DoubleTensor (function foo)
W0710 12:47:16.689601 24274 main.cpp:12] Warning: Error with torch.DoubleTensor (function foo)
W0710 12:47:16.690302 24274 main.cpp:12] Warning: Error with torch.DoubleTensor (function foo)
ok
ERROR
======================================================================
ERROR: tearDownClass (__main__.TestCppExtensionJIT)
----------------------------------------------------------------------
Traceback (most recent call last):
File "test_cpp_extensions_jit.py", line 60, in tearDownClass
remove_build_path()
File "test_cpp_extensions_jit.py", line 36, in remove_build_path
shutil.rmtree(default_build_root)
File "/sw/installed/Python/3.7.4-GCCcore-8.3.0/lib/python3.7/shutil.py", line 494, in rmtree
_rmtree_safe_fd(fd, path, onerror)
File "/sw/installed/Python/3.7.4-GCCcore-8.3.0/lib/python3.7/shutil.py", line 436, in _rmtree_safe_fd
onerror(os.rmdir, fullname, sys.exc_info())
File "/sw/installed/Python/3.7.4-GCCcore-8.3.0/lib/python3.7/shutil.py", line 434, in _rmtree_safe_fd
os.rmdir(entry.name, dir_fd=topfd)
OSError: [Errno 39] Directory not empty: 'test_backward_deadlock'
```
## Expected behavior
Tests succeed
## Environment
```
PyTorch version: 1.6.0-rc2
Is debug build: N/A
CUDA used to build PyTorch: N/A
OS: Red Hat Enterprise Linux Server release 7.8 (Maipo)
GCC version: (GCC) 8.3.0
CMake version: version 3.15.3
Python version: 3.7
Is CUDA available: N/A
CUDA runtime version: 10.1.243
GPU models and configuration: Could not collect
Nvidia driver version: Could not collect
cuDNN version: 7.6.4.38
Versions of relevant libraries:
[pip3] numpy==1.17.3
```
cc @suo @gmagogsfm
|
oncall: jit,module: POWER
|
low
|
Critical
|
654,721,923 |
pytorch
|
[discussion] Unite nn modules nn.Something*d to just nn.Something with appropriate options
|
I proposed it in https://github.com/pytorch/pytorch/issues/41081#issuecomment-655148816, but maybe a separate issue is a better place to discuss this:
A more radical proposal: unite BatchNorm\*d / SyncBatchNorm in one module BatchNorm (with option sync = False|True, especially given that SyncBatchNorm is now established in core and is reasonably common; and maybe an explicit "arity" argument if needed). I think now BatchNorm*d are all same except shape checking (and already implemented in a common parent class _BatchNorm). Then switching on sync would become
```python
for bn in [module for module in model.modules() if isinstance(module, torch.nn.BatchNorm)]:
bn.sync_mode = True # or bn.sync_mode(True)
```
There may be same proposal to unite Conv\*d into Conv, MaxPool\*d into MaxPool etc. Dimensionality may be defined by shape of kernel_size.
Fusion code may become simpler, several versions of fused modules (that now separately exist ConvBn\*d, LinearBn\*d etc) would likely be simplified and unified.
If not for all, maybe for some cases this can be done (like we already have only one nn.SyncBatchNorm, and functional forms are F.batch_norm and not three separate functions)
I think this design already exists in ONNX.
Even if some of these \*d variants could stay for some clarity, it would be good to have also usable public base class for them (with maybe some `num_dims` argument that would specify this arity) that would enable "polymorphism"
cc @albanD @mruberry
|
module: nn,triaged
|
low
|
Major
|
654,730,806 |
pytorch
|
torch.tan(complex) on CUDA doesn't handle nonfinite values properly
|
```
t = torch.tensor(complex(float('-inf'), 0))
print(torch.tan(t))
print(torch.tan(t.cuda()))
print(np.tan(t.numpy()))
```
will print:
```
tensor(nan+0.j)
tensor(nan+nanj, device='cuda:0')
(nan+0j)
```
The CUDA result (in the middle) is incorrect.
cc @ezyang @anjali411 @dylanbespalko @ngimel
|
module: cuda,triaged,module: complex
|
low
|
Minor
|
654,733,389 |
pytorch
|
torch.sign is divergent from numpy.sign on NaN
|
```
torch.sign(torch.tensor(float('nan')))
: tensor(0.)
np.sign(float('nan'))
: nan
```
cc @mruberry @rgommers
|
triaged,module: numpy
|
low
|
Minor
|
654,743,731 |
pytorch
|
torch.abs(complex) is divergent from NumPy on vectorized NaN values
|
```
c = complex(float('nan'), float('inf'))
l = [c] * 300
t = torch.tensor(l)
torch.abs(t)
```
will produce a tensor with mostly `nan` values. NumPy, however, and non-vectorized `torch.abs`, will produce `inf` values.
cc @ezyang @anjali411 @dylanbespalko
|
triaged,module: complex,module: NaNs and Infs
|
low
|
Minor
|
654,761,829 |
flutter
|
AVIF codec support
|
<!-- 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
AVIF aims to be the successor of jpg on the web and is growing very fast. AVIF images also compress with much better results then jpg. So it would be very good for flutter applications to support AVIF.
## Proposal
Add AVIF codec support to flutter.
|
c: new feature,engine,customer: crowd,a: images,c: proposal,P3,team-engine,triaged-engine
|
high
|
Critical
|
654,775,788 |
react-native
|
React native does not support screen refresh rate
|
Please provide all the information requested. Issues that do not follow this format are likely to stall.
## Description
When running a react native app on a device that has a refresh rate other than 60, react native apps still run at 60 fps.
This means that if you have a device like a Samsung Galaxy S20, which runs at 120 fps, every react native app will feel sluggish, slow and underperforming.
React native should honor the screen's refresh rate and use that for rendering.
## React Native version:
Run `react-native info` in your terminal and copy the results here.
```
System:
OS: macOS 10.15.5
CPU: (8) x64 Intel(R) Core(TM) i7-7820HQ CPU @ 2.90GHz
Memory: 791.56 MB / 16.00 GB
Shell: 3.2.57 - /bin/bash
Binaries:
Node: 12.16.1 - ~/.nvm/versions/node/v12.16.1/bin/node
Yarn: 1.19.1 - ~/.yarn/bin/yarn
npm: 6.13.4 - ~/.nvm/versions/node/v12.16.1/bin/npm
Watchman: 4.9.0 - /usr/local/bin/watchman
Managers:
CocoaPods: 1.9.1 - /usr/local/bin/pod
SDKs:
iOS SDK:
Platforms: iOS 13.5, DriverKit 19.0, macOS 10.15, tvOS 13.4, watchOS 6.2
Android SDK:
API Levels: 21, 22, 23, 24, 25, 26, 27, 28, 29
Build Tools: 25.0.0, 26.0.1, 26.0.2, 27.0.2, 27.0.3, 28.0.2, 28.0.3, 29.0.0, 29.0.2
System Images: android-16 | Google APIs Intel x86 Atom, android-19 | Google APIs Intel x86 Atom, android-21 | Google APIs Intel x86 Atom_64, android-22 | Google APIs Intel x86 Atom, android-23 | Google APIs Intel x86 Atom, android-23 | Google APIs Intel x86 Atom_64, android-24 | Google Play Intel x86 Atom, android-25 | Google APIs Intel x86 Atom, android-25 | Google Play Intel x86 Atom, android-26 | Google Play Intel x86 Atom, android-29 | Google APIs Intel x86 Atom, android-29 | Google Play Intel x86 Atom, android-P | Google Play Intel x86 Atom, android-Q | Google APIs Intel x86 Atom
Android NDK: 21.0.6113669
IDEs:
Android Studio: Not Found
Xcode: 11.5/11E608c - /usr/bin/xcodebuild
Languages:
Java: 1.8.0_152 - /usr/bin/javac
Python: 2.7.16 - /usr/bin/python
npmPackages:
@react-native-community/cli: Not Found
react: 16.11.0 => 16.11.0
react-native: 0.62.2 => 0.62.2
npmGlobalPackages:
*react-native*: Not Found
```
## Steps To Reproduce
Provide a detailed list of steps that reproduce the issue.
1. Use a phone with a higher refresh rate than 60 (e.g. Samsung Galaxy S20, Oneplus 7T (or higher))
2. Start any react native app
3. See that the app is slower than every app on the phone
## Expected Results
I expect that the app runs smooth on all refresh rates above 60.
## Snack, code example, screenshot, or link to a repository:
|
Needs: Triage :mag:
|
high
|
Major
|
654,785,544 |
flutter
|
[flutter doctor] Cocoapods was not found even after sudo gem install cocoapods
|
```
Doctor summary (to see all details, run flutter doctor -v):
[✓] Flutter (Channel master, 1.20.0-8.0.pre.41, on Mac OS X 10.14.6 18G5033,
locale zh-Hans-CN)
[✓] Android toolchain - develop for Android devices (Android SDK version 29.0.2)
[!] Xcode - develop for iOS and macOS (Xcode 11.3)
! Unknown CocoaPods version installed.
Flutter is unable to determine the installed CocoaPods's version.
Ensure that the output of 'pod --version' contains only digits and . to
be recognized by Flutter.
To upgrade:
sudo gem install cocoapods
[✓] Android Studio (version 3.4)
[!] IntelliJ IDEA Ultimate Edition (version 2020.1)
✗ Flutter plugin not installed; this adds Flutter specific functionality.
✗ Dart plugin not installed; this adds Dart specific functionality.
[✓] VS Code (version 1.46.1)
[✓] Connected device (1 available)
! Doctor found issues in 2 categories.
➜ ~ pod --version
Resolving dependencies...
1.9.3
```
|
platform-ios,tool,t: flutter doctor,a: first hour,P3,team-ios,triaged-ios
|
low
|
Major
|
654,786,805 |
go
|
x/build: some Gerrit CLs are getting into a state where they can't be updated
|
From `gopherbot` logs, there are a number of CLs where a reviewer fails to be added due to the error "Change nnn may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.":
```
2020/07/10 13:27:29 No reviewers or cc: https://go-review.googlesource.com/c/text/+/238119
2020/07/10 13:27:29 Setting review on https://go-review.googlesource.com/c/text/+/238119: {Message: Labels:map[] Comments:map[] Reviewers:[{Reviewer:[email protected] State:}]}
2020/07/10 13:27:29 Could not set review for change "text~238119": HTTP status 409 Conflict; Change 238119 may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.
2020/07/10 13:27:29 No reviewers or cc: https://go-review.googlesource.com/c/mobile/+/212839
2020/07/10 13:27:29 Setting review on https://go-review.googlesource.com/c/mobile/+/212839: {Message: Labels:map[] Comments:map[] Reviewers:[{Reviewer:[email protected] State:}]}
2020/07/10 13:27:29 Could not set review for change "mobile~212839": HTTP status 409 Conflict; Change 212839 may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.
2020/07/10 13:27:29 No reviewers or cc: https://go-review.googlesource.com/c/mobile/+/231681
2020/07/10 13:27:29 Setting review on https://go-review.googlesource.com/c/mobile/+/231681: {Message: Labels:map[] Comments:map[] Reviewers:[{Reviewer:[email protected] State:}]}
2020/07/10 13:27:29 Could not set review for change "mobile~231681": HTTP status 409 Conflict; Change 231681 may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.
2020/07/10 13:27:29 No reviewers or cc: https://go-review.googlesource.com/c/mobile/+/234239
2020/07/10 13:27:29 Setting review on https://go-review.googlesource.com/c/mobile/+/234239: {Message: Labels:map[] Comments:map[] Reviewers:[{Reviewer:[email protected] State:}]}
2020/07/10 13:27:29 Could not set review for change "mobile~234239": HTTP status 409 Conflict; Change 234239 may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.
2020/07/10 13:27:29 No reviewers or cc: https://go-review.googlesource.com/c/mobile/+/240009
2020/07/10 13:27:30 Setting review on https://go-review.googlesource.com/c/mobile/+/240009: {Message: Labels:map[] Comments:map[] Reviewers:[{Reviewer:[email protected] State:}]}
2020/07/10 13:27:30 Could not set review for change "mobile~240009": HTTP status 409 Conflict; Change 240009 may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.
2020/07/10 13:27:30 No reviewers or cc: https://go-review.googlesource.com/c/go/+/223118
2020/07/10 13:27:30 Setting review on https://go-review.googlesource.com/c/go/+/223118: {Message: Labels:map[] Comments:map[] Reviewers:[{Reviewer:[email protected] State:}]}
2020/07/10 13:27:30 Could not set review for change "go~223118": HTTP status 409 Conflict; Change 223118 may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.
2020/07/10 13:27:30 No reviewers or cc: https://go-review.googlesource.com/c/tools/+/223779
2020/07/10 13:27:30 Setting review on https://go-review.googlesource.com/c/tools/+/223779: {Message: Labels:map[] Comments:map[] Reviewers:[{Reviewer:[email protected] State:}]}
2020/07/10 13:27:30 Could not set review for change "tools~223779": HTTP status 409 Conflict; Change 223779 may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.
2020/07/10 13:27:30 No reviewers or cc: https://go-review.googlesource.com/c/tools/+/231464
2020/07/10 13:27:30 Setting review on https://go-review.googlesource.com/c/tools/+/231464: {Message: Labels:map[] Comments:map[] Reviewers:[{Reviewer:[email protected] State:}]}
2020/07/10 13:27:30 Could not set review for change "tools~231464": HTTP status 409 Conflict; Change 231464 may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.
2020/07/10 13:27:30 No reviewers or cc: https://go-review.googlesource.com/c/tools/+/239498
2020/07/10 13:27:31 Setting review on https://go-review.googlesource.com/c/tools/+/239498: {Message: Labels:map[] Comments:map[] Reviewers:[{Reviewer:[email protected] State:}]}
2020/07/10 13:27:31 Could not set review for change "tools~239498": HTTP status 409 Conflict; Change 239498 may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.
2020/07/10 13:27:31 No reviewers or cc: https://go-review.googlesource.com/c/net/+/187478
2020/07/10 13:27:31 Setting review on https://go-review.googlesource.com/c/net/+/187478: {Message: Labels:map[] Comments:map[] Reviewers:[{Reviewer:[email protected] State:}]}
2020/07/10 13:27:31 Could not set review for change "net~187478": HTTP status 409 Conflict; Change 187478 may not exceed 5000 updates. It may still be abandoned or submitted. To continue working on this change, recreate it with a new Change-Id, then abandon this one.
```
It's not possible to leave review comments on those CLs.
Need to investigate why this is happening.
From initial observations, it might be CLs that GerritBot imported from PRs, where PRs have had large review comments posted (e.g., https://github.com/golang/tools/pull/215#pullrequestreview-445781521), which might have caused some sort of fail-and-repeat loops to make an excess of 5000 API calls to the Gerrit CL, causing Gerrit to lock those CLs. But I've only looked very briefly so far, more investigation is needed to confirm this or find another explanation.
/cc @golang/osp-team
|
Builders,NeedsInvestigation
|
low
|
Critical
|
654,799,729 |
node
|
Undeprecate _stream_wrap
|
**Is your feature request related to a problem? Please describe.**
_stream_wrap was [recently deprecated](https://github.com/nodejs/node/pull/26245) because it "doesn't have a reasonable use outside of node".
I think I have two reasonable use cases, and I'd really like this to be undeprecated. As far as I can tell it's going to exist internally anyway, since the underlying [JSStreamSocket](https://github.com/nodejs/node/blob/master/lib/internal/js_stream_socket.js) implementation is actively used by TLSSocket, it works perfectly, and it already supports my use cases, so this is purely a question of whether it should be accessible externally.
My use cases:
1. I'm trying to sniff packets to detect HTTP/2 before the http2 server takes over, as described in #17132 (in my case: because I want to support HTTP/1, HTTP/2, HTTPS/1 and HTTPS/2 all on the same port). As described in that issue that's not normally possible, understandably, because sockets are tightly integrated with the stream internals in a way that disallows these kinds of interactions, and changing that has performance implications for the common case. However, this is possible if you provide some psuedo-socket, backed by a separate stream, such as `_stream_wrap`.
There are other users with this use case, including the original author of that issue, and every user of [httpolyglot](https://www.npmjs.com/package/httpolyglot), which does this for HTTP/1 (but which can't HTTP/2 using the native http2 module because of this issue). Issues like #20999 exist because this isn't available, so people have to implement it externally for themselves.
2. I'm trying to tunnel HTTP connections through a stream (in this case, an Http2Stream). I'm building a mitm proxy, which accepts h2 CONNECT requests and then locally handles them. That means after the CONNECT I get an h2 stream, which I need to treat as a fresh net.Socket connection. `_stream_wrap` does this perfectly.
This is less common, afaik, but it's an example of wanting the networking APIs to be generally more composeable, so that socket-based code can fully interoperate with streams or stream transformations.
**Describe the solution you'd like**
Make `_stream_wrap` a public API again, removing the deprecations.
Maybe even remove the underscore too, and make it a 1st class API with docs etc?
**Describe alternatives you've considered**
The alternative I can see is to reimplement `_stream_wrap` from scratch externally.
That's not really straightforward, as implementation here requires native code (https://github.com/nodejs/node/blob/master/src/js_stream.cc), which I'd also need to duplicate, and my pipeline to usefully build & distribute all this ends up being pretty complicated, especially to get it working across node versions.
It would also be far less reliable (I'd much rather use the real & well-maintained version of this, especially since it touches lots of node internals), and it's annoying to have to do all this when a fully working implementation is shipping within node anyway.
|
stream
|
medium
|
Major
|
654,807,790 |
godot
|
Option button giving false item count C#
|
3.2.2
I have created an OptionButton and added 3 items in the editor.
I run the program and us a FOR loop to change the text of each item.
I will get the following 12 errors:
```
E 0:00:16.906 get_item_text: Index p_idx = 3 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 4 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 5 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 6 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 7 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 8 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 9 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 10 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 11 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 12 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 13 is out of bounds (items.size() = 3).
E 0:00:16.906 get_item_text: Index p_idx = 14 is out of bounds (items.size() = 3).
```
When I check what the extra items are, it gives me the other properties of each item. For instance, it is counting the item ID and the Enabled property as individual items
|
topic:dotnet,topic:gui
|
low
|
Critical
|
654,814,640 |
opencv
|
Stitcher generates panorama smaller than input
|
OpenCV 4.3.0
I am currently trying the OpenCV Stitcher on very degraded images just to check for the limits of mosaicing.
For the algorithm not to hang, I have to set the bundleAdjuster to NoBundleAdjuster,
I also set WaveCorrection to false because it is not a linear travelling
Is it expected behaviour that the pano of two images can be smaller than input ?
For two 640x480 images, the resulting pano is 279x257. (with status being OK, not an error like ERR_NEED_MORE_IMGS)
Since the compositionResol is at -1 (ORIG_RESOL), I do not understand how it can happen.
Is there anything else to tune to force the resulting pano to be at least the size of the largest input ?
[edit]
The two images are almost identical (the camera did not move yet, only lighting and noise changed a little)
I tried with different bundle adjusters, and still the same : the pano is smaller, while it should have matched the two images almost perfectly
|
category: stitching,needs reproducer
|
low
|
Critical
|
654,847,318 |
go
|
x/build/cmd/gopherbot, x/build/maintner: "close stale WaitingForInfo" task failing on non-existing issue
|
From `gopherbot` logs:
```
2020/07/10 13:33:30 close stale WaitingForInfo: POST https://api.github.com/repos/golang/go/issues/39453/comments: 422 Validation Failed [{Resource:IssueComment Field:data Code:unprocessable Message:Could not resolve to a node with the global id of 'MDU6SXNzdWU2MzM4NzY4NTM='.}]
2020/07/10 13:33:30 gopherbot ran in 9.88987365s
```
Issue https://github.com/golang/go/issues/39453 is 404. `gopherbot` should handle this situation as a non-error. Treating it as an error adds an unnecessary sleep:
```
2020/07/10 13:33:30 sleeping 30s after previous error.
```
It's possible this is a bug or unexpected behavior on the GitHub API side (might be worth reporting it to them), but if it persists, we should just update `gopherbot` to handle it.
It's also worth investigating if this is an issue that can and is better to fix in maintner, rather than gopherbot.
/cc @golang/osp-team
|
Builders,NeedsInvestigation
|
low
|
Critical
|
654,862,967 |
opencv
|
opencv build error, could NOT find python2. Is this something I should be worried about?
|
- OpenCV = latest version
- Operating System / Platform => Windows 64 Bit
- Compiler => Visual Studio 2019
--
Compiled opencv with cmake, is this error something to worry about, or can I ignore it?
Help Required
I downloaded opencv today from the master brand and I am using python 3.7.7. I have cmake 3.18.0 and I am on windows 10. I got this error after building opencv:
CMake Warning at cmake/OpenCVDetectPython.cmake:81 (message):
CMake's 'find_host_package(PythonInterp 2.7)' found wrong Python version:
PYTHON_EXECUTABLE=C:/Python37/python.exe
PYTHON_VERSION_STRING=3.7.7
Consider providing the 'PYTHON2_EXECUTABLE' variable via CMake command line
or environment variables
Call Stack (most recent call first):
cmake/OpenCVDetectPython.cmake:271 (find_python)
CMakeLists.txt:599 (include)
Could NOT find Python2 (missing: Python2_EXECUTABLE Interpreter)
Reason given by package:
Interpreter: Wrong major version for the interpreter "C:/Python37/python.exe"
Do I have to use a different version of python? I plan on using opencv with the darknet framework/yolov4.
|
category: build/install,future
|
low
|
Critical
|
654,869,860 |
pytorch
|
Add nn.functional.swish alias for nn.functional.silu
|
SiLU activation function added here https://github.com/pytorch/pytorch/pull/41034 is also known as Swish (see https://github.com/pytorch/pytorch/pull/41034#issuecomment-654496698 for a discussion about it).
To help users find the function we should add an alias for SiLU by the name of Swish once the aliasing system @mruberry is working on is complete.
cc @albanD @mruberry
|
module: nn,triaged
|
low
|
Minor
|
654,919,748 |
rust
|
rustdoc generates broken "Read more" links when the destination crate doesn't have a URL
|
I tried this code:
`foo.rs`:
```rust
#![crate_type = "rlib"]
pub struct FooStruct;
pub trait FooTrait {
/// docs
///
/// more docs
fn foo_method();
}
impl FooTrait for FooStruct {
fn foo_method() {}
}
```
`bar.rs`:
```rust
extern crate foo;
pub use foo::FooStruct;
```
running `rustc foo.rs && rustdoc -L . bar.rs` so only `bar` is documented.
I expected to see this happen:
When the docs for `bar` are documented the page for `FooStruct` shouldn't contain any broken "Read more" links.
Instead, this happened:
The "Read more" link for `foo_method` on the page for `FooStruct` points to `#tymethod.foo_method` which doesn't go anywhere. It should point to the documentation for `FooTrait` but as that page doesn't exist the "Read more" link shouldn't exist at all. The link works fine if docs for `foo` are also generated or `pub use foo::FooTrait;` is added to `bar.rs`.
### Meta
`rustc --version --verbose`:
```
rustc 1.46.0-nightly (5db778aff 2020-07-09)
binary: rustc
commit-hash: 5db778affee7c6600c8e7a177c48282dab3f6292
commit-date: 2020-07-09
host: x86_64-pc-windows-msvc
release: 1.46.0-nightly
LLVM version: 10.0
```
|
T-rustdoc,C-bug,A-cross-crate-reexports,A-rustdoc-ui
|
low
|
Critical
|
654,923,115 |
flutter
|
[flutter_tools] IDE plugin validators should be deprecated
|
Supporting the filepaths to the android studio plugins is a moving target, that has proven difficult to keep up to date. In addition, the warning makes Flutter users think that the SDK will not function properly, when this is a completely informational warning; in other words, **Flutter does not directly use the IDE plugins, so doctor failing to detect them does not affect any behavior** (it's just annoying).
|
tool,c: proposal,P2,team-tool,triaged-tool
|
low
|
Major
|
654,943,601 |
excalidraw
|
It's not clear that the Library auto-expands or that a placeholder is empty
|
(1) If I open the library initially, nothing indicates that there isn't a component to re-use. Perhaps hide empty ones while nothing is selected? If it's completely empty, it's good to keep a single placeholder with a flashing + button, perhaps with a tooltip.
(2) After adding three elements, it's unclear that the library will auto-expand if I selected another element:
It could be covered by always showing one empty placeholder with a visible + button, at least after adding Nth element while the Library is still open.
Thanks!
|
enhancement
|
low
|
Minor
|
654,944,569 |
vscode
|
Open remote settings GUI when extension executes workbench.action.openSettings with query in remote session
|
<!-- ⚠️⚠️ Do Not Delete This! bug_report_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: https://code.visualstudio.com/insiders/ -->
<!-- Use Help > Report Issue to prefill these. -->
- VSCode Version: 1.47.0
- OS Version: Windows_NT x64 10.0.18363
Steps to Reproduce:
I work on the Python extension. Currently, in a remote session, the 'Preferences: Open Settings (UI)' command correctly opens the settings GUI with the Remote setting tab selected.
However, in a remote session (WSL and Remote-SSH), `vscode.commands.executeCommand('workbench.action.openSettings', 'query string')` opens the GUI with the User tab, instead of the Remote setting tab, selected. This behavior is unexpected for the user, as described here: https://github.com/microsoft/vscode-python/issues/12862.
Our use case is that we'd like to direct the user to a filtered view of the settings GUI, as described in #84355, which happens when a query string is provided with `workbench.action.openSettings`. It would be great if `workbench.action.openSettings` could detect that it's being executed in a remote environment and open the GUI with the Remote settings tab selected as well.
|
feature-request,settings-editor,confirmation-pending
|
low
|
Critical
|
655,002,948 |
vscode
|
encodedSemanticClassifications-full does not request cancellation when completions are requested
|
<!-- ⚠️⚠️ Do Not Delete This! bug_report_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Also please test using the latest insiders build to make sure your issue has not already been fixed: https://code.visualstudio.com/insiders/ -->
<!-- Use Help > Report Issue to prefill these. -->
- VSCode Version:
- OS Version:
Steps to Reproduce:
1.
2.
<!-- Launch with `code --disable-extensions` to check. -->
Does this issue occur when all extensions are disabled?: Yes/No
Found this while investigating https://github.com/microsoft/TypeScript/issues/35364 which turns out that semantic colorization takes approximately 10s but completions are quick with or without that request but seem to take longer as colorization is not cancelled.
|
typescript,javascript,debt,perf,keep
|
low
|
Critical
|
655,015,695 |
go
|
path/filepath: clarify VolumeName returns a path not a name
|
### What version of Go are you using (`go version`)?
<pre>
$ go version
1..14.4
</pre>
### Does this issue reproduce with the latest release?
Yes.
### What operating system and processor architecture are you using (`go env`)?
Operating system and architecture independent.
### What did you do?
Read the docs.
### What did you expect to see?
```
func VolumeRoot(path string) string
```
or something similar that reflects what `VolumeName` actually returns.
-or-
Documentation to the effect that `VolumeName` in fact returns a volume name.
A volume name on Windows is also known as a volume label.
### What did you see instead?
The function certainly does not return a volume name, hence either the name or its behavior and documentation are wrong.
Having `VolumeName("C:\A")` return "C:" is confusing and inconsistent. It could also be considered wrong, since "C:" is a relative path which would resolve to `%CD%` or `C:.` and `Abs` is also documented to never return `C:` for probably this reason. However this issue is about the function returning a path, not a name, as one would not be overly unjustified expect.
|
NeedsInvestigation
|
low
|
Minor
|
655,049,260 |
go
|
x/crypto/acme: Retry on LetsEncrypt rate limit
|
### What version of Go are you using?
<pre>
go version go1.14 darwin/amd64
</pre>
### What did you do?
I called `acme.Client.AuthorizeOrder` when the renewal rate limit for the desired domain has been exceeded.
See: https://letsencrypt.org/docs/rate-limits/
### What did you expect to see?
I expected the call to fail quickly and return the rate limit error. There is no value in retrying for a rate limit failure.
### What did you see instead?
The acme HTTP client code will retry for 429 errors.
<pre>
acme/http.go
func isRetriable(code int) bool {
return code <= 399 || code >= 500 || code == http.StatusTooManyRequests
}
</pre>
In the case of a LetsEncrypt rate violation LetsEncrypt returns 429. For example:
<pre>
429 urn:ietf:params:acme:error:rateLimited: Error creating new order :: too many certificates already issued for exact set of domains: example.com: see https://letsencrypt.org/docs/rate-limits/
</pre>
This results in the retry code executing until the context timeout is reached. If the context has no timeout then the code will make calls to LetsEncrypt forever.
|
NeedsInvestigation
|
low
|
Critical
|
655,058,213 |
terminal
|
New icons for UI elements
|
<!--
🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
I ACKNOWLEDGE THE FOLLOWING BEFORE PROCEEDING:
1. If I delete this entire template and go my own path, the core team may close my issue without further explanation or engagement.
2. If I list multiple bugs/concerns in this one issue, the core team may close my issue without further explanation or engagement.
3. If I write an issue that has many duplicates, the core team may close my issue without further explanation or engagement (and without necessarily spending time to find the exact duplicate ID number).
4. If I leave the title incomplete when filing the issue, the core team may close my issue without further explanation or engagement.
5. If I file something completely blank in the body, the core team may close my issue without further explanation or engagement.
All good? Then proceed!
-->
# Description of the new feature/enhancement
We need some new iconography for some of our UI elements, specifically the command palette and settings UI. Below are the requested icons:
| Command / Title | Icon proposal |
| --------------- | ------------- |
| Launch (SUI) | Rocket ship |
| Split horizontally | Window with horizontal pane split |
| Split vertically | Window with vertical pane split |
<!--
A clear and concise description of what the problem is that the new feature would solve.
Describe why and how a user would use this new functionality (if applicable).
-->
# Proposed technical implementation details (optional)
<!--
A clear and concise description of what you want to happen.
-->
|
Issue-Feature,Area-UserInterface,Product-Terminal
|
medium
|
Critical
|
655,067,062 |
pytorch
|
[docs] Strange arg names of torch.bmm/mm and tensor.bmm/mm
|
https://pytorch.org/docs/master/generated/torch.bmm.html#torch.bmm:
`torch.bmm(input, mat2, deterministic=False, out=None) → Tensor`
`mat2` is quite strange
https://pytorch.org/docs/master/tensors.html?highlight=bmm#torch.Tensor.bmm:
`bmm(batch2) → Tensor`
`batch2` is strange as well (and does not even match torch.bmm)
It seems that it's a remnant of torch.addbmm, there batch1/batch2 make more sense. But combinations input/mat2 and self/batch2 don't make much sense
Same for torch.mm/tensor.mm
cc @jlin27
|
module: docs,triaged,enhancement
|
low
|
Minor
|
655,078,789 |
pytorch
|
add a reparameterized version of inverse Gaussian distribution
|
## 🚀 Feature
<!-- A clear and concise description of the feature proposal -->
Like the gamma distribution, the inverse Gaussian distribution is also implicitly re-parameterizable.
## Motivation
This distribution can be used in variational inference such as Pyro. Moreover, inverse Gaussian is usually more numerically stable than gamma.
<!-- Please outline the motivation for the proposal. Is your feature request related to a problem? e.g., I'm always frustrated when [...]. If this is related to another GitHub issue, please link here too -->
## Pitch
<!-- A clear and concise description of what you want to happen. -->
The `autograd` implementation including some test cases is available at `https://github.com/yorkerlin/iBayesLRule/blob/master/inverse_gauss/inv_gauss_sampler.py`
The writeup can be found at
Appendix H.1 (page 27) of `https://arxiv.org/pdf/2002.10060.pdf` (to appear at ICML 2020)
An efficient natural-gradient variational inference for mixture of full Gaussians (MoG) is also discussed in our paper.
@fritzo
## Alternatives
<!-- A clear and concise description of any alternative solutions or features you've considered, if any. -->
## Additional context
<!-- Add any other context or screenshots about the feature request here. -->
cc @vincentqb @fritzo @neerajprad @alicanb @vishwakftw
|
module: distributions,triaged,enhancement
|
low
|
Minor
|
655,093,646 |
pytorch
|
List[Any] support in TorchScript
|
Forwarded from internal post
> Why is returning [1] for List[Any] not ok, while returning 1 for Any is fine?
```
To work around this, I had to use
untyped_output: List[Any] = []
for i in range(self.output_size):
untyped_output.append(output[i])
```
cc @ezyang @gchanan @zou3519 @suo @gmagogsfm
|
high priority,triage review,oncall: jit,weeks
|
low
|
Major
|
655,094,937 |
pytorch
|
Return list of tuples from custom operator
|
Forwarded from internal post
> Not having the ability to return List<Tuple<something, something>> from a custom operator means we have to return Tuple<List<something>, List<something>>, and convert on the torchscript side, which I think is expensive. Whereas, if we could just return the list of tuples, we could directly just use the result.
Adding a custom operator that returns a list of tuples is not supported. I'm using c10::List<std::tuple<at::Tensor, at::Tensor>> and getting Tried to use undefined class as output.
There is a workaround, namely returning a tuple of two lists.
cc @suo @gmagogsfm
|
oncall: jit
|
low
|
Minor
|
655,096,491 |
pytorch
|
[RFC, Tracker] DataLoader improvements
|
This issue created to track all current problems of `DataLoader` (and related components such as `DataSet`, `Sampler`, `Transforms`). It is focused on **what** we want to archive. Implementation details will follow after prioritizing.
### Known problems
- Issues interpretability. DataLoader periodically ends in a situation when the process hangs for an unknown to the user reason. Root cause analysis of such cases is overcomplicated due to cryptic Exception traces and/or inability to reproduce quickly. Examples #39570 #36375 #33296 #31758 #30147 #25522. We plan to:
- [ ] Implement proper timeouts handling and timeouts control.
- [ ] Make DataLoader return readable and actionable exceptions.
- [ ] Make DataLoader return usable traces in the case of Ctrl+C and similar OS signals #22924.
- Issues with CPU Utilization. Usage of DataLoader frequently ends with oversubscribing to CPU threads or CPU underutilization. We plan to:
- [ ] Make it easier to manage and understand the number of used threads/processes.
- GPU underutilization. Multiple users reported problems with the CUDA context and multiprocessing. Examples #40403 #35759. We plan to:
- [ ] Review the process of spawning processes to make sure it is easy/understandable to use. We plan to:
- [ ] Unlock the ability to use GPU `Transforms` easily.
- Recreation of the processes. Spawning processes is a costly procedure, especially in the case of CUDA usage. We plan to:
- [ ] Reuse processes between training epochs.
- Memory usage issues. In case of unbalanced loading/training speed DataLoader behavior leads to OOM. Examples #31101 #25691. We plan to:
- [ ] Add ability to control amount of pre-loaded data.
### Documentation
- Clear Initialization process doc. Some users getting confused at what time DataSet / DataLoader getting initialized. The order must be clear and understandable.
- Clear Multiprocessing / Multithreading docs. With recipes (torchaudio is a good example). Especially covering cases when threading meets forking and when cuda meets forking.
### Benchmarking
- To make sure that performance improvements are trackable and we are not introducing slow downs we need to write and document benchmark scripts / methodology.
- Add documentation how to measure performance of data loading, so users can identify bottlenecks.
- Ability to benchmark CPU vs GPU Transforms.
### Improvements
- Controllable size of batch on every step of pipeline. Various features (such as WebData, Batch Loading in general, Minibatces, batch level Transforms) requires various sizes of batch (1-N) on every step of the process. We need to make it possible to specify it on every level of loading graph.
- WebData #38419. As datasets become larger and larger, storing training samples as individual files becomes impractical and inefficient. This can be addressed using sequential storage formats and sharding.
- Related issues: #12672 #24985 #26957 #42405
- Improved Sampling. Sampling is important component of data loading pipeline. Next requirements should be considered in case of architecture changes.
- #25162
- Cascading Sampling
- Seq2Seq requirements #25743
- #28743
- Introduce ability to choose between multi-threading and multi-processing. Right now we provide only multi-processing solution, making multi-threading possible will potentially unlock more performant pipelines.
- Preprocessing/Networking. We currently support Distributed training but it is impossible to do distributed data preprocessing at this moment. We should consider this option.
- Infinite DataSet/DataLoader. There is significant number of requests to provide DataSet/DataLoader which never stops to produce batches. Ex: https://discuss.pytorch.org/t/implementing-an-infinite-loop-dataset-dataloader-combo/35567 https://www.google.com/search?q=pytorch+Infinite+DataSet
- Extend ConcatDataSet functionality. There are more than one way to combine two DataSets.
- Caching support. In some cases caching is the easiest way to store entire DataSet in the memory. We can unify caching API to avoid multiple non optimal solutions. #35642 #39274
- State saving / restoration for DataSet / DataLoader / Sampler. Restarting training from specific checkpoint is problematic when size of single epoch is too large. We can introduce method to save/restore data pipeline state. #36650.
- Improve collate_fn experience
- #33181
- #27617
- Unify Transforms Interface.
- Unlock ability to make JIT-able transforms
- Make it painless to make GPU transforms
- Make it possible to execute transforms on batch
### Considering Improvements
- C++ implementation of DataLoader components for performance/c++ parity
cc @SsnL
cc @vadimkantorov
|
module: dataloader,triaged
|
medium
|
Major
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.