id
int64 393k
2.82B
| repo
stringclasses 68
values | title
stringlengths 1
936
| body
stringlengths 0
256k
β | labels
stringlengths 2
508
| priority
stringclasses 3
values | severity
stringclasses 3
values |
|---|---|---|---|---|---|---|
311,876,046 | flutter | How to write custom ImageFilters? | I'm interested in writing custom ImageFilters as a possible approach to a photo filter application, as well as other possible use-cases.
I see that the current options for ImageFilter are bounded and that they're implemented in native land.
Is it even possible to write new ImageFilters, either in Dart or on the native side? If so, are there any resources to help figure out how to do that?
| c: new feature,framework,engine,a: images,P2,team-engine,triaged-engine | low | Major |
311,889,268 | flutter | flutter install -d all doesn't work to install on all devices even tho it says to. | ```
$ flutter devices
2 connected devices:
KFSUWI β’ G000NQ0373110E58 β’ android-arm64 β’ Android 5.1.1 (API 22)
Android SDK built for x86 β’ emulator-5554 β’ android-x86 β’ Android P (API 27) (emulator)
$ flutter install
More than one device connected; please specify a device with the '-d <deviceId>' flag, or use '-d all' to act on all devices.
KFSUWI β’ G000NQ0373110E58 β’ android-arm64 β’ Android 5.1.1 (API 22)
Android SDK built for x86 β’ emulator-5554 β’ android-x86 β’ Android P (API 27) (emulator)
No target device found
$ flutter install -d all
More than one device connected; please specify a device with the '-d <deviceId>' flag.
KFSUWI β’ G000NQ0373110E58 β’ android-arm64 β’ Android 5.1.1 (API 22)
Android SDK built for x86 β’ emulator-5554 β’ android-x86 β’ Android P (API 27) (emulator)
No target device found
```
This was on windows with an emulator and an amazon fire | c: new feature,tool,a: quality,P3,team-tool,triaged-tool | low | Minor |
311,961,782 | go | net: writing to TCP socket returns incorrect number of bytes upon timeout on Windows | Please answer these questions before submitting your issue. Thanks!
### What version of Go are you using (`go version`)?
go version go1.10.1 windows/amd64
### Does this issue reproduce with the latest release?
Yes - it seems to have been there for ages.
### What operating system and processor architecture are you using (`go env`)?
I use Windows 10 64 bit.
### What did you do?
The easiest way to see and reproduce the problem is by using the proof-of-concept tools.
First the server part - you can download it from here: https://pastebin.com/NUdxuMDw
Start the server on any PC that has TCP port 11111 open and accessible.
OS of the server machine does not matter.
Now download the actual test client: https://pastebin.com/SFQyyHfb
Run the test client on Windows.
You just need to specify one parameter on the command line: IP or host name of your running server (the one with open TCP port 11111).
The test is supposed to send 1MB of data to the server.
But you will see that it actually sends more....
The reason for it, is that whenever the Write() function returns a timeout, it always returns n=0.
Which seems to be wrong, as you can sometimes see some data going through.
This problem does not occur when you run the same code on Linux (there you can actually see n>0 being returned).
I have not checked other platforms.
### What did you expect to see?
Always 1MB of data received at the server side.
### What did you see instead?
More than 1MB of data received at the server side. | help wanted,OS-Windows,NeedsInvestigation | low | Major |
311,995,825 | youtube-dl | Support for Asiancammodels, etc. streams | ## Please follow the guide below
- You will be asked some questions and requested to provide some information, please read them **carefully** and answer honestly
- Put an `x` into all the boxes [ ] relevant to your *issue* (like this: `[x]`)
- Use the *Preview* tab to see what your issue will actually look like
---
### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2018.04.03*. If it's not, read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
- [x] I've **verified** and **I assure** that I'm running youtube-dl **2018.04.03**
### Before submitting an *issue* make sure you have:
- [x] At least skimmed through the [README](https://github.com/rg3/youtube-dl/blob/master/README.md), **most notably** the [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
- [x] [Searched](https://github.com/rg3/youtube-dl/search?type=Issues) the bugtracker for similar issues including closed ones
- [x] Checked that provided video/audio/playlist URLs (if any) are alive and playable in a browser
### What is the purpose of your *issue*?
- [ ] Bug report (encountered problems with youtube-dl)
- [x] Site support request (request for adding support for a new site)
- [ ] Feature request (request for a new functionality)
- [ ] Question
- [ ] Other
---
### Description of your *issue*, suggested solution and other information
I would like it if Youtube-dl could support streams from Asiancammodels.com and its clones. Examples of links to streams are as follows:
https://www.asiancammodels.com/chat/Sexyhornygirl
https://m.asiancammodels.com/Sexyhornygirl?chat | nsfw | low | Critical |
312,006,724 | vscode | [html][css] Investigate cross css/html, multi file support | Approach:
- merge HTML & CSS servers to avoid cross LSP communication
- in the server build an (in-memory) index of for the contents of css and html files in the current folder
Benefits:
In HTML
- code completion in 'class' attributes using the collected class names (same for ids)
- goto & peek references on 'class' attributes (same for ids)
In CSS/LESS/SCSS
- code completion for class names based on already used class names in HTML and CSS (same for ids)
- goto & peek references on 'class' attributes (same for ids)
- code completion for variables defined in other css files
| feature-request,css-less-scss,html | medium | Critical |
312,011,998 | pytorch | worker assignments in torch.utils.dataloader.py |
In https://github.com/pytorch/pytorch/blob/60a16e5663f35f2df17ae1d5b060d0e1accdbefd/torch/utils/data/dataloader.py#L297, all items from the dataset for a given minibatch are assigned to the same worker process. This means that the first minibatch always takes the full sequential processing time, and it can similarly lead to wait times later if the consumer ever races ahead to consume all prefeched minibatches quickly. Wouldn't it be better to distribute individual items (rather than entire minibatches) in a round robin fashion?
Thank you!
cc @SsnL @VitalyFedyunin | module: dataloader,triaged | low | Minor |
312,022,002 | rust | [parallel-queries] Refactor error message handling to better integrate with parallel queries | At the moment error message handling uses a [callback-based setup](https://github.com/rust-lang/rust/blob/2253216ec500601f9c2998196cb1b0a1ed0ae576/src/librustc_errors/lib.rs#L285-L286) in order to handle diagnostics tracking during query evaluation. This could probably be handled in a cleaner way.
Also, the current system has no way of making error message output somewhat deterministic under parallel execution.
Both problems should be solved by re-factoring the system to better fit the changed needs. | C-enhancement,T-compiler,A-parallel-queries | low | Critical |
312,056,743 | TypeScript | Goto definition after field exclusion should go to prop signature | <!-- π¨ STOP π¨ π¦π§π’π£ π¨ πΊπ»πΆπ· π¨ -->
<!--
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 CONTRIBUTING guidelines: https://github.com/Microsoft/TypeScript/blob/master/CONTRIBUTING.md
* Read the FAQ: https://github.com/Microsoft/TypeScript/wiki/FAQ
-->
<!-- If you have a QUESTION:
THIS IS NOT A FORUM FOR QUESTIONS.
Ask questions at http://stackoverflow.com/questions/tagged/typescript
or https://gitter.im/Microsoft/TypeScript
-->
<!-- If you have a SUGGESTION:
Most suggestion reports are duplicates, please search extra hard before logging a new suggestion.
See https://github.com/Microsoft/TypeScript-wiki/blob/master/Writing-Good-Design-Proposals.md
-->
<!-- If you have a BUG:
Please fill in the *entire* template below.
-->
<!-- Please try to reproduce the issue with `typescript@next`. It may have already been fixed. -->
**TypeScript Version:** master
**Code**
```ts
type OmitKeys<A, B> = {[P in Exclude<keyof A, B>]: A[P]}
interface Foo {
a: number;
b: string;
c: boolean;
}
var c: OmitKeys<Foo, 'c'>;
c.a // goto definition doesn't work
```
**Expected behavior:**
Goto definition on `c.a` takes me to the `Foo.a` property
**Actual behavior:**
Goto definition doesn't work
| Suggestion,Experience Enhancement,Domain: Symbol Navigation | low | Critical |
312,062,935 | TypeScript | In JS, there's no way to assert that a property is definitely assigned | Keywords: JSDoc, Salsa, JavaScript, definite, initialization, initialized, assigned, assertion
```js
class ElementWrapper {
constructor() {
this.init();
}
init() {
this.element = document.createElement('div');
}
getElementStyle() {
this.element.style // error: element is possibly undefined
}
}
```
We need a way to convince TypeScript in `strictNullChecks` that `this.element` is initialized in `getElementStyle`. Basically a definite initialization assertion.
Related is #23405 which tracks non-null assertions on the expression level. | Suggestion,Needs Proposal,Domain: JavaScript | low | Critical |
312,063,933 | go | cmd/go: clarify that special directories can still be used for package paths | Hi,
The current documentation is a bit unfortunate when it comes to the special directories understood by the go tool.
https://golang.org/cmd/go/#hdr-Test_packages
> The go tool will ignore a directory named "testdata", making it available to hold ancillary data needed by the tests.
https://golang.org/cmd/go/#hdr-Package_lists
> Directory and file names that begin with "." or "_" are ignored by the go tool, as are directories named "testdata".
However, those directories can currently be used just as well for compiling. Unfortunately, I don't have a better wording yet that would describe that only certain operations will behave in this special mode for these special rules/directories.
This comes up often enough as tooling can suggest completion symbols from these directories then users can interpret this as a problem with the tooling rather than special cases in certain go tool usages.
Thank you. | Documentation,NeedsInvestigation | low | Major |
312,065,924 | pytorch | [feature request] Unpooling layer in Caffe2 | It would be really useful to have unpooling layers in Caffe2, much like torch.nn.MaxUnpool2d etc in PyTorch.
| caffe2,feature | low | Minor |
312,070,078 | TypeScript | Do not type check .js files in referenced node_modules under `--checkJs` and `--maxNodeModuleJsDepth` > 0 | - Create a `jsconfig.json`
- Set `checkJs` to `true`
- run `tsc --p ./jsconfig.json`
Expected:
no errors in `node_modules\*\*.js`.
There is no really good fix for these issues. i would suggest not checking these files
| Bug,checkJs,Domain: JavaScript | medium | Critical |
312,086,199 | TypeScript | Inconsistent type compatibility for a type with a call signature and and index signature |
<!-- Please try to reproduce the issue with `typescript@next`. It may have already been fixed. -->
**TypeScript Version:** 2.9.0-dev
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:** index signature intersection
**Code**
```ts
type INOk = {
(): string;
[name: string] : number;
}
type IOk = {
(): string;
} & {
[name: string] : number;
}
declare let val:(() => "") & { foo: number; }
let ok: IOk = val; // This works
let nok: INOk = val; // This does not
```
**Expected behavior:**
Both `IOk` and `INOk` have the same public structure, they both have a call signature and are indexable, both assign statements should be valid.
**Actual behavior:**
The second assign statement fails with the message `Index signature is missing in type '(() => "") & { foo: number; }'.`
**Playground Link:** [link](https://www.typescriptlang.org/play/index.html#src=type%20INOk%20%3D%20%7B%0A%20%20%20%20()%3A%20string%3B%0A%20%20%20%20%5Bname%3A%20string%5D%20%3A%20number%3B%0A%7D%0A%0Atype%20IOk%20%3D%20%7B%0A%20%20%20%20()%3A%20string%3B%0A%7D%20%26%20%7B%0A%20%20%20%20%5Bname%3A%20string%5D%20%3A%20number%3B%0A%7D%0A%0Adeclare%20let%20val%3A(()%20%3D%3E%20%22%22)%20%26%20%7B%20foo%3A%20number%3B%20%7D%0Alet%20ok%3A%20IOk%20%3D%20val%3B%20%2F%2F%20This%20works%0Alet%20nok%3A%20INOk%20%3D%20val%3B%20%2F%2F%20This%20does%20not%0A)
**Related Issues:** https://github.com/Microsoft/TypeScript/issues/15300
**Notes**
Looking at the checker code, it appears that for `IOk` type compatibility is checked for each constituent of the intersection type, so we will have :
```ts
isRelatedTo (typeof val, IOk) =
isRealtedTo(typeof val, () => "")) // == True, since val has a call signature
( // Above equal to
isRelatedTo(() => "", () => "") // == True
||
isRelatedTo({ foo: number; }, () => "") // == False but it's does not matter
)
&&
isRealtedTo(typeof val, { [name: string] : number }) // = True since the { foo: number} part of typeof val has an inferable index (ie isObjectTypeWithInferableIndex( { foo: number} ) will return true)
( // Above equal to
isRelatedTo(() => "", { [name: string] : number }) // == False but does not matter
||
isRelatedTo({ foo: number; }, { [name: string] : number }) // == True,
)
```
While for `INOk` the relation is checked directly, since `INOk` can't be split into constituents and we have
```ts
isRelatedTo (typeof val, INOk) =
isRelatedTo(() => "", INOk) // == False INOk has index, but ()=> "" does not
||
isRelatedTo({ foo: number; }, INOk) // == False, No compatible call signature
```
So then the checker falls back to structural checking (`recursiveTypeRelatedTo`) but this fails as well because when checking for index compatibility (inside `indexTypesRelatedTo`), it decides that the `typeof value` (`(() => "") & { foo: number; }`) does not have an inferable index (`isObjectTypeWithInferableIndex` returns false because the intersection type does not have a `symbol` and even if it did the condition for inferable index checks that the type does not have a call signature (`!typeHasCallOrConstructSignatures(type)`)) | Suggestion,Awaiting More Feedback | low | Major |
312,130,796 | go | net/http: consider making Transport clean up after itself | If a Transport falls out of scope with idle connections, it will leak.
This is surprising for users (for example #24719), and obscure/inconsistent. `os.File` closes resources when falling out of scope, and Transport doesn't even have a Close method, only CloseIdleConnections. Idle connections are an implementation detail.
Implementing a finalizer would require making sure that all blocked goroutines don't keep pointers to the Transport itself, which might be hard.
Some previous discussion at #16005. | Thinking | low | Major |
312,139,996 | flutter | Make MethodChannel non-final or make it implement an interface to allow Mockito mocking | When writing plugin test code, it's useful to be able to mock out the `MethodChannel` to assert calls. Mockito, the popular mocking library for Java, doesn't allow mocking of final classes. Propositions:
- Make `MethodChannel` non-final so it can be mocked. I understand why you may not want to do this though.
- Make `MethodChannel` implement an interface so that anyone can mock it. | a: tests,c: new feature,engine,P2,a: plugins,team-engine,triaged-engine | low | Minor |
312,173,419 | go | spec: inconsistency about allowable conversions of numeric constants to strings | In section "Conversions", the Go spec says:
> A constant value x can be converted to type T if x is representable by a value of T. As a special case, an integer constant x can be converted to a string type using the same rule as for non-constant x.
Note that "integer constants" are distinct from "rune constants", "float constants", etc.
However, the spec also includes two examples of conversions of rune literals to type string: `string('x')` and `string('a')`.
cmd/compile and go/types allow these conversions. I assume gccgo does too, as utf8.RuneError is an untyped rune constant, and `string(utf8.RuneError)` appears a few times in the standard library.
I think the wording should be tweaked to "an integer **or rune** constant x".
Aside: I'd think we should relax it further to allow any integer *value*. We allow `make([]byte, 1.0)`, so it seems inconsistent to not allow `string(1.0)`.
/cc @griesemer | Documentation,NeedsFix | low | Critical |
312,175,037 | kubernetes | Introduce exponential moving average to autoscaling controller | <!-- This form is for bug reports and feature requests ONLY!
If you're looking for help check [Stack Overflow](https://stackoverflow.com/questions/tagged/kubernetes) and the [troubleshooting guide](https://kubernetes.io/docs/tasks/debug-application-cluster/troubleshooting/).
-->
**Is this a BUG REPORT or FEATURE REQUEST?**:
> Uncomment only one, leave it on its own line:
>
> /kind bug
/kind feature
**What happened**:
Current implementation of the pod autoscaler controller is basically calculating the desired replicas via the instantaneous observed metrics value and comparing it with the current replicas and then do the actual scaling by the difference. This will cause some ramification especially when the metrics value varies frequently over time. This may also potentially influence the ok-rate or performance of some services.
IMHO, introducing `EMA(exponential moving average)` and storing a tiny time-series of metrics in the controller can help fixes this problem. It's also adopted by Jenkins for calculating load statistics:
Some introduction to `Exponential Smoothing`:
[https://en.wikipedia.org/wiki/Exponential_smoothing](https://en.wikipedia.org/wiki/Exponential_smoothing)
Prior art by Jenkins: [https://github.com/kohsuke/jenkins/blob/55203ebeed1b7e182878d3e3c1184ac042f20473/core/src/main/java/hudson/slaves/NodeProvisioner.java#L727](https://github.com/kohsuke/jenkins/blob/55203ebeed1b7e182878d3e3c1184ac042f20473/core/src/main/java/hudson/slaves/NodeProvisioner.java#L727)
We can add annotations like `autoscaling.kubernetes.io/ema-decay: 0.5` which is decay coefficient of EMA. Each autoscaler may have its own decay, if the annotation is absent then we should default the decay to `0` so that it's compatible with the current logic.
**Anything else we need to know?**:
**Environment**:
- Kubernetes version (use `kubectl version`):
- Cloud provider or hardware configuration:
- OS (e.g. from /etc/os-release):
- Kernel (e.g. `uname -a`):
- Install tools:
- Others:
| sig/autoscaling,kind/feature,lifecycle/frozen | medium | Critical |
312,199,103 | opencv | OpenCv 3.4.1 for Android: Bad JavaCamera2View performance | #I recently upgraded to 3.4.1 and build it to support new `JavaCamera2View`.
On my Nexus 7 (2013) I resolved a FPS decrease over time bug with `JavaCameraView`.
`JavaCamera2View` now gives slow, but stable ~10 FPS . Not acceptable but maybe improvable.
Performance on my Galaxy S8 differs much. ~30 FPS with `JavaCameraView`, without any processing.
But only **2** FPS! with `JavaCamera2View`, which is not usable for any needs.
In both test cases, `JavaCamera2View` has a very improved behaviour in resolution settings, scaling / zooming behaviour and image quality (removed some blurry and bad quality effects) and maybe that is some kind of the problem.
As far as I can say the resolution, quality and autofocus behaviour is so good, that it drops the FPS to a minimum. But that is just my "newbie" opinion.
Buffer managment could also be a problem, but I can't imagine why the performance is better on an older Device than S8 (Nexus 7).
really need to solve this issue because I'd love to use the new `JavaCamera2View` since the quality of camera is much better
ty in advance
##### System information (version)
- OpenCV => 3.4.1 with contrib
- Operating System / Platform => Ubuntu 16.04
- Compiler Settings => build_sdk.py with ndk-config.16 - downgraded android SDK
#### Edit
the devices get really really hot after a few minutes of running a simple Java2CameraView | platform: android | low | Critical |
312,200,673 | rust | strange end span in borrowck error | This code legitimately errors:
```rust
#![allow(warnings)]
use std::rc::Rc;
struct Data {
vector: Vec<Datum>
}
struct Datum {
/* ... */
}
struct Consumer {
data: Rc<Data>
}
impl Consumer {
fn process_data(&mut self) {
for datum in &self.data.vector {
self.process_datum(datum);
}
}
fn process_datum(&mut self, datum: &Datum) {
/* ... */
}
}
fn main() {
}
```
but the span looks strange:
```
error[E0502]: cannot borrow `*self` as mutable because `self.data` is also borrowed as immutable
--> src/main.rs:19:7
|
18 | for datum in &self.data.vector {
| --------- - immutable borrow ends here
| |
| immutable borrow occurs here
19 | self.process_datum(datum);
| ^^^^ mutable borrow occurs here
```
Why is "immutable borrow ends here" highlighting the end of the word `vector`? | C-enhancement,A-diagnostics,T-compiler | low | Critical |
312,203,024 | ant-design | Allow something other than the field "name" property to be the default when showing error messages. | - [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?
By default, when an error message is displayed is uses `"%s is required"`, where `%s` is the name of the field. The field name isn't necessarily something to present to the user. For example, my field name might be `parentId`, but to the user I want to display `Parent`. In essence, across the entirety of my app I want to use the label as opposed to the name to display errors.
### What does the proposed API look like?
I had a browse around the source, it seems as though `fullField` is an attempt to allow this, however it doesn't appear to work correctly. See: https://codesandbox.io/s/v06vq5zlx0 which attempts to use fullField.
A potential API would be something like:
```
rules: [{
required: true,
errorProp: 'Parent'
}]
```
<!-- generated by ant-design-issue-helper. DO NOT REMOVE --> | π‘ Feature Request,Inactive | low | Critical |
312,211,692 | neovim | win: revisit os_setperm() / setfperm | (migrated from https://github.com/neovim/neovim/issues/5229 )
`os_setperm()` needs to be re-thought; it is nonsense on Windows because of uv_fs_chmod:
https://github.com/libuv/libuv/blob/e76b8838e51f6e8f1944b6c6d50c3948ed764a0b/src/win/fs.c#L1072-L1081
ACL support may be needed, e.g.
- https://github.com/neovim/neovim/issues/1200
- https://github.com/neovim/neovim/pull/4060
# Status
- Implemented in #10379 but that PR needs a revival. | platform:windows,complexity:low,system | low | Minor |
312,213,078 | godot | Crash signal 7 (SIGBUS), code 1 (BUS_ADRALN) _ZN10StringName5unrefEv | Hi,
I'm getting the following crash on my game (Godot 2.1.4)
*** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
pid: 0, tid: 0 >>> it.megasoft78.fallingword <<<
backtrace:
#00 pc 00000000008a8de4 /data/app/it.megasoft78.fallingword-2/lib/arm/libgodot_android.so (_ZN10StringName5unrefEv+56)
#01 pc 00000000008a9b3c /data/app/it.megasoft78.fallingword-2/lib/arm/libgodot_android.so (_ZN10StringNameD2Ev+8)
#02 pc 00000000008669f4 /data/app/it.megasoft78.fallingword-2/lib/arm/libgodot_android.so (_ZN12MessageQueue5flushEv+356)
#03 pc 000000000006b978 /data/app/it.megasoft78.fallingword-2/lib/arm/libgodot_android.so (_ZN4Main9iterationEv+456)
#04 pc 000000000002f448 /data/app/it.megasoft78.fallingword-2/lib/arm/libgodot_android.so (_ZN10OS_Android17main_loop_iterateEv+20)
#05 pc 000000000003d448 /data/app/it.megasoft78.fallingword-2/lib/arm/libgodot_android.so (Java_org_godotengine_godot_GodotLib_step+3140)
#06 pc 00000000005b47bd /data/app/it.megasoft78.fallingword-2/oat/arm/base.odex
On the following devices:
- Samsung Galaxy J3(2017) (j3y17lte), 2048MB RAM, Android 7.0
- Samsung Galaxy S7 (herolte), 4096MB RAM, Android 7.0
- Samsung Galaxy J5(2016) (j5xnlte), 2048MB RAM, Android 7.1
No idea how to reproduce. :(
| bug,platform:android,topic:porting | medium | Critical |
312,216,025 | go | x/build: create VERSION file for cmd/dist to read | https://build.golang.org/log/89d71a5b0369a691899c3435abc7df3d4abd4074
```
##### API check
Error running API checker: exit status 1
Go version is "builder host-linux-armel-cross", ignoring -next /home/builder/stage0scratch/workdir/go/api/next.txt
[ a lot of +pkg ... lines ]
exit status 1
2018/04/07 13:27:59 Failed: exit status 1
2018/04/07 13:28:04 FAILED
```
The fact that the version does not contain `devel` implies that `+pkg` lines, i.e. new APIs, are a test failure:
```
fail = !compareAPI(bw, features, required, optional, exception,
*allowNew && strings.Contains(runtime.Version(), "devel"))
```
The fact that no other builder hits this tells me that the version in this ARM builder is wrong; it should contain `devel`. | Testing,Builders | low | Critical |
312,241,061 | go | misc/cgo/testtls: skip test if C toolchain exhibits known TLS bugs | ```
fatal error: unexpected signal during runtime execution
[signal SIGSEGV: segmentation violation code=0x1 addr=0x1b pc=0xe56e4]
runtime stack:
runtime.throw(0x16739f, 0x2a)
/workdir/go/src/runtime/panic.go:598 +0x54
runtime.sigpanic()
/workdir/go/src/runtime/signal_unix.go:372 +0x22c
goroutine 5 [syscall, locked to thread]:
runtime.cgocall(0xe5678, 0x429f78, 0xe5208)
/workdir/go/src/runtime/cgocall.go:128 +0x64 fp=0x429f60 sp=0x429f48 pc=0x123c0
_/workdir/go/misc/cgo/testtls._Cfunc_getTLS(0x0)
_cgo_gotypes.go:43 +0x38 fp=0x429f74 sp=0x429f60 pc=0xe5134
_/workdir/go/misc/cgo/testtls.testTLS(0x478120)
/workdir/go/misc/cgo/testtls/tls.go:21 +0x34 fp=0x429fb0 sp=0x429f74 pc=0xe5214
_/workdir/go/misc/cgo/testtls.TestTLS(0x478120)
/workdir/go/misc/cgo/testtls/tls_test.go:12 +0x1c fp=0x429fb8 sp=0x429fb0 pc=0xe50ec
testing.tRunner(0x478120, 0x168c34)
/workdir/go/src/testing/testing.go:791 +0xac fp=0x429fe4 sp=0x429fb8 pc=0xb3074
runtime.goexit()
/workdir/go/src/runtime/asm_arm.s:840 +0x4 fp=0x429fe4 sp=0x429fe4 pc=0x63184
created by testing.(*T).Run
/workdir/go/src/testing/testing.go:838 +0x240
goroutine 1 [runnable]:
testing.(*T).Run(0x478120, 0x1604a5, 0x7, 0x168c34, 0x410540)
/workdir/go/src/testing/testing.go:839 +0x260
testing.runTests.func1(0x478090)
/workdir/go/src/testing/testing.go:1081 +0x50
testing.tRunner(0x478090, 0x44bef8)
/workdir/go/src/testing/testing.go:791 +0xac
testing.runTests(0x40c040, 0x206598, 0x1, 0x1, 0x458100)
/workdir/go/src/testing/testing.go:1079 +0x21c
testing.(*M).Run(0x458100, 0x0)
/workdir/go/src/testing/testing.go:996 +0x134
main.main()
_testmain.go:42 +0x12c
exit status 2
FAIL _/workdir/go/misc/cgo/testtls 0.046s
```
Some recent occurences:
https://build.golang.org/log/b7e1728e48a73089be64d42ea3b7e581eeae029c
https://build.golang.org/log/56a9b6b774c15433baff5b11a2f2eeb8aaa936a1 | Testing,help wanted,NeedsFix | medium | Critical |
312,266,058 | rust | incremental compilation: could not create session directory lock file: No locks available (os error 37) | I was unable to incrementally compile using cargo build.
I had added the rand dependency to the toml file, per the advice of the Guessing Game exercise in Rust Programming Language (Second Edition).
Before this particular error, I had just finished setting up a synced folder with NFS on my vagrant box (ubuntu 16.04) where rust was installed. There is some issue using the default vagrant shared folder setup with rust.
I tried this code:
```rust
use std::io;
fn main() {
println!("Guess the number");
println!("Please input your guess.");
let mut guess = String::new();
io::stdin().read_line(&mut guess)
.expect("Failed to read line");
print!("You guessed: {}", guess);
}
```
I expected to see this happen:
compile
Instead, this happened:
```
Updating registry `https://github.com/rust-lang/crates.io-index`
Downloading rand v0.3.22
Downloading libc v0.2.40
Downloading rand v0.4.2
Compiling libc v0.2.40
Compiling rand v0.4.2
Compiling rand v0.3.22
Compiling guessing_game v0.1.0 (file:///Rust/projects/guessing_game)
error: incremental compilation: could not create session directory lock file: No locks available (os error 37)
thread 'rustc' panicked at 'librustc/session/mod.rs:727: Trying to get session directory from IncrCompSession `NotInitialized`', librustc/session/mod.rs:1180:26
note: Run with `RUST_BACKTRACE=1` for a backtrace.
error: internal compiler error: unexpected panic
note: the compiler unexpectedly panicked. this is a bug.
note: we would appreciate a bug report: https://github.com/rust-lang/rust/blob/master/CONTRIBUTING.md#bug-reports
note: rustc 1.25.0 (84203cac6 2018-03-25) running on x86_64-unknown-linux-gnu
error: Could not compile `guessing_game`.
```
`rustc --version --verbose`:
```
rustc 1.25.0 (84203cac6 2018-03-25)
binary: rustc
commit-hash: 84203cac67e65ca8640b8392348411098c856985
commit-date: 2018-03-25
host: x86_64-unknown-linux-gnu
release: 1.25.0
LLVM version: 6.0
```
Backtrace:
```
thread 'rustc' panicked at 'librustc/session/mod.rs:727: Trying to get session directory from IncrCompSession `NotInitialized`', librustc/session/mod.rs:1180:26
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.
stack backtrace:
0: std::sys::unix::backtrace::tracing::imp::unwind_backtrace
at libstd/sys/unix/backtrace/tracing/gcc_s.rs:49
1: std::sys_common::backtrace::_print
at libstd/sys_common/backtrace.rs:71
2: std::panicking::default_hook::{{closure}}
at libstd/sys_common/backtrace.rs:59
at libstd/panicking.rs:380
3: std::panicking::default_hook
at libstd/panicking.rs:396
4: std::panicking::rust_panic_with_hook
at libstd/panicking.rs:576
5: std::panicking::begin_panic
6: rustc::session::opt_span_bug_fmt::{{closure}}
7: rustc::ty::context::tls::with_opt
8: rustc::session::opt_span_bug_fmt
9: rustc::session::bug_fmt
10: rustc::session::Session::incr_comp_session_dir
11: rustc_incremental::persist::fs::garbage_collect_session_directories
12: rustc_driver::driver::phase_2_configure_and_expand_inner::{{closure}}
13: rustc_driver::driver::phase_2_configure_and_expand_inner
14: rustc_driver::driver::compile_input
15: rustc_driver::run_compiler
``` | I-ICE,T-compiler,A-incr-comp,C-bug | medium | Critical |
312,266,576 | godot | [TRACKER] Clean up and reformat C# source code | **Godot version:**
`master` / 9e7cee2ce
**OS/device including version:**
_Manjaro KDE Linux 17.1_
**Issue description:**
As discussed in godotengine/godot-docs#1296, I'm planning to do a mass clean up of existing C# source files(`/modules/mono/glue/cs_files/*.cs`) by using a code inspection tool.
The purpose is to improve code quality and make it more consistent with the standard code style and naming convention of the language.
The initial result reports 1,604 issues, so it'd be obviously dangerous do an automatic clean up. So, I'm going to list them by categories here, and incrementally fix them in case by case basis:
- [ ] Code Style (530 / 839 remaining).
- [ ] Potential Code Quality Issues (129 / 140 remaining).
- [ ] Common Practices and Code Improvements (97 / 97 remaining).
- [ ] Language Usage Opportunities (69 / 66 remaining, 3 more found).
- [x] Redundancies in Code (0 / 83 remaining).
- [ ] Constraints Violations (71 / 72 remaining).
- [ ] Redundancies in Symbol Declarations (313 / 307 remaining, 6 more found).
Total 1,209 / 1,604 issues found. | enhancement,tracker,topic:dotnet | low | Major |
312,292,752 | pytorch | [caffe2] How to freeze a layer? | Hi, I'm working on caffe2.
I want to know how to freeze a certain layer, which means I don`t want to update a parameter (weight, bias) of a certain conv layer. How can I implement this feature? I found Stopgradient, Negategradient, Zerogradient in caffe2 doc, however none of them seems useful for me. | caffe2 | low | Major |
312,314,662 | flutter | Add details to Canvas.transform API doc, specifically format of the matrix | This is necessary because the matrix used in Canvas is 4x4 whereas in Skia (and SVG) it is 3x3.
Something like this would do it:
https://developer.mozilla.org/en-US/docs/Web/SVG/Attribute/transform | framework,engine,d: api docs,P3,team-engine,triaged-engine | low | Minor |
312,316,855 | rust | Compiler allows mutating immutable variables in unreachable code | IMO, this should give a compiler error:
https://play.rust-lang.org/?gist=13095cf519ab250005688bf214e2482f&version=stable
```rust
fn main() {
let x = 1;
return;
x += 2;
}
``` | C-enhancement,A-diagnostics,T-compiler | low | Critical |
312,324,572 | pytorch | [caffe2] Prefetching blobs for memory optimization | Hi,
I am wondering if there is a way to free, prefetch blobs and manually feed the outputs of layers to one another so that I am able to optimize memory use on GPUs.
Moreover, I am currently trying to implement something like the vDNN (virtualized Deep Neural Network) mentioned in this paper [https://arxiv.org/pdf/1602.08124.pdf](url). Is this a possibility in caffe2? | caffe2 | low | Minor |
312,329,168 | vue | add `options` property to the render context of functional component | ### What problem does this feature solve?
Custom properties in the options of functional component can't be accessed easily.
### What does the proposed API look like?
In the render function of functional component, options can be accessed by `context.options`, just like `vm.$options`(https://vuejs.org/v2/api/index.html#vm-options)
related issue #7492
<!-- generated by vue-issues. DO NOT REMOVE --> | feature request | medium | Major |
312,337,077 | go | cmd/compile: unexpected memequal call in short string comparison | (Using Go tip, 8818b4d27)
Compiling this code on amd64,
```go
type BoxType [4]byte
func (t BoxType) EqualString(s string) bool {
return string(t[:]) == s
}
```
I would expect that to compile to something pretty minimal, since the memory comparison must always be 4 bytes, but instead I see a call to memequal:
```
"".BoxType.EqualString STEXT size=103 args=0x20 locals=0x28
0x0000 00000 (./bmff/bmff.go:62) TEXT "".BoxType.EqualString(SB), $40-32
0x0000 00000 (./bmff/bmff.go:62) MOVQ (TLS), CX
0x0009 00009 (./bmff/bmff.go:62) CMPQ SP, 16(CX)
0x000d 00013 (./bmff/bmff.go:62) JLS 96
0x000f 00015 (./bmff/bmff.go:62) SUBQ $40, SP
0x0013 00019 (./bmff/bmff.go:62) MOVQ BP, 32(SP)
0x0018 00024 (./bmff/bmff.go:62) LEAQ 32(SP), BP
0x001d 00029 (./bmff/bmff.go:62) FUNCDATA $0, gclocalsΒ·a20105803dd226ab8faa525f9ceddf12(SB)
0x001d 00029 (./bmff/bmff.go:62) FUNCDATA $1, gclocalsΒ·69c1753bd5f81501d95132d08af04464(SB)
0x001d 00029 (./bmff/bmff.go:63) MOVQ "".s+64(SP), AX
0x0022 00034 (./bmff/bmff.go:63) CMPQ AX, $4
0x0026 00038 (./bmff/bmff.go:63) JEQ 56
0x0028 00040 (./bmff/bmff.go:63) XORL AX, AX
0x002a 00042 (./bmff/bmff.go:63) MOVB AL, "".~r1+72(SP)
0x002e 00046 (./bmff/bmff.go:63) MOVQ 32(SP), BP
0x0033 00051 (./bmff/bmff.go:63) ADDQ $40, SP
0x0037 00055 (./bmff/bmff.go:63) RET
0x0038 00056 (./bmff/bmff.go:63) LEAQ "".t+48(SP), AX
0x003d 00061 (./bmff/bmff.go:63) MOVQ AX, (SP)
0x0041 00065 (./bmff/bmff.go:63) MOVQ "".s+56(SP), AX
0x0046 00070 (./bmff/bmff.go:63) MOVQ AX, 8(SP)
0x004b 00075 (./bmff/bmff.go:63) MOVQ $4, 16(SP)
0x0054 00084 (./bmff/bmff.go:63) PCDATA $0, $1
0x0054 00084 (./bmff/bmff.go:63) CALL runtime.memequal(SB)
0x0059 00089 (./bmff/bmff.go:63) MOVBLZX 24(SP), AX
0x005e 00094 (./bmff/bmff.go:63) JMP 42
0x0060 00096 (./bmff/bmff.go:63) NOP
0x0060 00096 (./bmff/bmff.go:62) PCDATA $0, $-1
0x0060 00096 (./bmff/bmff.go:62) CALL runtime.morestack_noctxt(SB)
0x0065 00101 (./bmff/bmff.go:62) JMP 0
```
Ideally we'd just do a 4 byte (potentially unaligned) load from the string into a register and compare with the receiver?
/cc @josharian @randall77 @rasky | Suggested,Performance,help wanted,NeedsFix,compiler/runtime | low | Major |
312,375,539 | godot | Watch for edited resources in editor | So I am writing a C++ plugin for godot. The user assigns a material to my custom Spatial node. Now I want to be notified when this material is changed in the editor.
It seems the add_change_receptor API nor the CoreStringNames::get_singleton()->changed signal are emitted when say the materials albedo colour changes for example.
So my next avenue was to try to connect to the property editor "variant_changed" signal, but there seems to be no way to determine what object the Property editor is editing.
I would like some way to watch when the material is modified.
| enhancement,topic:editor,topic:plugin | low | Major |
312,392,104 | rust | libproc_macro: get crate for span or support $crate metavar in quote!{} | [One Reddit user was asking how to reference the invoking crate in a proc-macro](https://www.reddit.com/r/rust/comments/88xvgg/hey_rustaceans_got_an_easy_question_ask_here/dx0kp1f/), suggesting that `quote!{}` should support the `$crate` metavar. I don't think this is too unreasonable since `$` is already used for unquoting variables but `crate` couldn't be a user variable since it's a keyword.
Alternately, I think `Span` could have a method telling us what crate it comes from as a `String`.
cc @alexcrichton @dtolnay | T-compiler,C-feature-request,A-macros-1.2 | low | Minor |
312,412,631 | TypeScript | Documentation: Clarify declaration files and the declare keyword | **TypeScript Version:** 2.8.0
**Search Terms:**
document declaration files
** Issue**
It doesn't look like you actually define what a declaration file is in your documentation. For instance, your introduction to these at https://www.typescriptlang.org/docs/handbook/declaration-files/introduction.html tells me you're going to help me learn how to write them, but not what they are or why I'd want one. But, I don't think this makes it easy for a newbie to understand.
Similarly, you don't seem to explain when the `declare` keyword is, or when it should be used (as far as I can tell, it is used when you need to tell the compiler that the module/library/whatever has a value of some type).
You also seem to make some references to the global namespace (`declare global { ... }`)), but don't seem to directly explain it anywhere.
**Code**
n/a
**Expected behavior:**
n/a
**Actual behavior:**
n/a
**Playground Link:**
n/a
**Related Issues:**
n/a | Docs | low | Minor |
312,447,800 | rust | Tracking issue for RFC 2295, "Extend Pattern API to OsStr" | This is a tracking issue for the RFC "Extend Pattern API to OsStr" (rust-lang/rfcs#2295).
**Steps:**
- [ ] Implement the RFC (cc @kennytm)
- [ ] Adjust documentation ([see instructions on forge][doc-guide])
- [ ] Stabilization PR ([see instructions on forge][stabilization-guide])
[stabilization-guide]: https://forge.rust-lang.org/stabilization-guide.html
[doc-guide]: https://forge.rust-lang.org/stabilization-guide.html#updating-documentation
**Unresolved questions:**
None
<!-- TRIAGEBOT_START -->
<!-- TRIAGEBOT_ASSIGN_START -->
<!-- TRIAGEBOT_ASSIGN_DATA_START$${"user":"pitaj"}$$TRIAGEBOT_ASSIGN_DATA_END -->
<!-- TRIAGEBOT_ASSIGN_END -->
<!-- TRIAGEBOT_END --> | A-FFI,B-RFC-approved,T-libs-api,C-tracking-issue,Libs-Tracked | low | Major |
312,453,602 | rust | Tracking issue for RFC 2102, "Unnamed fields of struct and union type" | This is a tracking issue for the RFC "Unnamed fields of struct and union type" (rust-lang/rfcs#2102).
The feature gate for the issue is `#![feature(unnamed_fields)]`.
### About tracking issues
Tracking issues are used to record the overall progress of implementation.
They are also used as hubs connecting to other relevant issues, e.g., bugs or open design questions.
A tracking issue is however *not* meant for large scale discussion, questions, or bug reports about a feature.
Instead, open a dedicated issue for the specific matter and add the relevant feature gate label.
### Steps
- [ ] Implement the RFC (see https://github.com/rust-lang/rust/issues/49804#issuecomment-798768285 for mentoring instructions)
- [ ] Adjust documentation ([see instructions on forge][doc-guide])
- [ ] Formatting for new syntax has been added to the [Style Guide] ([nightly-style-procedure])
- [ ] Stabilization PR ([see instructions on forge][stabilization-guide])
[stabilization-guide]: https://forge.rust-lang.org/stabilization-guide.html
[doc-guide]: https://forge.rust-lang.org/stabilization-guide.html#updating-documentation
[Style Guide]: https://github.com/rust-lang/rust/tree/master/src/doc/style-guide
[nightly-style-procedure]: https://github.com/rust-lang/style-team/blob/master/nightly-style-procedure.md
### Unresolved questions
*None so far.*
### Implementation history
* #115131
* #115367
* #121198
<!-- TRIAGEBOT_START -->
<!-- TRIAGEBOT_ASSIGN_START -->
<!-- TRIAGEBOT_ASSIGN_DATA_START$${"user":"frank-king"}$$TRIAGEBOT_ASSIGN_DATA_END -->
<!-- TRIAGEBOT_ASSIGN_END -->
<!-- TRIAGEBOT_END --> | B-RFC-approved,T-lang,E-help-wanted,C-tracking-issue,S-tracking-impl-incomplete,S-tracking-design-concerns,F-unnamed_fields | medium | Critical |
312,522,143 | vue | The values of some attributes are replaced with the attribute name, which breaks MathML | ### Version
2.5.16
### Reproduction link
[https://gist.github.com/VelizarHristov/edaf4cf0d7b9f90c1e5d6070c027237f](https://gist.github.com/VelizarHristov/edaf4cf0d7b9f90c1e5d6070c027237f)
### Steps to reproduce
Open the HTML in a browser (Internet connection is required to fetch the libraries).
### What is expected?
[] should be displayed
### What is actually happening?
open] is displayed
---
It comes from this line of code: https://github.com/vuejs/vue/blob/aa1356e83de1112660e7a88ff955f49d64bb5b1f/src/platforms/web/server/modules/attrs.js#L49
The list of rewritten attributes is here: https://github.com/vuejs/vue/blob/7116af4e07520040ed7328c39d0a456808bfe1e1/src/platforms/web/util/attrs.js#L22
This is a general problem when including a block of XML in the HTML.
<!-- generated by vue-issues. DO NOT REMOVE --> | improvement | medium | Minor |
312,534,678 | go | x/sys/windows: Windows Service does not stop | Please answer these questions before submitting your issue. Thanks!
### What version of Go are you using (`go version`)?
go version go1.10.1 windows/amd64
### Does this issue reproduce with the latest release?
Yes
### What operating system and processor architecture are you using (`go env`)?
set GOARCH=amd64
set GOBIN=
set GOCACHE=C:\Users\azret\AppData\Local\go-build
set GOEXE=.exe
set GOHOSTARCH=amd64
set GOHOSTOS=windows
set GOOS=windows
set GOPATH=C:\Users\azret\go
set GORACE=
set GOROOT=C:\Go
set GOTMPDIR=
set GOTOOLDIR=C:\Go\pkg\tool\windows_amd64
set GCCGO=gccgo
set CC=gcc
set CXX=g++
set CGO_ENABLED=1
set CGO_CFLAGS=-g -O2
set CGO_CPPFLAGS=
set CGO_CXXFLAGS=-g -O2
set CGO_FFLAGS=-g -O2
set CGO_LDFLAGS=-g -O2
set PKG_CONFIG=pkg-config
set GOGCCFLAGS=-m64 -mthreads -fmessage-length=0 -fdebug-prefix-map=C:\Users\azret\AppData\Local\Temp\go-build856806762=/tmp/go-build -gno-record-gcc-switches
### What did you do?
https://github.com/golang/sys/blob/master/windows/svc/example/service.go in Line 42 I fixed a bug - the service is not stopping but just exits from the loop.
If possible, provide a recipe for reproducing the error.
A complete runnable program is good.
A link on play.golang.org is best.
### What did you expect to see?
I expected the Windows Service to stop.
### What did you see instead?
It did not stop and even did not stop when I tried to force stop it from Service Management Tool.
| OS-Windows,NeedsInvestigation,compiler/runtime | low | Critical |
312,595,977 | pytorch | [utils.bottleneck] List of improvements | Leaving this here as a reminder to self.
Things to add:
- [ ] Add sample output to docs
- [ ] Clarify CUDA profiling gotchas in docs
- [x] Test stability: If user isn't using `python` (ie, if they're using `python3`), the tests will fail.
- [ ] Use average keys (if not already using) | todo,triaged,module: bottleneck | low | Minor |
312,629,355 | go | x/build/cmd/gopherbot: βcongratulationsβ message posted long after a change was abandoned | See https://go-review.googlesource.com/c/sys/+/103858#message-7f834fe611a18c00129b9fef17acee53ecf05a6a
The βcongratsβ message is posted April 9, but it was abandoned April 4.
/cc @kevinburke | Builders,NeedsInvestigation | low | Minor |
312,642,829 | go | types2: type-check `syntax` nodes directly (don't use types2 maps) | [edit: By setting type information directly in `syntax` nodes rather than storing in `types2` maps, we should be able to recover some of the compiler speed loss of Go 1.18.]
We should move type-checking to operate directly on the new syntax nodes. Placeholder issue.
Set for Go1.12 so it stays on the radar. Unclear yet if that is a realistic goal.
| NeedsInvestigation | low | Major |
312,647,602 | opencv | Fails to compile "hal_internal.cpp" | ##### System information (version)
- OpenCV => 3.4.1
- Operating System / Platform => CentOs & Redhat 7.4 64 Bit
- Compiler => g++ 5.5
- Compilation by Linuxbrew with all default settings
##### Detailed description
Compilation fails when reaches "hal_internal.cpp".
Note, two "complex" header files are included in "hal_internal.cpp":
"complex.h" and "complex"
```
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp: In instantiation of 'int lapack_gemm_c(const fptype*, size_t, const fptype*, size_t, fptype, const fptype*, size_t, fptype, fptype*, size_t, int, int, int, int) [with fptype = float; size_t = long unsigned int]':
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:592:119: required from here
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:440:9: warning: unused variable 'ldsrc3' [-Wunused-variable]
int ldsrc3 = (int)(src3_step / sizeof(std::complex<fptype>));
^
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:435:98: warning: unused parameter 'alpha' [-Wunused-parameter]
lapack_gemm_c(const fptype *src1, size_t src1_step, const fptype *src2, size_t src2_step, fptype alpha,
^
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp: In instantiation of 'int lapack_gemm_c(const fptype*, size_t, const fptype*, size_t, fptype, const fptype*, size_t, fptype, fptype*, size_t, int, int, int, int) [with fptype = double; size_t = long unsigned int]':
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:599:119: required from here
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:440:9: warning: unused variable 'ldsrc3' [-Wunused-variable]
int ldsrc3 = (int)(src3_step / sizeof(std::complex<fptype>));
^
copy_matrix((std::complex<fptype>*)src3, ldsrc3, (std::complex<fptype>*)dst, lddst, c_m, c_n);
^
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:497:31: **error**: expected ')' before '_Complex'
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:502:25: **error**: expected unqualified-id before '_Complex'
set_value((std::complex<fptype>*)dst, lddst, std::complex<fptype>(0.0, 0.0), d_m, d_n);
^
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:502:25: **error**: expected ')' before '_Complex'
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:505:106: **error**: 'cAlpha' was not declared in this scope
cblas_cgemm(CblasRowMajor, transA, transB, a_m, d_n, a_n, (float*)reinterpret_cast<fptype(&)[2]>(cAlpha), (float*)src1, ldsrc1, (float*)src2, ldsrc2, (float*)reinterpret_cast<fptype(&)[2]>(cBeta), (float*)dst, lddst);
^
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:505:198: **error**: 'cBeta' was not declared in this scope
cblas_cgemm(CblasRowMajor, transA, transB, a_m, d_n, a_n, (float*)reinterpret_cast<fptype(&)[2]>(cAlpha), (float*)src1, ldsrc1, (float*)src2, ldsrc2, (float*)reinterpret_cast<fptype(&)[2]>(cBeta), (float*)dst, lddst);
^
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:507:107: **error**: 'cAlpha' was not declared in this scope
cblas_zgemm(CblasRowMajor, transA, transB, a_m, d_n, a_n, (double*)reinterpret_cast<fptype(&)[2]>(cAlpha), (double*)src1, ldsrc1, (double*)src2, ldsrc2, (double*)reinterpret_cast<fptype(&)[2]>(cBeta), (double*)dst, lddst);
^
/tmp/opencv-20180409-18036-14evg85/opencv-3.4.1/modules/core/src/hal_internal.cpp:507:202: **error**: 'cBeta' was not declared in this scope
cblas_zgemm(CblasRowMajor, transA, transB, a_m, d_n, a_n, (double*)reinterpret_cast<fptype(&)[2]>(cAlpha), (double*)src1, ldsrc1, (double*)src2, ldsrc2, (double*)reinterpret_cast<fptype(&)[2]>(cBeta), (double*)dst, lddst);
^
[ 38%] Building CXX object modules/core/CMakeFiles/opencv_core.dir/src/kmeans.cpp.o
``` | bug,category: core,category: build/install | low | Critical |
312,647,756 | pytorch | Examine contiguity requirements for gradInput | For THNN updateGradInput functions, am not 100% sure if gradInput is always initialized as an empty tensor.
cc @li-roy | triaged | low | Major |
312,683,457 | pytorch | [caffe2] Double precision for operators? | A lot of the operators are not defined for double precision. I am planning on experimenting with double precision calculations and will add them as necessary. Would there be interest in me submitting a PR request with them? Is there a reason they are not currently enabled? The only drawback is larger library file size and slower compilation time but I suppose not by much.
Thank you,
Svet | caffe2 | low | Major |
312,685,717 | godot | Cannot create more than 65536 PoolByteArrays | **Godot version:**
3.0.1
**OS/device including version:**
Windows 7, 64bit
**Issue description:**
Attempted to create a very large number of [PoolByteArrays](http://docs.godotengine.org/en/3.0/classes/class_poolbytearray.html) but as soon as more than 65536 (maximum for a 16 bit number) of them are created the game becomes unresponsive and crashes.
**Steps to reproduce:**
The following code crashes but if you change the variable `maxNum` to 65535 or less it works as expected.
```
extends Node
var testArr = []
var randNumber
var randNumberTwo
var maxNum = 65536 # <- reduce by one and this code works
func _ready():
for i in maxNum:
randNumber = randi() % 100
randNumberTwo = randi() % 200
var bytePool = PoolByteArray()
bytePool.push_back(randNumber)
bytePool.push_back(randNumberTwo)
testArr.push_back(bytePool)
print("Done!")
print(testArr.size())
```
**Minimal reproduction project:**
[PoolByteArrays.zip](https://github.com/godotengine/godot/files/1891791/PoolByteArrays.zip)
<!-- Recommended as it greatly speeds up debugging. Drag and drop a zip archive to upload it. -->
| bug,discussion,topic:core,confirmed | low | Critical |
312,712,481 | flutter | UX for finder errors in driver needs improvement | My team is frequently confused by errors that driver tests provide.
Example:
Let's say we have this line:
```
await driver.tap(find.byTooltip('Back'));
```
And we remove a screen that has the back button. The error we get is:
```
DriverError: Error in Flutter application: Timeout while executing tap: TimeoutException after 0:00:05.000000: Future not completed...
```
We could improve this error message by saying something like:
```
DriverError: Tap failed. Timed out after waiting 5 seconds to find tooltip with text "Back".
```
| a: tests,tool,framework,t: flutter driver,c: proposal,a: error message,P2,team-framework,triaged-framework | low | Critical |
312,743,358 | angular | asynchronous providers | <!--
PLEASE HELP US PROCESS GITHUB ISSUES FASTER BY PROVIDING THE FOLLOWING INFORMATION.
ISSUES MISSING IMPORTANT INFORMATION MAY BE CLOSED WITHOUT INVESTIGATION.
-->
## I'm submitting a...
<!-- Check one of the following options with "x" -->
<pre><code>
[ ] Regression (a behavior that used to work and stopped working in a new release)
[ ] Bug report <!-- Please search GitHub for a similar issue or PR before submitting -->
[ X] Feature request
[ ] Documentation issue or request
[ ] Support request => Please do not submit support request here, instead see https://github.com/angular/angular/blob/master/CONTRIBUTING.md#question
</code></pre>
## Current behavior
<!-- Describe how the issue manifests. -->
Angular doesn't seem to support providers which asynchronously create their object / service. For those who want to load config from a server dynamically, they seem to be encouraged to use a kludge comprised of creating a `ConfigService` and then using `APP_INITIALIZER` to delay the bootstrapping just long enough to make an async call and inject the results into the `ConfigService`. But, this is a hack, and is buggy with race conditions - the `ConfigService` may potentially be injected into other services, and be used / read from before this bootstrapping happens, causing bugs because the object gets used before it's ready.
Here's a plunk which demonstrates the problem: https://plnkr.co/edit/plWAT5i5BFtjastSdaVy?p=preview
Notice the problem is that another service will read from the `ConfigService` before it's truly finished being "constructed" IMO, since we need to construct the object, and then finish initializing it's state the `APP_INITIALIZER` step. I think reading config in a constructor is pretty common programming practice.
## Expected behavior
<!-- Describe what the desired behavior would be. -->
I think it would be great to be able to do something like:
```
export function asyncConfigServiceFactory(api: ApiService) {
return new Promise((resolve, reject) => {
api.get('my-config.json').then(response => {
resolve(new ConfigService(response));
});
});
}
@NgModule({
providers: [
ApiService,
{
provide: ConfigService,
useAsyncFactory: asyncConfigServiceFactory,
deps: [ApiService]
},
],
})
```
(notice the `useAsyncFactory` idea to tell ng that it should wait on the promise, and use the value it produces as the service)
I think that's a pretty straight forward way to do it, but whatever you think is the best way to address it.
## What is the motivation / use case for changing the behavior?
<!-- Describe the motivation or the concrete use case. -->
My use case is to fetch config from a server, and stick it into a service, and then inject this `ConfigService` into other services. I do not want to inject `ConfigService` into other stuff until it's fully constructed and ready to be used - in other words, until the asynchronous request to the server has completed and my `ConfigService` has all the values populated into it.
I've seen quite a few other users trying to accomplish the same thing on Stackoverflow. This kludge is becoming the defacto way to do it, and SO answers are getting highly upvoted, and multiple articles are being published on other websites - all recommending a bug-prone method. That's not good for the framework's reputation long term. It should be easy to do it "the right way", whatever that is.
## Environment
<pre><code>
Angular version: 5.1
<!-- Check whether this is still an issue in the most recent Angular version -->
Others:
<!-- Anything else relevant? Operating system version, IDE, package manager, HTTP server, ... -->
</code></pre>
Other issues that would be well served by async providers:
https://github.com/angular/angular/issues/19448
https://github.com/angular/angular/issues/20383 | feature,area: core,core: di,feature: under consideration | high | Critical |
312,744,166 | flutter | [a11y] contrast between caption and white to low | In the Cupertino theme the contrast between a white background and the caption text style is only 4.3. It needs to be 4.5 or better.
We currently use black54 for this, but according to spec grey700 (for a contrast ratio of 6) should be used. | framework,a: accessibility,f: cupertino,from: a11y review,P2,team-design,triaged-design | low | Major |
312,758,951 | rust | better error messages when documentation tests fail | Few days ago my code started to failing for some reason. After some research I found that everything wrapped in ticks (code) is tested by default (new feature? not failing on stable). Error message was not very helpful and I was wondering if there was a way to make error message nicer.
```
Doc-tests slime
Running `rustdoc --test /home/travis/build/jaroslaw-weber/slime/src/lib.rs --crate-name slime -L dependency=/home/travis/build/jaroslaw-weber/slime/target/debug/deps -L native=/home/travis/build/jaroslaw-weber/slime/target/debug/build/backtrace-sys-1971e506b73c6055/out/.libs -L dependency=/home/travis/build/jaroslaw-weber/slime/target/debug/deps --extern toml=/home/travis/build/jaroslaw-weber/slime/target/debug/deps/libtoml-bc45e8f131e8cc1d.rlib --extern slime=/home/travis/build/jaroslaw-weber/slime/target/debug/deps/libslime-7825b794b397c4d0.rlib --extern serde_json=/home/travis/build/jaroslaw-weber/slime/target/debug/deps/libserde_json-5ae15ac5c0b54e23.rlib --extern failure=/home/travis/build/jaroslaw-weber/slime/target/debug/deps/libfailure-b4d75b472e3bd251.rlib --extern fs_extra=/home/travis/build/jaroslaw-weber/slime/target/debug/deps/libfs_extra-b486394ae399718d.rlib --extern handlebars=/home/travis/build/jaroslaw-weber/slime/target/debug/deps/libhandlebars-f60000f1622466b3.rlib --extern serde=/home/travis/build/jaroslaw-weber/slime/target/debug/deps/libserde-aaecc1e461c12055.rlib`
running 1 test
test src/lib.rs - Slime (line 25) ... FAILED
failures:
---- src/lib.rs - Slime (line 25) stdout ----
error: expected one of `!`, `.`, `::`, `;`, `?`, `{`, `}`, or an operator, found `new`
--> src/lib.rs:26:7
|
3 | cargo new --bin crate_name
| ^^^ expected one of 8 possible tokens here
error[E0425]: cannot find value `cargo` in this scope
--> src/lib.rs:26:1
|
3 | cargo new --bin crate_name
| ^^^^^ not found in this scope
thread 'src/lib.rs - Slime (line 25)' panicked at 'couldn't compile the test', librustdoc/test.rs:306:13
note: Run with `RUST_BACKTRACE=1` for a backtrace.
```
It would be nice to have some kind of explanation:
```
Documentation test have failed. Every code block (wrapped with ticks) in documentation is tested. To not test this code of block use "ignore" after the ticks. More info at: https://doc.rust-lang.org/stable/rustdoc/documentation-tests.html . Details about failed tests:
```
| T-rustdoc,C-enhancement,A-diagnostics,A-doctests | low | Critical |
312,794,454 | vscode | Expose `glyphMargin` support to extensions | It would be great if extension were able to use the built-in glyphMargin decorations -- i.e. clickable decorations in the gutter. | feature-request,api | medium | Major |
312,887,074 | kubernetes | kubelet should track tcp_mem stats also along with cpu/ram/disk | /kind feature
/sig node
**What happened**:
A program started leaking TCP memory, which filled up the node's TCP stack memory. The network performance on the node degraded and connections to pods running on the node either times out or will hang for a long time.
Node's `dmesg` had lines mentioning `TCP: out of memory -- consider tuning tcp_mem`
Further reading and investigation reveals that this could happen when TCP stack runs out of memory pages allocated by kernel or when there are lot of orphaned/open sockets.
TCP stack limits: max 86514
```bash
$ cat /proc/sys/net/ipv4/tcp_mem
43257 57676 86514
# min pressure max
```
Usage when issue happened: mem 87916
```
$ cat /proc/net/sockstat
sockets: used 1386
TCP: inuse 24 orphan 0 tw 58 alloc 863 mem 87916
UDP: inuse 3 mem 3
UDPLITE: inuse 0
RAW: inuse 0
FRAG: inuse 0 memory 0
```
kubelet posts node status as ready.
**What you expected to happen**:
kubelet should say node is not ready.
It would be great if `kubelet` could track the `tcp_mem` stats also along with CPU/RAM/disk as network is also an important factor. If `tcp_mem` limit is hit, for some reason, the node is not usable. Notifying the user that node has some issue can help debugging and further identifying the cause.
**How to reproduce it (as minimally and precisely as possible)**:
1. Create a GKE cluster
2. Label a node to carry out tests
```bash
$ kubectl label node <node-name> node=leak-test
```
3. Create an nginx deployment with Loadbalancer, with can serve a large file
```bash
$ kubectl create -f https://raw.githubusercontent.com/shahidhk/k8s-tcp-mem-leak/master/nginx.yaml
```
4. Check if you can download the large file
```bash
$ curl -o large-file <ip>/large-file
```
5. Create a deployment that can fill up the TCP stack memory
```bash
$ kubectl create -f https://raw.githubusercontent.com/shahidhk/k8s-tcp-mem-leak/master/leak-repro.yaml
```
6. SSH into the node and observe `cat /proc/sys/net/ipv4/tcp_mem` and `cat /proc/net/sockstat` and scale the deployment until the current mem exceeds the limit
7. Try downloading the large file again. It will either become very slow or will not happen at all
**Anything else we need to know?**:
This is more of a feature request for kubelet rather than a bug. TCP mem can get filled if the node is running a lot of TCP heavy workloads, need not necessarily be a leak. Since kubelet is ultimately responsible for reporting node's health, network should also be a parameter.
**Environment**:
- Kubernetes version (use `kubectl version`): `v1.9.6-gke.0`
- Cloud provider or hardware configuration: GKE, 1 node, n1-standard-1
- OS (e.g. from /etc/os-release): Container-Optimized OS
- Kernel (e.g. `uname -a`): 4.4.111+
- Install tools: -
- Others: -
| priority/backlog,sig/network,sig/node,kind/feature,triage/accepted | high | Critical |
312,891,127 | rust | Error message about immutable binding is not actually caused by a binding | See the following code:
```rust
use std::collections::HashMap;
struct Test {
v: u32,
}
fn main() {
let mut map = HashMap::new();
map.insert("a", Test { v: 0 });
let values = map.values();
for mut t in values {
t.v += 1;
}
}
```
It raises this compiler error:
> error[E0594]: cannot assign to field `t.v` of immutable binding
--> /tmp/test.rs:14:9
|
14 | t.v += 1;
| ^^^^^^^^ cannot mutably borrow field of immutable binding
The fix is to use `values_mut()` instead of `values()`.
But the error message is confusing as the problem here is not the binding as `t` itself is `mut` making it harder to debug for newcomers.
| C-enhancement,A-diagnostics,P-medium,T-compiler,E-medium | low | Critical |
312,949,404 | pytorch | [caffe2] ChannelShuffle example | Is there any example for the ChannelShuffle operator?
There is no doc right now.
I have a stack of v rgba images, where the alpha contain a mask. I would like to transform from:
rgbargbargba to rgbrgbrgbaaa so that my convolutions do not get messed up by the alpha.
Thanks, | caffe2,module: docs | low | Minor |
312,952,944 | go | runtime: use register instead of TLS for G pointer for darwin/arm and darwin/arm64 | Need to decide how to handle TLS on darwin/arm. On x86, TLS was moved from a pthread_get/setspecific implementation to a reserved (by Apple) slot in the TLS (see #23617). That's not possible for arm/arm64 so we still use the old method. Since these archs have more registers, we should probably just reserve a register for the G pointer.
Reserving a register might invalidate some existing assembly code. Kind of a wrench in the plan, need to investigate how bad it might be.
Or maybe someone can think of a better solution.
We might need a solution before we can do #17490 for arm.
| OS-Darwin,compiler/runtime | low | Major |
313,026,231 | flutter | Improve error message for CocoaPods error "The sandbox is not in sync with the Podfile.lock" | See the transcript below for full details, but briefly:
flutter run on examples/flutter_view failed with this error message:
```
error: The sandbox is not in sync with the Podfile.lock. Run 'pod install' or update your CocoaPods installation.
```
`pod install` in the current directory fails. You appear to need to `cd ios; pod install`.
It would be more helpful if the error message could indicate that in some way.
```
leafp-macbookpro:flutter_view leafp$ ../../bin/flutter run
Launching lib/main.dart on iPhone X in debug mode...
Running Xcode clean... 1.0s
Starting Xcode build...
Xcode build done. 1.0s
Failed to build iOS app
Error output from Xcode build:
β³
** BUILD FAILED **
Xcode's output:
β³
=== BUILD TARGET Runner OF PROJECT Runner WITH CONFIGURATION Debug ===
Debug.xcconfig line 2: Unable to find included file "Pods/Target Support Files/Pods-Runner/Pods-Runner.debug.xcconfig"
Debug.xcconfig line 2: Unable to find included file "Pods/Target Support Files/Pods-Runner/Pods-Runner.debug.xcconfig"
=== BUILD TARGET Runner OF PROJECT Runner WITH CONFIGURATION Debug ===
diff: /Podfile.lock: No such file or directory
diff: /Manifest.lock: No such file or directory
error: The sandbox is not in sync with the Podfile.lock. Run 'pod install' or update your CocoaPods installation.
Could not build the application for the simulator.
Error launching application on iPhone X.
leafp-macbookpro:flutter_view leafp$ pod install
[!] No `Podfile' found in the project directory.
leafp-macbookpro:flutter_view leafp$ ls
README.md assets ios pubspec.lock
android build lib pubspec.yaml
leafp-macbookpro:flutter_view leafp$ cd ios/
leafp-macbookpro:ios leafp$ ls
Flutter Podfile Runner.xcodeproj ServiceDefinitions.json
Frameworks Runner Runner.xcworkspace
leafp-macbookpro:ios leafp$ pod install
Analyzing dependencies
Downloading dependencies
Installing MaterialControls (1.2.2)
Generating Pods project
Integrating client project
Sending stats
Pod installation complete! There is 1 dependency from the Podfile and 1 total pod installed.
``` | tool,t: xcode,P2,a: plugins,team-tool,triaged-tool | low | Critical |
313,042,459 | go | x/build/cmd/relui: occasionally see TLS timeouts when fetching source archive | When recently building on windows-amd64 and also during the 1.10.1 release:
```
2018/04/10 14:14:48 windows-amd64: Error: 500 Internal Server Error; body: fetching URL https://go.googlesource.com/go/+archive/master.tar.gz: Get https://go.googlesource.com/go/+archive/master.tar.gz: net/http: TLS handshake timeout
```
@bradfitz @FiloSottile | Builders,NeedsInvestigation | low | Critical |
313,093,530 | rust | compiletest: compare-mode cannot handle mixed success + failure | Sub-issue of #48879
Currently, if you have a ui test `src/test/ui/foo.rs` that signals a diagnostic error under the AST borrowck, but has no error under the NLL borrowck, then there is no way to make a `foo.nll.stderr` or `foo.nll.stdout` such that running `compiletest --compare-mode=nll` will accept it.
The reason: `compiletest` complains that the compilation itself was successful, and by default, all ui tests are assumed to signal a compilation failure.
The usual way such a case (of a test that compiles without an error) is handled in `compiletest` is to add a comment of the form `// run-pass` in the test itself. But we do not want the tests to be marked as always passing; they *only* pass under the special mode (NLL in this case) that is still under development in the compiler itself.
There are a number of reasonable ways of addressing this.
1. We could modify such tests to ensure that ~~they end with a function~~ *their `fn main`* is tagged with `#[rustc_error]`, so that they will *never* successfully compile.
2. We could attempt to have some way to encode a mode-predicated run-pass marker, e.g. `// [nll] run-pass`.
3. We could have `compiletest` *infer* that a test, when running under compare-mode, must have an implicit `// run-pass` marker, if there exists a `foo.mode.stdout` file (even an empty one).
I'm not such a big fan of options 1 or 2 because I like the fact that (so far) compare-mode does not have to impact the existing `.rs` nor `.stderr`/`.stdout` files; you just add your alternative `.nll.stderr` file and no one else is the wiser.
So currently I would like us to adopt option 3 (or some other option I haven't thought of yet) that allows the `.rs`/`.stderr`/`.stdout` to go on oblivious to the existence of compare-mode.
**But**, in the short term, to let us get work done, I plan to follow option 1 for all existing `ui/` tests. (This is intended to be a **temporary hack**; all such uses of this hack should point back to this issue.) | A-testsuite,C-enhancement,T-bootstrap,A-compiletest,E-needs-design,E-needs-investigation | low | Critical |
313,120,139 | flutter | Text overflow handling is inconsistent and sometimes wrong. | It looks like our overflow handling is pretty weird (if not just downright broken).
For instance, the following doesn't result in any overflow handling other than "clip", regardless of the overflow mode:
```dart
new Text(
'''Lorem ipsum dolor sit amet, consectetur adipiscing elit.\n Phasellus sodales eros nec tortor maximus molestie.\nSed metus metus, viverra vel libero sed, commodo consequat ipsum. Sed commodo ultricies ligula, sed ultricies diam dignissim vitae. ''',
softWrap: true,
maxLines: 2,
textAlign: TextAlign.start,
overflow: TextOverflow.ellipsis,
),
```
I would expect that since there is more text to be displayed than the `maxLines`, that we do show the ellipsis at the end of the second line.
Also, if I remove the hard newlines, then the ellipsis does appear at the end of the second line, which is correct, but inconsistent. And if I then switch to overflow to fade, it fades the first line and the second, instead of just the second. | framework,a: typography,has reproducible steps,P2,found in release: 3.7,found in release: 3.9,team-framework,triaged-framework | low | Critical |
313,167,081 | angular | Parent ngIf not respecting child's animation to finish before parent is removed from DOM | <!--
PLEASE HELP US PROCESS GITHUB ISSUES FASTER BY PROVIDING THE FOLLOWING INFORMATION.
ISSUES MISSING IMPORTANT INFORMATION MAY BE CLOSED WITHOUT INVESTIGATION.
-->
## I'm submitting a...
<!-- Check one of the following options with "x" -->
<pre><code>
[ ] Regression (a behavior that used to work and stopped working in a new release)
[X] Bug report <!-- Please search GitHub for a similar issue or PR before submitting -->
[ ] Feature request
[ ] Documentation issue or request
[ ] Support request => Please do not submit support request here, instead see https://github.com/angular/angular/blob/master/CONTRIBUTING.md#question
</code></pre>
## Current behavior
Parent ngIf does not wait for child animation to finish before killing it, apparently only if child's animation does not use :leave.
Issue [#15798](https://github.com/angular/angular/issues/15798) and [#18305](https://github.com/angular/angular/issues/18305) are about the same thing. The user [dharders](https://github.com/dharders) said:
> I should have specified that animateChild() is needed only if you have :leave animations in a child component and the parent is removed from the DOM.
but I would like to have the usage of that method for custom states.
## Expected behavior
Parent ngIf waits for child's animation to finish before removing it form the DOM.
## Minimal reproduction of the problem with instructions
I have this for my child component
```
animations: [
trigger('slideUpDown', [
state('hide', style({transform: 'translateY(100%)'})),
state('show', style({transform: 'translateY(0)'})),
transition('hide <=> show', animate('500ms ease-in-out'))
])
]
...
state: 'show'|'hide' = 'hide';
toggleView() {
this.state = this.state === 'show' ? 'hide' : 'show';
}
```
and for my parent component
- html:
```
<button (click)="toggle()">Click</button>
<div *ngIf="isVisible">
<child #childComponent></child>
</div>
```
- ts
```
@ViewChild('childComponent') childComponent: ChildComponent;
toggle() {
this.childComponent.toggleState();
}
```
## What is the motivation / use case for changing the behavior?
So child animations are respected when parent is about to be hidden.
## Environment
<pre><code>
Angular version: Angular-CLI 1.7.4 - "@angular/core": "^5.2.0"
<!-- Check whether this is still an issue in the most recent Angular version -->
Browser:
- [X] Chrome (desktop) version 65.0.3325.181
- [ ] Chrome (Android) version XX
- [ ] Chrome (iOS) version XX
- [ ] Firefox version XX
- [ ] Safari (desktop) version XX
- [ ] Safari (iOS) version XX
- [ ] IE version XX
- [ ] Edge version XX
For Tooling issues:
- Node version: v8.11.1
- Platform: Windows 10 64bit
Others:
</code></pre>
| type: bug/fix,area: animations,freq2: medium,P3 | medium | Critical |
313,220,452 | pytorch | [Caffe2] Boolean Tensor not supported for mobile_exporter.Export | The following sample results in an error.
# Code
```python
#!/usr/bin/env python
import onnx
import numpy as np
import caffe2.python.onnx.frontend as oc2f
import caffe2.python.predictor.predictor_exporter as pe
import caffe2.python.predictor.mobile_exporter as c2me
from caffe2.python import model_helper, workspace, brew, utils
IN_DATA_NAME = "in_data"
OUT_DATA_NAME = "out_data"
DATA_SHAPE = (1, 3, 3, 3)
model = model_helper.ModelHelper(name="not_net")
model.net.Not(IN_DATA_NAME, OUT_DATA_NAME)
val = (np.random.standard_normal(DATA_SHAPE) > 0).astype(np.bool)
model.param_init_net.GivenTensorBoolFill([], IN_DATA_NAME, values=val, shape=DATA_SHAPE)
workspace.RunNetOnce(model.param_init_net)
init_net, predict_net = c2me.Export(workspace, model.net, [IN_DATA_NAME])
value_info = {
}
onnx_model = oc2f.caffe2_net_to_onnx_model(
predict_net,
init_net,
value_info)
onnx.checker.check_model(onnx_model)
print(onnx_model)
with open("not.onnx", 'wb') as f:
f.write(onnx_model.SerializeToString())
```
# Error
```
Traceback (most recent call last):
File "./gen_not.py", line 33, in <module>
init_net, predict_net = c2me.Export(workspace, model.net, [IN_DATA_NAME])
File "/home/masato/repo/pytorch/build/caffe2/python/predictor/mobile_exporter.py", line 70, in Export
add_tensor(init_net, blob_name, blob)
File "/home/masato/repo/pytorch/build/caffe2/python/predictor/mobile_exporter.py", line 35, in add_tensor
kTypeNameMapper[blob.dtype],
KeyError: dtype('bool')
```
I am able to finish properly by adding the following line to mobile_exporter.py, but I'm not sure if this is the correct fix.
```diff
diff --git a/caffe2/python/predictor/mobile_exporter.py b/caffe2/python/predictor/mobile_exporter.py
index 07f88def0..9095f9c10 100644
--- a/caffe2/python/predictor/mobile_exporter.py
+++ b/caffe2/python/predictor/mobile_exporter.py
@@ -20,6 +20,7 @@ def add_tensor(net, name, blob):
np.dtype('int32'): "GivenTensorIntFill",
np.dtype('int64'): "GivenTensorInt64Fill",
np.dtype('uint8'): "GivenTensorStringFill",
+ np.dtype('bool'): "GivenTensorBoolFill",
}
shape = blob.shape
```
# Environment
- OS: Ubuntu 16.04.4 x86_64
- Python: 3.5.2
- Pytorch version: 2e156f3eabb6fa4ae88c2e831aea7da344cf43a3
- onnx version: b8c423889b3d15825e90fd5bd8fa1766ba2200ce
| caffe2 | low | Critical |
313,248,323 | scrcpy | Set envvar on device, so other tools can react to scrcpy | I use Tasker for device automation and other neat little gimmicks.
I wonder if it's possible to have scrcpy set an envvar when connecting to the device so I can have tasker do stuff like disabling display timeout and such. | feature request | low | Minor |
313,339,852 | vscode | Allow maximizing/restoring a single terminal or editor pane | I'd be surprised if this hasn't been asked before, but tmux supports a thing it calls "pane zooming", if I recall correctly.
Basically once you start splitting a window, vertically, horizontally, it doesn't matter, you press a certain keyboard shortcut and the current pane is "maximized", hiding all the other panes. Text is reflowed, so you actually see the former pane as a full window. This is incredibly useful once you launch some command that has a ton of unexpected output that you want to check out.
When you're done you press the same keyboard shortcut and the pane "reverts", showing the previous layout. The text in the pane is reflowed again to make it fit in the smaller pane size.
Very, very handy when you're moving through terminals quickly and you want to check out something. | feature-request,workbench-views | high | Critical |
313,346,255 | pytorch | [Caffe2] Onnx to Caffe2 Blob error | Trying to convert a model from PyTorch to Caffe2. I have the model in Onnx form, however when I try to go from Onnx to Caffe2 using the tool in caffe2.python.onnx.bin.conversion, I get an output but running them in Caffe2 produces:
'W0411 14:04:06.918320 13911 workspace.cc:174] Blob 1 not in the workspace.
WARNING:caffe2.python.workspace:Original python traceback for operator `1647504248` in network
torch-jit-export_predict^
0
1
25"Conv*
strides00*
dilations00*
group*
kernels00*
pads00003
56PadImage*
pads0000*
mode"reflect^
6
3
47"Conv*
strides00*
dilations00*
group*
kernels00*
pads0000
78"Relu:0:1:2:3:4B8` in exception above (most recent call last):
Traceback (most recent call last):
File "Caffe2WCT.py", line 32, in <module>
workspace.RunNetOnce(predict_net)
File "/usr/local/lib/python2.7/dist-packages/caffe2/python/workspace.py", line 199, in RunNetOnce
StringifyProto(net),
File "/usr/local/lib/python2.7/dist-packages/caffe2/python/workspace.py", line 178, in CallWithExceptionIntercept
return func(*args, **kwargs)
RuntimeError: [enforce fail at operator.cc:42] blob != nullptr. op Conv: Encountered a non-existing input blob: 1'
OS: Ubuntu 16.04
Caffe2: Built from source this morning
Python: 2.7
| caffe2 | low | Critical |
313,511,957 | flutter | Support drawing text along a curved path and on a mesh | I would like to be able to draw text on a non-linear path, e.g. like so: https://fiddle.skia.org/c/@draw_text_on_path
It doesn't appear to be possible to do that today - the `Paragraph` and `Canvas.drawParagraph()` classes/methods only appear to support drawing text baselined on a straight line (and paragraph in particular is heavily geared to laying out potentially multi-line text in straight lines).
As linked above, Skia has a method for drawing text on a path, but that seems to fall outside of the text/paragraph painting model used by Flutter and I'm not entirely sure it'd be appropriate for Flutter's design - however, it seems like it would be fairly easy to implement if it could fit in with the design.
It would be nice to have the ability to apply various styles to the text along the path, if that can be done cheaply (I have some doubts that it could be done cheaplym and I could live without it).
This would help to support #1831 and possibly #15501 | c: new feature,engine,a: typography,P3,team-engine,triaged-engine | medium | Critical |
313,516,892 | kubernetes | Realtime process inside container | /kind feature
I want to run RT process inside container as it is a requirement for my application. I am able to do this in docker/container using --cpu-rt-runtime=95000 --ulimit rtprio=99 but it doesn't work with k8s as I did not find a parameter to pass the same from YAML file.
**Environment**:
- Kubernetes version (use `kubectl version`):
Server Version: version.Info{Major:"1", Minor:"9", GitVersion:"v1.9.1+2.0.2.el7"
- OS (e.g. from /etc/os-release):
Oracle Linux Server release 7.4
- Kernel (e.g. `uname -a`):
4.1.12-112.14.15.el7uek.x86_64 #2 SMP Thu Feb 8 09:58:19 PST 2018 x86_64 x86_64 x86_64 GNU/Linux
- Install tools:
- Others:
| sig/node,kind/feature,lifecycle/frozen | medium | Minor |
313,518,114 | angular | Elements adds ng-version attribute to every Angular element on the page | Currently every top-level custom element tag gets a ng-version attribute
Ex.
```
...
<my-element name="Hello World!!" ng-version="6.0.0-rc.1" >
<div>Hello World!!</div>
</my-element>
...
```
For server side rendered custom elements this adds extra bytes for every custom element on the page. | type: bug/fix,freq3: high,area: server,area: elements,state: confirmed,P4 | low | Major |
313,520,880 | rust | Comparison of Option<NonZeroU*> is not fully optimized | Compare:
```rust
#![feature(nonzero)]
use std::num::NonZeroU32;
pub fn f(a: Option<NonZeroU32>, b: Option<NonZeroU32>) -> bool {
a < b
}
pub fn g(a: u32, b: u32) -> bool {
a < b
}
```
The functions `f` and `g` have equivalent effect, and should compile to the same code. However, the translated `f` is much more complex, involving needless comparison of the discriminant.
<details><summary>LLVM IR and ASM (in release mode)</summary>
```llvm
; playground::f
; Function Attrs: norecurse nounwind readnone uwtable
define zeroext i1 @_ZN10playground1f17hd628eff49d2ff60dE(i32, i32) unnamed_addr #0 {
start:
%2 = icmp ne i32 %0, 0
%3 = icmp ne i32 %1, 0
%4 = xor i1 %2, %3
br i1 %4, label %bb7.i, label %bb6.i
bb6.i: ; preds = %start
%5 = icmp ult i32 %0, %1
%spec.select = and i1 %2, %5
ret i1 %spec.select
bb7.i: ; preds = %start
%6 = xor i1 %2, true
%7 = and i1 %3, %6
ret i1 %7
}
; playground::g
; Function Attrs: norecurse nounwind readnone uwtable
define zeroext i1 @_ZN10playground1g17hea1c394facd055aeE(i32 %a, i32 %b) unnamed_addr #0 {
start:
%0 = icmp ult i32 %a, %b
ret i1 %0
}
```
```asm
playground::f:
test edi, edi
setne al
test esi, esi
setne cl
xor cl, al
je .LBB0_1
test esi, esi
setne cl
test edi, edi
sete al
and al, cl
ret
.LBB0_1:
test edi, edi
setne cl
cmp edi, esi
setb al
and al, cl
ret
playground::g:
cmp edi, esi
setb al
ret
```
</details>
Note that specializing `PartialOrd` etc for `Option<NonZero*>` is not a valid solution, since custom wrapper types of NonZero* will still have the same problem.
```rust
#[derive(PartialOrd, PartialEq)]
pub struct S(NonZeroU32);
pub fn f(a: Option<S>, b: Option<S>) -> bool {
a < b
}
```
cc #49137.
<!-- TRIAGEBOT_START -->
<!-- TRIAGEBOT_ASSIGN_START -->
<!-- TRIAGEBOT_ASSIGN_DATA_START$${"user":"veera-sivarajan"}$$TRIAGEBOT_ASSIGN_DATA_END -->
<!-- TRIAGEBOT_ASSIGN_END -->
<!-- TRIAGEBOT_END --> | A-LLVM,I-slow,C-enhancement,T-compiler,WG-llvm,C-optimization | medium | Critical |
313,537,743 | flutter | iOS double-tap reachability under Flutter | iOS has a reachability feature where you can double-tap the home button to shift the entire screen contents down within thumb reach: 
Under Flutter, the double-tapping works fine to shift the screen down, but unlike in other apps, the screen doesn't immediately shift back up once you tap to navigate within the app. Is there a way to push the screen back up when the Navigator changes routes?
I couldn't find any relevant Apple methods other than some private API in UIApplication https://github.com/nst/iOS-Runtime-Headers/blob/master/Frameworks/UIKit.framework/UIApplication.h#L323
Note that there does seem to be a timer which triggers the screen to shift up a few seconds following the last tap; however, in other apps the shift occurs immediately upon tapping/navigating to a different screen. | c: new feature,platform-ios,framework,a: accessibility,customer: mulligan (g3),P2,team-ios,triaged-ios | low | Minor |
313,554,536 | flutter | Flutter Doctor Error - SocketException: Write failed (OS Error: Broken pipe, errno = 32) |
I followed the "Get Started Tutorial" and when I run the βFlutter: New Projectβ show this message:
```
$ flutter doctor -v
[β] Flutter (Channel beta, v0.2.8, on Linux, locale en_US.UTF-8)
β’ Flutter version 0.2.8 at /home/raulino/.development/flutter
β’ Framework revision b397406561 (9 dias atrΓ‘s), 2018-04-02 13:53:20 -0700
β’ Engine revision c903c217a1
β’ Dart version 2.0.0-dev.43.0.flutter-52afcba357
-Unhandled exception:
SocketException: Write failed (OS Error: Broken pipe, errno = 32), port = 0
#0 _rootHandleUncaughtError.<anonymous closure> (dart:async/zone.dart:1114)
#1 _microtaskLoop (dart:async/schedule_microtask.dart:41)
#2 _startMicrotaskLoop (dart:async/schedule_microtask.dart:50)
#3 _Timer._runTimers (dart:isolate-patch/dart:isolate/timer_impl.dart:391)
#4 _Timer._handleMessage (dart:isolate-patch/dart:isolate/timer_impl.dart:416)
#5 _RawReceivePortImpl._handleMessage (dart:isolate-patch/dart:isolate/isolate_patch.dart:165)
```
Otherwise when I run the 'flutter doctor' in terminal runs normal like this:
```
$ flutter doctor -v
[β] Flutter (Channel beta, v0.2.8, on Linux, locale pt_BR.UTF-8)
β’ Flutter version 0.2.8 at /home/raulino/.development/flutter
β’ Framework revision b397406561 (9 dias atrΓ‘s), 2018-04-02 13:53:20 -0700
β’ Engine revision c903c217a1
β’ Dart version 2.0.0-dev.43.0.flutter-52afcba357
[β] Android toolchain - develop for Android devices (Android SDK 27.0.3)
β’ Android SDK at /home/raulino/Android/Sdk
β’ Android NDK location not configured (optional; useful for native profiling support)
β’ Platform android-27, build-tools 27.0.3
β’ ANDROID_HOME = /home/raulino/Android/Sdk
β’ Java binary at: /home/raulino/.development/android-studio/jre/bin/java
β’ Java version OpenJDK Runtime Environment (build 1.8.0_152-release-1024-b01)
β’ All Android licenses accepted.
[β] Android Studio (version 3.1)
β’ Android Studio at /home/raulino/.development/android-studio
β’ Java version OpenJDK Runtime Environment (build 1.8.0_152-release-1024-b01)
[β] VS Code (version 1.22.1)
β’ VS Code at /usr/share/code
β’ Dart Code extension version 2.11.2
[!] Connected devices
! No devices available
! Doctor found issues in 1 category.
```
| tool,t: flutter doctor,a: first hour,P2,team-tool,triaged-tool | low | Critical |
313,583,914 | pytorch | γTrain issue with caffe2 detectronγAborted at 1523501813 (unix time) try "date -d @1523501813" if you are using GNU date *** PC: @ 0x7fa35103733a (unknown) | I try to run resnet50_trainer.py to build my model by using 100 images, but I get the following error messages. My train method is drawn on the following web, https://github.com/royhuang9/Detectron. How to fix the error and run correctly using Caffe2? I am not sure if this is a Caffe2 issue.
Train command:
python2 tools/train_net.py --cfg experiments/e2e_faster_rcnn_resnet-50-FPN_pascal2007.yaml OUTPUT_DIR experiments/output
Error meassage:
json_stats: {"accuracy_cls": 1.000000, "eta": "1:59:55", "iter": 26200, "loss": 0.000865, "loss_bbox": 0.000183, "loss_cls": 0.000680, "loss_rpn_bbox_fpn2": 0.000000, "loss_rpn_bbox_fpn3": 0.000000, "loss_rpn_bbox_fpn4": 0.000003, "loss_rpn_bbox_fpn5": 0.000003, "loss_rpn_bbox_fpn6": 0.000000, "loss_rpn_cls_fpn2": 0.000000, "loss_rpn_cls_fpn3": 0.000005, "loss_rpn_cls_fpn4": 0.000004, "loss_rpn_cls_fpn5": 0.000002, "loss_rpn_cls_fpn6": 0.000000, "lr": 0.002500, "mb_qsize": 64, "mem": 2451, "time": 0.212877}
*** Aborted at 1523501813 (unix time) try "date -d @1523501813" if you are using GNU date ***
PC: @ 0x7fa35103733a (unknown)
*** SIGSEGV (@0x0) received by PID 6778 (TID 0x7fa2edffb700) from PID 0; stack trace: ***
@ 0x7fa351390390 (unknown)
@ 0x7fa35103733a (unknown)
@ 0x7fa351039184 __libc_malloc
@ 0x7fa348101c51 (unknown)
@ 0x7fa348164232 (unknown)
@ 0x7fa3481642ea (unknown)
@ 0x7fa3481f29d4 (unknown)
@ 0x7fa3481f2da0 (unknown)
@ 0x4c195b PyEval_EvalFrameEx
@ 0x4c136f PyEval_EvalFrameEx
@ 0x4b9ab6 PyEval_EvalCodeEx
@ 0x4d54b9 (unknown)
@ 0x4eebee (unknown)
@ 0x4a577e PyObject_Call
@ 0x4c5e10 PyEval_CallObjectWithKeywords
@ 0x7fa33adca3d0 pybind11::detail::object_api<>::operator()<>()
@ 0x7fa33adcbb51 caffe2::python::PythonOpBase<>::RunOnDevice()
@ 0x7fa33ad932cb caffe2::Operator<>::Run()
@ 0x7fa33addce3a _ZN6caffe213GPUFallbackOpINS_6python8PythonOpINS_10CPUContextELb0EEENS_11SkipIndicesIJEEEE11RunOnDeviceEv
@ 0x7fa33add95d5 caffe2::Operator<>::Run()
@ 0x7fa336bcef0a caffe2::DAGNet::RunAt()
@ 0x7fa336bcdbbe caffe2::DAGNetBase::WorkerFunction()
@ 0x7fa34a5cbc80 (unknown)
@ 0x7fa3513866ba start_thread
@ 0x7fa3510bc41d clone
@ 0x0 (unknown)
| caffe2 | low | Critical |
313,673,188 | rust | one_time_diagnostics and --teach doesn't work with incremental compilation | Both `one_time_diagnostics` and `--teach` suppresses diagnostic information if it was already emitted. However with incremental compilation if cause of the other diagnostic is removed without affecting the cause of other such diagnostics, the information will not be emitted at all. | A-lints,T-compiler,A-incr-comp,C-bug | low | Minor |
313,680,520 | TypeScript | JSDOC inconsistance casting behavior | <!-- π¨ STOP π¨ π¦π§π’π£ π¨ πΊπ»πΆπ· π¨ -->
<!--
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 CONTRIBUTING guidelines: https://github.com/Microsoft/TypeScript/blob/master/CONTRIBUTING.md
* Read the FAQ: https://github.com/Microsoft/TypeScript/wiki/FAQ
-->
<!-- If you have a QUESTION:
THIS IS NOT A FORUM FOR QUESTIONS.
Ask questions at http://stackoverflow.com/questions/tagged/typescript
or https://gitter.im/Microsoft/TypeScript
-->
<!-- If you have a SUGGESTION:
Most suggestion reports are duplicates, please search extra hard before logging a new suggestion.
See https://github.com/Microsoft/TypeScript-wiki/blob/master/Writing-Good-Design-Proposals.md
-->
<!-- If you have a BUG:
Please fill in the *entire* template below.
-->
<!-- Please try to reproduce the issue with `typescript@next`. It may have already been fixed. -->
**TypeScript Version:** 2.8.1
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:**
jsdoc, casting
**Code**
utils.d.ts
```typescript
declare namespace utils {
export type PartialTypeGuard<T, U extends T> = (value: T) => value is U;
export type TypeGuard<T> = PartialTypeGuard<any, T>;
}
```
test.js
```javascript
// @ts-check
/// <reference path="./utils.d.ts" />
/**
* @type {utils.TypeGuard<string>}
*/
let pass = ((a)=>typeof a === 'string');
/**
* @type {utils.TypeGuard<string>}
*/
let fail = (a)=>typeof a === 'string';
```
**Expected behavior:**
They both throw error due to casting between item with incompatible signature without explicit casting
**Actual behavior:**
The first one pass for unknown reason , the second one failed as expect
**Playground Link:** <!-- A link to a TypeScript Playground "Share" link which demonstrates this behavior -->
**Related Issues:**
| Bug,Domain: JavaScript | low | Critical |
313,685,454 | rust | unsized generic field not allowed in repr(packed) struct | For some reason, using a generic `T: ?Sized` does not work in a `repr(packed)` struct, while a non-generic unsized type is OK.
```
#[repr(packed)]
struct Bar([u8]); // ok
#[repr(packed)]
struct Foo<T: ?Sized>(T): // err
```
An non-`packed` is also OK:
```struct Foo<T: ?Sized>(T); // ok```
The full error:
```
error[E0277]: the trait bound `T: std::marker::Sized` is not satisfied
|
5 | struct Foo<T: ?Sized>(T); // err
| ^^ `T` does not have a constant size known at compile-time
|
= help: the trait `std::marker::Sized` is not implemented for `T`
= help: consider adding a `where T: std::marker::Sized` bound
= note: only the last field of a struct may have a dynamically sized type
```
This may be related to #47068?
## Meta
```
$ rust --version --verbose
rustc 1.25.0 (84203cac6 2018-03-25)
binary: rustc
commit-hash: 84203cac67e65ca8640b8392348411098c856985
commit-date: 2018-03-25
host: x86_64-unknown-linux-gnu
release: 1.25.0
LLVM version: 6.0
``` | A-type-system,C-enhancement,T-types,A-repr-packed | low | Critical |
313,687,788 | angular | HttpErrorResponse statusText always default to 'OK' regardless of statusCode | ## I'm submitting a...
<!-- Check one of the following options with "x" -->
<pre><code>
[ ] Regression (a behavior that used to work and stopped working in a new release)
[x] Bug report <!-- Please search GitHub for a similar issue or PR before submitting -->
[ ] Performance issue
[ ] Feature request
[ ] Documentation issue or request
[ ] Support request => Please do not submit support request here, instead see https://github.com/angular/angular/blob/master/CONTRIBUTING.md#question
[ ] Other... Please describe:
</code></pre>
## Current behavior
<!-- Describe how the issue manifests. -->
When server response does not contain any statusText, [HttpErrorResponse.statusText](https://angular.io/api/common/http/HttpResponseBase#statusText) is [automatically set to `OK`](https://github.com/angular/angular/blob/5.2.9/packages/common/http/src/xhr.ts#L140) regardless of the status code.
Thus you can end up with `403 OK` error response for instance, which I believe is weird. In the [previous documentation](https://angular.io/api/http/Response#statusText) it was clearly stated that `OK` was the default value, but I'm not sure to understand the reason behind it -- an empty string would be less confusing I think.
## Expected behavior
<!-- Describe what the desired behavior would be. -->
Either the statusString should be left empty, or even better check for status code and either use the proper http status for it, or show `ok` only for success status code only (and.. something has to be done for others)?
## Minimal reproduction of the problem with instructions
<!--
For bug reports please provide the *STEPS TO REPRODUCE* and if possible a *MINIMAL DEMO* of the problem via
https://stackblitz.com or similar (you can use this template as a starting point: https://stackblitz.com/fork/angular-gitter).
-->
## What is the motivation / use case for changing the behavior?
<!-- Describe the motivation or the concrete use case. -->
While debugging some 403 errors on a website, I looked at HttpErrorResponse object with a 403 OK status which made me think there was something going wrong with the server and/or the error handling, but actually that was not the error at all. I know that documentation states `Do not depend on this.`, but this is still misleading when you see that in logs IMO.
## Environment
<pre><code>
Angular version: 5.2.9, still applicable on latest version
<!-- Check whether this is still an issue in the most recent Angular version -->
Browser:
- [ ] Chrome (desktop) version XX
- [ ] Chrome (Android) version XX
- [ ] Chrome (iOS) version XX
- [x] Firefox version 59.0.2
- [ ] Safari (desktop) version XX
- [ ] Safari (iOS) version XX
- [ ] IE version XX
- [ ] Edge version XX
For Tooling issues:
- Node version: XX <!-- run `node --version` -->
- Platform: <!-- Mac, Linux, Windows -->
Others:
<!-- Anything else relevant? Operating system version, IDE, package manager, HTTP server, ... -->
</code></pre>
| type: bug/fix,breaking changes,freq1: low,area: common/http,state: confirmed,design complexity: low-hanging,P3 | medium | Critical |
313,696,036 | javascript | separate event handler functions for an event attached to different inputs or one event handler function with lot of if statements/ switch statement? | separate event handler functions for an event attached to different inputs or one event handler function with lot of if statements/ switch statement?
```js
// event handler 1
function validateEmail() {
// validate email here
}
// event handler 2
function validatePassword() {
//validate password here
}
// or
// generic event handler
function validateFields() {
/*
if(is email){
validate email
}
if(is password) {
validate password
}
*/
}
``` | question | low | Minor |
313,716,527 | pytorch | caffe2/cuda_rtc can throw during the destructor | Note that this line throws in the destructor:
https://github.com/pytorch/pytorch/blob/master/caffe2/cuda_rtc/common_rtc.h#L27
We saw those cases happening in TC so we rewrote our logic to unload first before destroying.
You may wan to consider the same. | caffe2 | low | Minor |
313,737,077 | rust | run-make/relocation-model fails on mips{el}-unknown-linux-gnu | ```
error: make failed
status: exit code: 2
command: "make"
stdout:
------------------------------------------
LD_LIBRARY_PATH="/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/lib:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-tools/mipsel-unknown-linux-gnu/release/deps:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-sysroot/lib/rustlib/mipsel-unknown-linux-gnu/lib:" '/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/bin/rustc' --out-dir /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu -L /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu -Clinker=gcc -C relocation-model=static foo.rs
LD_LIBRARY_PATH="/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/lib/rustlib/mipsel-unknown-linux-gnu/lib:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-tools/mipsel-unknown-linux-gnu/release/deps:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-sysroot/lib/rustlib/mipsel-unknown-linux-gnu/lib:" /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/foo
LD_LIBRARY_PATH="/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/lib:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-tools/mipsel-unknown-linux-gnu/release/deps:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-sysroot/lib/rustlib/mipsel-unknown-linux-gnu/lib:" '/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/bin/rustc' --out-dir /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu -L /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu -Clinker=gcc -C relocation-model=dynamic-no-pic foo.rs
LD_LIBRARY_PATH="/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/lib/rustlib/mipsel-unknown-linux-gnu/lib:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-tools/mipsel-unknown-linux-gnu/release/deps:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-sysroot/lib/rustlib/mipsel-unknown-linux-gnu/lib:" /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/foo
LD_LIBRARY_PATH="/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/lib:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-tools/mipsel-unknown-linux-gnu/release/deps:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-sysroot/lib/rustlib/mipsel-unknown-linux-gnu/lib:" '/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/bin/rustc' --out-dir /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu -L /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu -Clinker=gcc -C relocation-model=default foo.rs
LD_LIBRARY_PATH="/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/lib/rustlib/mipsel-unknown-linux-gnu/lib:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-tools/mipsel-unknown-linux-gnu/release/deps:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-sysroot/lib/rustlib/mipsel-unknown-linux-gnu/lib:" /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/foo
LD_LIBRARY_PATH="/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/lib:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-tools/mipsel-unknown-linux-gnu/release/deps:/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage0-sysroot/lib/rustlib/mipsel-unknown-linux-gnu/lib:" '/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/bin/rustc' --out-dir /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu -L /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu -Clinker=gcc -C relocation-model=dynamic-no-pic --crate-type=dylib foo.rs --emit=link,obj
Makefile:4: recipe for target 'all' failed
------------------------------------------
stderr:
------------------------------------------
error: linking with `gcc` failed: exit code: 1
|
= note: "gcc" "-Wl,--as-needed" "-Wl,-z,noexecstack" "-L" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/lib/rustlib/mipsel-unknown-linux-gnu/lib" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/foo.foo0.rcgu.o" "-o" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/libfoo.so" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/foo.crate.metadata.rcgu.o" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/foo.crate.allocator.rcgu.o" "-Wl,-z,relro,-z,now" "-nodefaultlibs" "-L" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu" "-L" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/stage2/lib/rustlib/mipsel-unknown-linux-gnu/lib" "-Wl,--start-group" "-Wl,-Bstatic" "-Wl,--whole-archive" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/rustc.7Ndt3AnBrnar/libstd-40c425b776fadd4b.rlib" "-Wl,--no-whole-archive" "-Wl,--whole-archive" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/rustc.7Ndt3AnBrnar/libpanic_unwind-8dcd6188e6a17652.rlib" "-Wl,--no-whole-archive" "-Wl,--whole-archive" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/rustc.7Ndt3AnBrnar/libunwind-9cbe08a673e59763.rlib" "-Wl,--no-whole-archive" "-Wl,--whole-archive" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/rustc.7Ndt3AnBrnar/liballoc_system-09b8c28e82ce8959.rlib" "-Wl,--no-whole-archive" "-Wl,--whole-archive" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/rustc.7Ndt3AnBrnar/liblibc-6564eda3124d9d78.rlib" "-Wl,--no-whole-archive" "-Wl,--whole-archive" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/rustc.7Ndt3AnBrnar/liballoc-12610f5c8bd6d62d.rlib" "-Wl,--no-whole-archive" "-Wl,--whole-archive" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/rustc.7Ndt3AnBrnar/libstd_unicode-0ecffe7df08180c9.rlib" "-Wl,--no-whole-archive" "-Wl,--whole-archive" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/rustc.7Ndt3AnBrnar/libcore-ecdb2c8a330fb142.rlib" "-Wl,--no-whole-archive" "-Wl,--end-group" "/home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/rustc.7Ndt3AnBrnar/libcompiler_builtins-b6c0cc1595c95da2.rlib" "-Wl,-Bdynamic" "-l" "dl" "-l" "rt" "-l" "pthread" "-l" "gcc_s" "-l" "c" "-l" "m" "-l" "rt" "-l" "pthread" "-l" "util" "-shared"
= note: /usr/bin/ld: /home/draganm/rust1/rust/build/mipsel-unknown-linux-gnu/test/run-make-fulldeps/relocation-model.stage2-mipsel-unknown-linux-gnu/foo.crate.allocator.rcgu.o: relocation R_MIPS_26 against `__rdl_alloc' can not be used when making a shared object; recompile with -fPIC
collect2: error: ld returned 1 exit status
```
The issue is that you cannot use non-pic (dynamic-no-pic) code to produce .so on MIPS. You might get away on other platforms that use same call sequence/relocations for pic/no-pic local calls. Is Windows platform original reason for using dynamic-no-pic for shlib target? Is it ok to change this test to not set relocation-model for shlib case or use pic for Linux ?
| O-linux,O-MIPS,T-compiler,T-bootstrap,C-bug | low | Critical |
313,743,858 | go | net/http: TestServerDuplicateBackgroundRead crashes netbsd-arm builder | One of the net/http tests, `TestServerDuplicateBackgroundRead`, has crashed the netbsd-arm-bsiegert builder machine several times. When this happens, the machine locks up, accepting no input on the console, and the CPU gets hot.
02328153520b10d6252064ef18d5eeb214f55824 marks this test as skipped.
This issue is tracked upstream in NetBSD as http://gnats.netbsd.org/53173.
//cc @zoulasc @bradfitz @tklauser | Testing,OS-NetBSD,NeedsFix | low | Critical |
313,761,424 | You-Dont-Know-JS | this & Object Prototypes - Chapter 3 : 'in' operator | There's a paragraph related to the **in** operator that goes like this :
> There's (currently) no built-in way to get a list of all properties which is equivalent to what the in operator test would consult (traversing all properties on the entire [[Prototype]] chain, as explained in Chapter 5). You could approximate such a utility by recursively traversing the [[Prototype]] chain of an object, and for each level, capturing the list from **Object.keys(..)** -- only enumerable properties.
Why do we need to capture only the **enumerable** properties. The **in** operator also checks for **non-enumerable** properties as well. So instead of using Object.keys(), shouldn't we use Object.getOwnPropertyNames() to build the utility?
| for second edition | low | Minor |
313,768,900 | pytorch | [Caffe2] - Run on different Cuda streams | Hi,
I loaded the model once on the GPU, and would like to run the network on different cuda streams, meaning having two videos running simultaneously using same model. I managed to implement this, but can't find where I need to change the cuda stream, for the Workspace to run on.
Maybe it will be more helpful for OpenCV I'm passing cv::cuda::Stream, is there a way that I can do the same with caffe2, while sharing the data or running the network? Also I'm using the C++ version not Python.
Thanks very much | caffe2 | low | Minor |
313,834,738 | go | x/build/cmd/gerritbot: auto-wrap description | Per https://go-review.googlesource.com/c/vgo/+/105335/9//COMMIT_MSG#9 it would be good/nice to have gerritbot auto-wrap the first comment to save people having to do this by hand (which in a webpage text box is painful)
cc @andybons | help wanted,Builders,NeedsFix | low | Major |
313,839,388 | vscode | Select editor text via touch | Issue Type: <b>Feature Request</b>
When using a touch screen, I want to be able to interact with VSCode in an intuitive manner.
When I touch and drag on text, I expect it to do something. I think the most intuitive option would be to select text.
When I touch and drag on empty space (space with no characters in the buffer), I expect the editor to scroll with my finger.
VS Code version: Code 1.22.1 (950b8b0d37a9b7061b6f0d291837ccc4015f5ecd, 2018-04-06T02:26:57.615Z)
OS version: Windows_NT x64 10.0.16299
<!-- generated by issue reporter --> | feature-request,editor-core,touch/pointer | medium | Critical |
313,840,631 | TypeScript | Review preserving internal trivia in code fixes | **TypeScript Version:** master
**Code**
```ts
exports.x = () /*comment*/ => f;
```
Run code fix to convert to an ES6 module.
**Expected behavior:**
```ts
export function x() /*comment*/ { return f; }
```
**Actual behavior:**
```ts
export function x() { return f; }
``` | Bug,Domain: Quick Fixes | low | Minor |
313,859,224 | flutter | Document the accuracy of native memory leak instrumentations | 
There are some Dart memory management related stacks that show up as leaks in Xcode Instruments when playing around with the Gallery.
We should document whether users that see these are legitimate concerns or just red herrings that are artifacts of our garbage collection mechanism.
cc @sethladd @cbracken @sivachandra
| team,engine,dependency: dart,d: api docs,P2,team-engine,triaged-engine | low | Major |
313,874,512 | pytorch | [Feature Request] Optimization with constraint (L-BFGS-B) | Is there an agenda for implementing bounded optimization method like L-BFGS-B?
Scipy already has a functional method here scipy.optimize.fmin_l_bfgs_b. | proposal accepted,todo,triaged | low | Major |
313,875,407 | rust | "File name too long" error when compiling files with long names | ```
could not write output to /home/russell/code/rust/rust/build/x86_64-unknown-linux-gnu/test/ui/nll/closure-requirements/region-lbr1-does-outlive-lbr2-because-implied-bound.region_lbr1_does_outlive_lbr2_because_implied_bound0-317d481089b8c8fe83113de504472633.rs.rcgu.o: File name too long
```
Can the test runner avoid trying to write file with names this long? I've moved my rust installation to a higher level directory to work around it. | C-enhancement,T-bootstrap | low | Critical |
313,884,968 | flutter | Potential customers run Flutter in Instruments and perceive it as leaking | My understanding is this occurs for two reasons.
1. Running Flutter in --debug, and see something (presumably tracing/observatory) taking up increasing memory over time. This occurred for customer:gold early in their initial eval.
2. Running flutter and loading lots of images and hitting: https://github.com/flutter/flutter/issues/13606. Customers see what looks like Flutter leaking memory every time an image is loaded. | platform-ios,engine,P3,team-ios,triaged-ios | low | Critical |
313,901,333 | vscode | Command - Undo Replace All | In Search Panel should be button "Undo Replace All", whitch undo all "replaced" changes in all files.
Should to be work like standard undo function, but only for files whitch "Replace All" made a changes.
It looks something simple to introduce. Am I right? | feature-request,search-replace,undo-redo | high | Critical |
313,923,370 | flutter | Add a per directory structure README.md to the engine repo | Add a directory map README.md at the root similar to https://fuchsia.googlesource.com/docs/+/master/development/source_code/layers.md or https://fuchsia.googlesource.com/docs/+/master/map.md describing what should go where or a README.md like https://github.com/chinmaygarde/flutter_engine/blob/02c1f084cbe6a4f4c7a99a9278e11c9b41d657ec/content_handler/README.md per directory describing the role of the directory.
cc @cbracken @chinmaygarde | team,engine,d: api docs,P2,team-engine,triaged-engine | low | Minor |
313,928,959 | go | x/build: finish, document Gerrit hashtags for managing open CLs | In chat with @andybons @FiloSottile @bcmills and @ianlancetaylor yesterday, I proposed we start using Gerrit's "hashtags" support, now that our hosted Gerrit supports it (notedb support is active for us) and PolyGerrit (the new web UI) supports it.
After discussion, we settled on using tags:
```
so, how about: "wait-author", "wait-release", "wait-issue", "wait-$PERSON"
where $PERSON is some substring of a name/email in the reviewer set.
and if somebody is named "Issue Release" and their email is [email protected], then you have to use their full email in $PERSON
we can mass remove the "wait-release" labels when tree opens
```
While triaging today, I found we should also add support for:
```
wait-cl-NNNNN -- this CL is blocked until NNN is resolved
```
I started looking into adding hashtag support to maintner and gopherbot.
Tracking that work here, then documenting all this on the wiki.
| Documentation,Builders,NeedsFix | medium | Major |
313,931,736 | go | x/build/cmd/gerritbot: impersonate reviewer accounts | The experience of only receiving messages from @gopherbot can be a bit impersonal and confusing. When migrating from Google Code we had gopherbot impersonate accounts. We could do the same for comments on PRs.
It would be opt-in but as long as the main reviewers are covered, I think it would achieve its goal.
/cc @andybons | Builders,NeedsInvestigation | low | Minor |
313,975,569 | terminal | Conhost Properties Dialog β "Find out more about new console features" leads to outdated documentation | * Setup:
* System 1:
* Windows 10 Pro Fall Creator's Update, Version 1709 (16299.371)
* Release ring: Semi-Annual Channel (targeted)
* Locale: English (United States) (en-US)
* System 2:
* Windows 10 Home Insider Preview, Version 1803 (17113.73)
* Release ring: Release Preview
* Locale: English (United States) (en-US)
* Steps to Reproduce:
* Via command prompt:
1. Launch `cmd` from the run dialog
2. Right click the title of the new window
3. Select `Properties`
4. Click the link in `Find out more about new console features`
* Via Powershell
1. Launch `powershell` from the run dialog
2. Right click the title of the new window
3. Select `Properties`
4. Click the link in `Find out more about new console features`
* What happens:
* Browser opens a link a technet page: https://go.microsoft.com/fwlink/?LinkId=507549 (https://technet.microsoft.com/en-us/library/mt427362.aspx)
* Inside the page:
`What's New in the Windows Console in Windows Server 2016 Technical Preview`
and
`Applies To: Windows Server Technical Preview`
* What I expected:
* Inside the page:
`What's New in the Windows Console in Windows 10 Fall Creator's Update and Windows Server 2016 version 1709` (or it's appropriate Windows Support wording)
and
`Applies To: Windows 10 version 1709, Windows Server 2016 version 1709`
* Severity: Very low/cosmetic/documentation (for english speaking users)
* Documentation is otherwise correct except for the above two issues
* Other languages require automatic translations to use the page
* Possible solutions:
* Link to: https://blogs.msdn.microsoft.com/commandline/2017/10/11/whats-new-in-windows-console-in-windows-10-fall-creators-update/ instead
* Issues:
* Is not a technet/windows support page/is a blog post
* Is only available in english
* Update: https://go.microsoft.com/fwlink/?LinkId=507549 (https://technet.microsoft.com/en-us/library/mt427362.aspx)
* Issues:
* Might retroactively change existing preview documentation
* Is also only available in english(?) (there are no french/german/spanish/chinese/etc translations of the page at the moment)
* Migrate doc to https://docs.microsoft.com/en-us/windows/console
* Issues:
* Requires an update of cmd.exe if a new LinkId is desired, if retroactively changing the existing LinkId is not desired
* Technet pages seem more consistent with under the hood/administrative tool pages | Product-Conhost,Issue-Docs,Area-UserInterface | low | Minor |
314,009,222 | node | Flexible stdios for IPC | * **Version**: master
* **Platform**: all
* **Subsystem**: child_process
A noted problem pattern is broken data flow when i) spawning OS commands that predate Node.js and that do not expect their stdios to be non-blocking. ii) Composite IPC that involves more than one Node process in the process chain.
The underlying issue is the mixed mode I/O in a number of variations based on number of parent-child combo that is possible.
While many of these can be mitigated by studying the I/O characteristics of parent and child and crafting a suitable IPC that works, inability to (programatically) control the blocking behavior (flow mode) of the stdio channel causes usability issues.
Proposals:
Node parent -> non-Node child
Node parent - Node child
- exec family functions to receive option to control the blocking behavior of child stdios (new option)
- adjust the child stdios accordingly, while retaining parent side of pipes non-blocking
- convert stdout and stderr to nonblocking prior to exit to help drain the buffered chunks (revisit #6773)
Node parent - non-Node child -> Node grand child
Node parent - non-Node child -> non-Node grand child
- restore stdio characteristics on exit (already being worked upon in #17737)
- convert stdout and stderr to nonblocking prior to exit to help drain the buffered chunks
references:
truncated stdout on porcess exit: #19218 #784 #6456 #6379
pre-fabricated duplex streams for IPC: #16553 #15714 #13542
inconsistent IPC with process chain: #9413
unexpected flow mode for stdio: #14752 #18446 #19838 | child_process,stdio | low | Critical |
314,024,460 | rust | Type inference fails to infer lifetimes. | ## Short summary
When I partially specify the `&mut` type of a `Fn` trait argument, my code compiles. If I let the compiler find it out, it fails. I cannot see a single, language-wise, reason why it should fail.
## Details
If I try the following code:
```
fn set(t : &mut usize) {}
fn reset(t : &mut usize) {}
fn main() {
let vec : Vec<&Fn(_)> = vec![&set, &reset];
let mut t = 0;
for f in vec {
f(&mut t);
}
}
```
Then it fails to compile with the message:
```
error[E0597]: `t` does not live long enough
--> src/main.rs:14:16
|
14 | f(&mut t);
| ^ borrowed value doeerror[E0597]: `t` does not live long enough
15 | }
16 | }
| - `t` dropped here while still borrowed
|
= note: values in a scope are dropped in the opposite order they are created
```
However, if I replace the underscore with `&mut _`, everything works fine.
I assumed that, in functions, an underscore should be strictly better for type interference than everything else, especially in this case where the borrow checker can just insert `&mut` itself. Apparently this is not the case, and I have no idea why.
## Meta
```
$ rustc --version --verbose
rustc 1.25.0 (84203cac6 2018-03-25)
binary: rustc
commit-hash: 84203cac67e65ca8640b8392348411098c856985
commit-date: 2018-03-25
host: x86_64-unknown-linux-gnu
release: 1.25.0
LLVM version: 6.0
``` | C-enhancement,A-diagnostics,A-lifetimes,A-closures,T-compiler,A-inference | low | Critical |
314,026,764 | vscode | Use fresh environment block on new terminals in Windows | At the moment, if any environment variables change, or new "apps" are installed while VS Code is running, in order for those changes to be recognized VS Code needs to be closed and re-open.
For instance, when running `npm i -g <something>` or `choco install <package>` in another prompt or powershell window, those new commands are not recognized until the app is rebooted.
It would be great if adding a new terminal instance with `+` would load these new variables/apps/notsurewhatterminologytouse. | feature-request,windows,terminal-process | high | Critical |
314,030,951 | TypeScript | JSDoc optional/required variant doesn't play well with multi-@param RecordType declarations | <!-- π¨ STOP π¨ π¦π§π’π£ π¨ πΊπ»πΆπ· π¨ -->
<!--
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 CONTRIBUTING guidelines: https://github.com/Microsoft/TypeScript/blob/master/CONTRIBUTING.md
* Read the FAQ: https://github.com/Microsoft/TypeScript/wiki/FAQ
-->
<!-- If you have a QUESTION:
THIS IS NOT A FORUM FOR QUESTIONS.
Ask questions at http://stackoverflow.com/questions/tagged/typescript
or https://gitter.im/Microsoft/TypeScript
-->
<!-- If you have a SUGGESTION:
Most suggestion reports are duplicates, please search extra hard before logging a new suggestion.
See https://github.com/Microsoft/TypeScript-wiki/blob/master/Writing-Good-Design-Proposals.md
-->
<!-- If you have a BUG:
Please fill in the *entire* template below.
-->
<!-- Please try to reproduce the issue with `typescript@next`. It may have already been fixed. -->
**TypeScript Version:** 2.9.0-dev.20180330
<!-- Search terms you tried before logging this (so others can find this issue more easily) -->
**Search Terms:** as mentioned in [this comment](https://github.com/Microsoft/TypeScript/issues/19645#issuecomment-377952907), I'm opening this issue because with this JSDoc syntax typescript doesn't recognize the params correctly.
**Code**
```ts
// A *self-contained* demonstration of the problem follows...
// Test this by running `tsc` on the command-line, rather than through another build tool such as Gulp, Webpack, etc.
/**
* @param {!object} args
* @param {!string} args.prop1
* @param {?string} args.prop2
*/
function patate(args) {
const { prop1, prop2 } = args;
}
```
**Expected behavior:** No errors
**Actual behavior:** With allowJs and checkJs, I get JSDoc '@param' tag has name 'prop1', but is no parameter with that name.
**Playground Link:** <!-- A link to a TypeScript Playground "Share" link which demonstrates this behavior -->
**Related Issues:** https://github.com/Microsoft/TypeScript/issues/19645
| Bug,Domain: JSDoc,Domain: JavaScript | low | Critical |
314,086,944 | rust | LLVM has rare assert failure (3% of time?) during async trans | I recently tried turning on LLVM assertions (and built it in Debug mode, etc; part of debugging something else) via the following in my `config.toml`:
```
[llvm]
# Indicates whether rustc will support compilation with LLVM
# note: rustc does not compile without LLVM at the moment
#enabled = true
# Indicates whether the LLVM build is a Release or Debug build
# optimize = true
optimize = false
# Indicates whether an LLVM Release build should include debug info
#release-debuginfo = false
release-debuginfo = true
# Indicates whether the LLVM assertions are enabled or not
assertions = true
# Indicates whether ccache is used when building LLVM
ccache = true
```
and then discovered that sometimes we get LLVM assertion faillures when running the *compile-fail* test suite.
This is of course surprising since one does not expect trans to run if the input has some problem with its static semantics.
Here is an example of such output.
```
failures:
---- [compile-fail] compile-fail/simd-type-generic-monomorphisation.rs stdout ----
error: Error: expected failure status (Some(101)) but received status None.
status: signal: 6
command: "/home/pnkfelix/Dev/Mozilla/rust.git/objdir-dbgopt/build/x86_64-unknown-linux-gnu/stage1/bin/rustc" "/home/pnkfelix/Dev/Mozilla/rust.git/src/test/compile-fail/simd-type-generic-monomorphisation.rs" "-L" "/home/pnkfelix/Dev/Mozilla/rust.git/objdir-dbgopt/build/x86_64-unknown-linux-gnu/test/compile-fail" "--target=x86_64-unknown-linux-gnu" "-Zui-testing" "-C" "prefer-dynamic" "-o" "/home/pnkfelix/Dev/Mozilla/rust.git/objdir-dbgopt/build/x86_64-unknown-linux-gnu/test/compile-fail/simd-type-generic-monomorphisation.stage1-x86_64-unknown-linux-gnu" "-Crpath" "-Zmiri" "-Zunstable-options" "-Lnative=/home/pnkfelix/Dev/Mozilla/rust.git/objdir-dbgopt/build/x86_64-unknown-linux-gnu/native/rust-test-helpers" "-L" "/home/pnkfelix/Dev/Mozilla/rust.git/objdir-dbgopt/build/x86_64-unknown-linux-gnu/test/compile-fail/simd-type-generic-monomorphisation.stage1-x86_64-unknown-linux-gnu.aux" "-A" "unused"
stdout:
------------------------------------------
------------------------------------------
stderr:
------------------------------------------
error: monomorphising SIMD type `Simd2<X>` with a non-machine element type `X`
error: aborting due to previous error
rustc: /home/pnkfelix/Dev/Mozilla/rust.git/src/llvm/lib/CodeGen/TargetPassConfig.cpp:805: virtual void llvm::TargetPassConfig::addMachinePasses(): Assertion `TPI && IPI && "Pass ID not registered!"' failed.
------------------------------------------
thread '[compile-fail] compile-fail/simd-type-generic-monomorphisation.rs' panicked at 'explicit panic', tools/compiletest/src/runtest.rs:2919:9
note: Run with `RUST_BACKTRACE=1` for a backtrace.
failures:
[compile-fail] compile-fail/simd-type-generic-monomorphisation.rs
test result: FAILED. 233 passed; 1 failed; 1 ignored; 0 measured; 2087 filtered out
```
Iteratively running that `rustc` command 1000 times yielded that LLVM assert 31 times (that's the basis of my 3% above).
I looked at a generated core file and got a backtrace from it, here: https://gist.github.com/7cb39f28bd7f4960130a4ef555589e98
The current hypothesis in `#rustc` about this is that it is caused from setting up async trans, and then something during a subsequent error from the compiler (in this case the error is signaled during trans, since its a SIMD issue) is tearing down LLVM state that the async trans needs to remain in place. | A-LLVM,T-compiler,C-bug | low | Critical |
314,138,916 | vue | Merge common class names of component and root element of that component | ### What problem does this feature solve?
It looks a little awkward if there are duplicate class names in a DOM element.
### What does the proposed API look like?
If you look into [this fiddle](https://jsfiddle.net/zLnzuh1f/).
The `div` element of `foo` component will have two `bar` classes in it. So it looks like.
```html
<div class="bar bar"></div>
```
If the two classes would have been merged into one.
```html
<div class="bar"></div>
```
It would have worked and behaved exactly in the same way but the code will look a lot cleaner.
<!-- generated by vue-issues. DO NOT REMOVE --> | improvement,has PR | low | Minor |
314,182,866 | vue | Vue Selected Bind Does Not Work on Option Tags | ### Version
2.5.16
### Reproduction link
[https://codepen.io/jbenner/pen/geJqex](https://codepen.io/jbenner/pen/geJqex)
### Steps to reproduce
1. Instantiate a Vue instance with a `data` property named `locations` which is an array consisting of `['all', 'north', 'south', 'east', 'west']`.
2. Iterate over the locations in an `option` tag like so `<option v-for="location in locations" :selected="location === 'all'">{{ location }}</option>`.
### What is expected?
The `selected` attribute to be set on the "all" `option` element.
### What is actually happening?
The `selected` attribute is not being set.
---
I've also tried the following while attempting to debug but to no avail:
```
<option v-for="(location, index) in locations" :selected="index === 0">{{ location }}</option>
```
<!-- generated by vue-issues. DO NOT REMOVE --> | improvement | low | Critical |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.