Fix ReadOnlySequence First#27691
Conversation
|
Centos.74.Amd64.Open:Release-x64 @dotnet-bot test Linux x64 Release Build |
|
LGTM |
| } | ||
|
|
||
| [Fact] | ||
| public void CanGetFirst() |
There was a problem hiding this comment.
We have other First tests right?
There was a problem hiding this comment.
Yes; though not mismatched multi segment buffer sizes (which this is)
|
Don't have permission to rerun Tizen; and OSX is just generally broken |
|
This file isn't included in the test csproj (and it needs to be): https://github.com/dotnet/corefx/blob/master/src/System.Memory/tests/ReadOnlyBuffer/ReadOnlySequenceTests.TryGet.cs And it needs to be updated to remove use of IMemoryList! Can you try and get code coverage numbers too please? |
| var buffer = new ReadOnlySequence<byte>(bufferSegment1, 0, bufferSegment4, 200); | ||
|
|
||
| Assert.Equal(500, buffer.Length); | ||
| var length = 500; |
What do you mean? @dotnet-bot test Tizen armel Debug Build |
|
dotnet bot ignores me in corefx for Tizen reruns (works in coreclr though). I think its a limited group because it gets so overloaded |
|
|
|
BuffersExtensions at 70% - we need to get that coverage up - I will add to https://github.com/dotnet/corefx/issues/26768 Anything from the |
|
Some argument throws for .ctor validation (empty array). A couple exceptions I'm not sure can ever be hit, some non-segment paths (pure array/string/OwnedMemory cast paths) |
|
added report to issue |
|
|
| Assert.Equal(i, buffer.First.Length); | ||
| buffer = buffer.Slice(1); | ||
| length--; | ||
| Assert.Equal(length, buffer.Length); |
There was a problem hiding this comment.
Might not be soon. In airport, then in plane for 10 hrs 😀
* Fix ReadOnlySequence First * Feedback Commit migrated from dotnet/corefx@8710caa
5 participants
Seen in dotnet/corefxlab#2143
Don't have enough bandwidth on the train to download and install a fresh coreclr to get passed apicheck; but the test definitely fails before the change. Hopefully it passes after :)
/cc @ahsonkhan @davidfowl @pakrym