Update ICSharpCode.Decompiler to 7.2.1.6856 (#67875) #69065
Conversation
| @@ -1198,7 +1198,7 @@ class C | |||
| instance void .ctor () cil managed | |||
| { | |||
| // Method begins at RVA 0x207f | |||
| // Code size 7 (0x7) | |||
| // Code size: 7 (0x7) | |||
There was a problem hiding this comment.
new C# decompiler syntax.
There was a problem hiding this comment.
Sorry for the change... I changed this for consistency with other info we print and did not realize in time that ildasm does not use a colon...
|
@dotnet/roslyn-compiler ptal. @genlu ptal. |
|
@dotnet/roslyn-compiler ptal. |
|
@dotnet/roslyn-infrastructure ptal. |
| @@ -21,7 +21,7 @@ namespace Microsoft.CodeAnalysis.Editor.UnitTests.MetadataAsSource | |||
| [UseExportProvider] | |||
| public abstract partial class AbstractMetadataAsSourceTests : IAsyncLifetime | |||
| { | |||
| protected static readonly string ICSharpCodeDecompilerVersion = "7.1.0.6543"; | |||
There was a problem hiding this comment.
Not a huge deal, but can't this code just read the version number from the file?
| {{ | ||
| return _message; | ||
| }} | ||
| /*Error: Empty body found. Decompiled assembly might be a reference assembly.*/ |
There was a problem hiding this comment.
This part of the output isn't really the point of the test, but any idea why this might have changed?
|
@dotnet/roslyn-compiler for minor updates to test files based on new decompiler update. |
|
|
||
| private static string RemoveHeaderComments(string value) | ||
| { | ||
| return s_headerCommentsRegex.Replace(value, ""); |
There was a problem hiding this comment.
Since we are already doing this, could we also remove the colon from // Code size: 6 (0x6) comments to reduce the diffs?
There was a problem hiding this comment.
that's nto the worst idea :) But the works already been done. so i'd prefer to keep as is :)
There was a problem hiding this comment.
I'm not sure I understand which IL baselines are impacted here. Basically, for a change like this I'd expect many many more baselines to be impacted. Is the issue that we use ildasm in some places and ICSharpCode.Decompiler in other places? I'm concerned about the inconsistency in baselines and about ongoing PRs being disrupted by the change.
There was a problem hiding this comment.
Yeah, i'm not sure what a good strategy is here. As long as the compiler is using this tool, and as long as this tool changes their outpus across releases, then we have this problem.
|
@dotnet/roslyn-compiler ptal. |
|
@dotnet/roslyn-compiler ptal. This fixes a bad memory oom :-) |
ghost
added this to the 
Revert "Merge pull request #67919 from dibarbet/revert_icsharpcodedecompiler"
This reverts commit 85eaf94, reversing changes made to 0f8a8ed.