Let TransparentCompiler respect hash directives like #nowarn

[dawedawe]

Description

This fixes the issue shown in this PR

As @0101 said, we need FSharpDiagnosticOptions depending on the ParsedInput of a file. But we loose that pretty deep in the call stack. Instead of changing returns all over the place, this PR makes use of the cached FSharpParsedFile to fix this.
I hope this is an acceptable tradeoff to move the TransparentCompiler forward.

(I don't think this needs release notes)

Checklist

• Test cases added

• Performance benchmarks added in case of performance changes

• Release notes entry updated:

  Please make sure to add an entry with short succinct description of the change as well as link to this pull request to the respective release notes file, if applicable.

  Release notes files:

  • If anything under src/Compiler has been changed, please make sure to make an entry in docs/release-notes/.FSharp.Compiler.Service/<version>.md, where <version> is usually "highest" one, e.g. 42.8.200
  • If language feature was added (i.e. LanguageFeatures.fsi was changed), please add it to docs/releae-notes/.Language/preview.md
  • If a change to FSharp.Core was made, please make sure to edit docs/release-notes/.FSharp.Core/<version>.md where version is "highest" one, e.g. 8.0.200.

  Information about the release notes entries format can be found in the documentation.
  Example:

  • More inlines for Result module. (PR #16106)
  • Correctly handle assembly imports with public key token of 0 length. (Issue #16359, PR #16363)
    *while! (Language suggestion #1038, PR #14238)

  If you believe that release notes are not necessary for this PR, please add NO_RELEASE_NOTES label to the pull request.

[github-actions]

⚠️ Release notes required, but author opted out

Warning

Author opted out of release notes, check is disabled for this pull request.
cc @dotnet/fsharp-team-msft

[0101]

If we can fix it like this then why not!

But some diagnostics refactoring might be needed eventually. We're already processing them with this config applied in ComputeTcIntermediate:

fsharp/src/Compiler/Service/TransparentCompiler.fs

Lines 1128 to 1134 in 93b8f0c

	lettcConfig=
	ApplyNoWarnsToTcConfig(tcConfig, parsedMainInput, Path.GetDirectoryName mainInputFileName)
	// update the error handler with the modified tcConfig
	errHandler.DiagnosticOptions <- tcConfig.diagnosticsOptions
	letdiagnosticsLogger= errHandler.DiagnosticsLogger

Maybe it's possible to somehow make use of that.

[majocha]

Maybe this code could work in the Finisher in processGraphNode?

[dawedawe]

The last push is just to correct the first approach taken.
I'm looking into a better way to do this deeper in the callstack.

[dawedawe]

Maybe it's possible to somehow make use of that.

Thanks for the pointer. I think we are good now.

[0101]

Huh, looks like it was close 😀

Maybe this code could work in the Finisher in processGraphNode?

That would make sense if we need some prior state for the processing. But probably we're fine at this stage just with proper filtering and formatting can still happen at the end when we're creating the result.

[dawedawe]

Huh, looks like it was close 😀

Oh yeah, just a little shake and shuffle was needed :)

[0101]

@dawedawe do you think you could temporarily trigger the tests in CI again? To see if this had any unexpected effects.

We really need the tests auto-enabled in main, I'll try to set it up here: #16591

[dawedawe]

Mmh, some expected Information diagnostics are reported as Warning diagnostics.
I'll look into that tomorrow.

[0101]

The tests are now in main so you can undo dawedawe@b47ba47