Run the F# analyzers over every project and script - #249
Merged
Merged
Conversation
Add an Analyze pipeline to build.fsx that runs fsharp-analyzers with Ionide.Analyzers and G-Research.FSharp.Analyzers over the nine solution projects, build.fsx and OldFsYaccTests.fsx in a single invocation, writing one analysis.sarif for the repository. Generated lexer and parser sources, AssemblyInfo.fs and the test SDK entry point are excluded, since findings there are not fixable here. The analyzer packages live in their own paket group with package storage, so no project references them and the script finds them at a fixed path without knowing the version. Pull requests get a separate ubuntu-only analyze job, and pushes to master run the same steps so code scanning has a baseline to diff against. Both upload the report to GitHub code scanning and do not fail the workflow. Remove tests/fsyacc/oldfsyacctests.fsx.lock, a leftover from when that script pulled FAKE in through paket.
The first analyzer run reported 153 findings, all warnings or lower. This resolves them so the code scanning baseline starts clean and a new finding on a pull request stands out. Most of it is mechanical: postfix generic syntax throughout the runtime and the tools, printf-style calls without format specifiers replaced by plain writes, ordinal string comparisons made explicit, list functions instead of Seq over lists, and a type argument on each bare `string` call. A few changes touch the shape of the code. The Rule, Production and ExplicitPrec union cases have named fields. Action is qualified access only, because its Error case shadowed Result.Error. Associativity and Domain are struct unions. TryDecodeUnicodeCategory returns a value option, so the UnicodeCategoryAP active pattern can be struct-returning without a conversion. The JSON example matches on the parse result rather than reading .Value. No behaviour changes. The fsyacc console hint about precedences reads the same, its %% escapes just became single % now that it is no longer a format string.
|
You are seeing this message because GitHub Code Scanning has recently been set up for this repository, or this pull request contains the workflow file for the Code Scanning tool. What Enabling Code Scanning Means:
For more information about GitHub Code Scanning, check out the documentation. |
The pack step hands RELEASE_NOTES.md to MSBuild as a /p:PackageReleaseNotes switch. MSBuild reads a newline, a semicolon or a comma in a property value as the start of the next switch, so the multi-line notes were cut into fragments and the Release pipeline failed with MSB1006 "Property is not valid" on every push to master since the move to Fun.Build. Each of those characters, and the others MSBuild gives meaning to, is now written as its %XX escape. MSBuild unescapes the value when it reads the property, so the nuspec carries the notes exactly as written.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Add an Analyze pipeline to build.fsx that runs fsharp-analyzers with
Ionide.Analyzers and G-Research.FSharp.Analyzers over the nine solution
projects, build.fsx and OldFsYaccTests.fsx in a single invocation, writing
one analysis.sarif for the repository. Generated lexer and parser sources,
AssemblyInfo.fs and the test SDK entry point are excluded, since findings
there are not fixable here.
The analyzer packages live in their own paket group with package storage,
so no project references them and the script finds them at a fixed path
without knowing the version.
Pull requests get a separate ubuntu-only analyze job, and pushes to master
run the same steps so code scanning has a baseline to diff against. Both
upload the report to GitHub code scanning and do not fail the workflow.
Remove tests/fsyacc/oldfsyacctests.fsx.lock, a leftover from when that
script pulled FAKE in through paket.