[error]Could not find exactly one of the appropriate labels on the PR.

[larryjoelane]

I have added a label to a pr named patch and I receive the following error output.

output from action run:

Run zwaldowski/match-label-action@v1
  with:
    allowed: major minor patch
  
  env:
    SOLUTION_FILE: MySolution-Main.sln
    NUGET: C:\hostedtoolcache\windows\nuget.exe\5.4.0\x64/nuget.exe
##[error]Could not find exactly one of the appropriate labels on the PR.

steps from workflow:

     - name: bump version
        id: bump
        uses: zwaldowski/match-label-action@v1
        with:
          allowed: >
            major
            minor
            patch
      - name: output version
        id: next_version
        uses: zwaldowski/semver-release-action@v1
        with:
          bump: ${{ steps.bump.outputs.match }}
          github_token: ${{ secrets.GITHUB_TOKEN }}

[crazy4pi314]

I'm having exactly the same issue, any ideas?
qsharp-community/qram#50

[zwaldowski]

As this is simply a composition on @actions/github, it’s not clear whether the action is being given the right context from the PR.

@crazy4pi314, would you mind changing your uses to zwaldowski/match-label-action@zwaldowski/debug-issue-9 and turning on debug logs as described here? That should give me a quick path to triaging. Thanks for your patience!

[unitario]

I'm having this issue as well. The problem is that new lines \n are stripped out (only tested in ubuntu-latest) causing the parseAllowed function to yield incorrect result.

To get around this issue, you need to organize the allowed labels in a comma separated list as opposed to using new lines as illustrated in the documentation.

Correct

uses: zwaldowski/match-label-action@v2
with:
  allowed: major, minor, patch

Incorrect

uses: zwaldowski/match-label-action@v2
with:
  allowed: >
    major
    minor
    patch

[zwaldowski]

Ah, yes, thanks @unitario. It seems like GitHub's runner doesn't enjoy the > syntax. I've updated the README and will investigate on making that syntax work later without regressing #7.

@larryjoelane please see @unitario's note (if you still care about using this of course lol, if not that's fine) and reopen if it's still giving issues. Thanks all!