Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Fix loss of
run-summary.md
in GHA Job Summary
Scala Steward's GitHub Action Job Summary, which details success or failure of each repo in the Scala Steward run, was added with scala-steward-org/scala-steward#3071 in June 2023, followed up by scala-steward-org/scala-steward#3088, which ensured that old `run-summary.md` files couldn't persist in the workspace cache (where they might become misleading), by purging them before the cache was saved. Unfortunately #631 (released with v2.67.0 in September 2024) unintentionally broke the outputting of the Job Summary, because it moved the `saveWorkspaceCache()` call (which among other things deletes the `run-summary.md` file) to a point _before_ the `run-summary.md` file was read and output to the GitHub Action Job Summary. We can see this in these two successive runs of the Guardian's Scala Steward workflow: * Run 6002 (using [email protected]) successfully output the scala-steward summary: https://github.com/guardian/scala-steward-public-repos/actions/runs/11591772739 https://github.com/guardian/scala-steward-public-repos/actions/runs/11591772739/job/32272232470#step:7:4512 * Run 6003 (using [email protected]) silently omitted to output the summary: https://github.com/guardian/scala-steward-public-repos/actions/runs/11592668893 https://github.com/guardian/scala-steward-public-repos/actions/runs/11592668893/job/32275051343#step:7:35 The fix is just to move the code writing the run-summary earlier, to _before_ the new `finally` block introduced by #631 which executes `saveWorkspaceCache()`. I've also renamed the method to make its actions a little clearer: * OLD: `workspace.saveWorkspaceCache()` * NEW: `workspace.purgeTempFilesAndSaveCache()`
- Loading branch information