Open Flamefire opened 3 years ago
Do you know if this is a regression by chance?
It might be. IIRC it did work some time ago, but can't tell for sure.
I tested the APPVEYOR_PULL_REQUEST_HEAD_COMMIT variable and that did work
Seemingly the Github action had a similar issue: https://github.com/codecov/codecov-action/issues/190 and they used a similar variable: https://github.com/codecov/codecov-action/commit/e77dab65bffcbc621e169f31ec9e12acac61a526
@Flamefire do you have another instance of this? I wonder if this has something to do with the commit SHAs on forks
I haven't tested it further as I believe that all evidence suggests that APPVEYOR_PULL_REQUEST_HEAD_COMMIT is the correct variable to use and I don't see any downside.
It looks like the appveyor commit SHA needs similar fixing as the Github Actions or Azure one.
In https://ci.appveyor.com/project/Flamefire/nowide-fr98b/builds/38560544/job/kb2fl8uw27ued5qs the commit is detected as
b6c2ea6c45b76b760189abe5246df472e701a1f9
while it should be0c853652e5409b32ef8d07939d93c4c5d7e55094
See the same CI on GHA: https://github.com/boostorg/nowide/pull/131/checks?check_run_id=2272049938
The correct env var for PRs seems to be APPVEYOR_PULL_REQUEST_HEAD_COMMIT