I created a documentation improvement PR with no issue number and with skip-issue set. There was no problem with bedevere on this, and I successfully merged the PR in due course. I then set back-port flags for 3.10 and 3.11; back-port PRs were duly created and correctly inherited the skip-issue label. There were no problems with bedevere on these, either. The 3.11 back-port PR tests ran successfully and I merged it, too. The 3.10 back-port PR is where the problem occurred, though not initially. The tests for this failed due to a suspicious.py check and, to see if this was some transient issue, I closed the PR and opened it again just to re-run the tests. At this point, bedevere failed the PR on the issue number. The problem is that the back-port PR contains the text (GH-96344) from the original PR against main; for some reason, bedevere interprets this as an invalid issue number, when it would seem that it should check for the presence of skip-issue first. This is the 3.10 back-port PR which shows the issue.
I created a documentation improvement PR with no issue number and with
skip-issueset. There was no problem with bedevere on this, and I successfully merged the PR in due course. I then set back-port flags for 3.10 and 3.11; back-port PRs were duly created and correctly inherited theskip-issuelabel. There were no problems with bedevere on these, either. The 3.11 back-port PR tests ran successfully and I merged it, too. The 3.10 back-port PR is where the problem occurred, though not initially. The tests for this failed due to asuspicious.pycheck and, to see if this was some transient issue, I closed the PR and opened it again just to re-run the tests. At this point, bedevere failed the PR on the issue number. The problem is that the back-port PR contains the text(GH-96344)from the original PR againstmain; for some reason, bedevere interprets this as an invalid issue number, when it would seem that it should check for the presence ofskip-issuefirst. This is the 3.10 back-port PR which shows the issue.