Sep 2021

3 commits
Fix some issues with TAP tests for postgres -C
22 Sep 2021Michael Paquiermaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% suremodified before commitlight changesfrom patchset #121 of 1 reviewer saw the final version

The commit preserves the pg_ctl invocation, expected failure output, reduced test count, and removal of online-cluster checks from #12. It additionally removes the Msys skip as proposed in #7 and corrects the privileged-account comment in src/bin/pg_checksums/t/002_actions.pl.

  • Andres Freundsaw the final versionLLM analysis, 100% sureAndres reported the privileged-account failure in #9, corrected the account description in #13, and confirmed that #12 passed the checksum tests in #14.
Fix places in TestLib.pm in need of adaptation to the output of Msys perl
21 Sep 2021Michael Paquiermaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1REL_14_STABLE14.0REL_13_STABLE13.5+4 more
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #72 of 2 reviewers saw the final version

The four TestLib.pm hunks match #7 exactly; the accompanying pg_checksums test re-enablement was omitted and landed separately in 1a9d802828.

  • Andrew Dunstansaw the final versiontrailerAndrew requested normalization before the loops and supplied the combined stdout/stderr expression in #6, which #7 adopted.
  • Álvaro Herrerasaw the final versiontrailerÁlvaro called #4 reasonable and pointed to earlier related fixes in #5, with only light revisions following.
Disable test for postgres -C on Msys
17 Sep 2021Michael Paquiermaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surewritten from the discussion, no matching patch0 of 1 reviewer saw the final version

Following Tom Lane's failure report in #2, Michael announced the temporary Msys skip in #3 and committed it before any attachment. The later attachments instead remove that skip, so their textual overlap is not evidence that this commit came from them.

  • Tom Lanesaw an earlier versionLLM analysis, 90% sureTom supplied the failing Windows test output in #2 before the temporary fix; his review predates all attachments and the subsequent pg_ctl-based test redesign in #12.

Credits across 3 commits

Author
Michael PaquierLLM2
Reviewer
Alvaro Herrera1
Andrew Dunstan1
Committer
Michael Paquier3

N LLM means N of the credits come from LLM analysis, not from commit trailers.