Mar 2023

1 commit
pg_regress: Emit TAP compliant output
31 Mar 2023Daniel Gustafssonmaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1Also in: making pg_regress less noisy by removing boilerplate
LLM analysis, 100% suremodified before commitlight changesfrom patchset #571 of 5 reviewers saw the final version

Applying #57 to the commit's parent reproduces the implementation exactly: the remaining differences are one explanatory comment in src/test/regress/pg_regress.c and two documentation additions in doc/src/sgml/regress.sgml. Daniel confirms those final comment/doc tweaks in #58.

  • Andres Freundsaw an earlier versiontrailerAndres reviewed output design, tested ECPG and process failures, and proposed stderr diagnostics and bailout naming in #17, #29, #32, and #36, before later emitter and error-path changes.
  • Tom Lanesaw an earlier versiontrailerTom insisted on preserving timing, parallel-group information, and stable schedule-order numbering in #9, #15, and #18, before the switch to a single TAP format.
  • Nikolay Shaplovsaw an earlier versiontrailerNikolay tested TAP consumption, requested simpler shared formatting and clearer bailout documentation, and declared the patch ready in #27, #34, #38, and #43, before subsequent functional revisions.
  • Dagfinn Ilmari MannsÃ¥kersaw an earlier versiontrailerIlmari proposed separate note/diag helpers and correct comment prefixes in #30, before the emitter was introduced and subsequently revised.
  • Peter Eisentrautsaw the final versiontrailerPeter questioned maintaining two formats in #8 and reviewed the late implementation in #52 and #55, after which only output-formatting, cleanup, comment, and documentation changes followed.

Feb 2023

1 commit
Consider a failed process as a failed test in pg_regress
23 Feb 2023Daniel Gustafssonmaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1Also in: pg_regress: Treat child process failure as test failure (landed as posted)

Credits across 2 commits

Author
Daniel GustafssonLLM2
Reviewer
Andres Freund1
Dagfinn Ilmari Mannsåker1
Nikolay Shaplov1
Peter Eisentraut1
Tom Lane1
Reported by
Andres Freund1
Committer
Daniel Gustafsson2

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