Feb 2022

1 commit
Disallow setting bogus GUCs within an extension's reserved namespace.
21 Feb 2022Tom Lanemaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 97% surelanded as postedfrom patchset #200 of 4 reviewers saw the final version

Combines the prerequisite reserved-prefix checks and API rename with the deletion delta from #20; its guc.c deletion logic and regression scenario match the posted delta. Rebase accommodation includes renaming the call in the newly added contrib/basic_archive module; the commit credits Florin Irion and Tom Lane.

  • Bharath Rupireddysaw an earlier versionLLM analysis, 88% sureReported continued acceptance of bogus extension settings in #4 and discussed error-handling compatibility in #6, before the reserved-prefix and deletion redesigns.
  • Fujii Masaosaw an earlier versionLLM analysis, 95% sureIdentified rejection of legitimate pre-load settings and a compilation failure in #8, then reported successful compilation but failing regression tests in #10.
  • Kyotaro Horiguchisaw an earlier versionLLM analysis, 94% sureExplained in #14 why validation must distinguish a reserved namespace from one whose extension has not yet loaded, before the implementation was rewritten.
  • Shinya Katosaw an earlier versionLLM analysis, 99% sureApproved Tom Lane's reserved-prefix patch in #18, before #20 added deletion of existing placeholders.

Dec 2021

4 commits
Revert changes about warnings/errors for placeholders.
27 Dec 2021Tom Lanemaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1reverts 2 commits here
LLM analysis, 100% surewritten from the discussion, no matching patch

Implements Tom Lane's announced revert in #19 after parallel-worker startup exposed failures involving bogus placeholders. No attached patch proposes this revert; its diff removes reserved-prefix enforcement, the API rename and associated tests.

Rename EmitWarningsOnPlaceholders() to MarkGUCPrefixReserved().
27 Dec 2021Tom Lanemaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

Implements the prose proposal in #16 to rename EmitWarningsOnPlaceholders to MarkGUCPrefixReserved and retain a compatibility macro in src/include/utils/guc.h. The attached patch does not contain that rename.

  • Shinya Katosaw the final versionLLM analysis, 96% sureEndorsed the rename proposal accompanying #16 with a separate '+1' in #18.
Rethink handling of settings with a prefix reserved by an extension.
27 Dec 2021Tom Lanemaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% suremodified before commitlight changesfrom patchset #161 of 4 reviewers saw the final version

The guc.c changes match #16, including find_option enforcement and exact prefix-length checking. The committed regression tests were relocated and expanded to cover loading an extension after creating a bogus placeholder, as explained in #19.

  • Bharath Rupireddysaw an earlier versionLLM analysis, 88% sureReported acceptance of bogus extension settings in #4 and reviewed error-handling compatibility in #6, before the reserved-prefix implementation was rewritten.
  • Fujii Masaosaw an earlier versionLLM analysis, 95% sureReported invalid rejection of legitimate extension settings and compilation problems in #8, followed by regression failures in #10.
  • Kyotaro Horiguchisaw an earlier versionLLM analysis, 94% sureExplained the need to preserve placeholders for unloaded extensions in #14, before Tom Lane replaced the validation approach.
  • Shinya Katosaw the final versionLLM analysis, 100% sureExplicitly said the replacement patch looked good in #18, after which only regression-test changes preceded this commit.
Add missing EmitWarningsOnPlaceholders() calls.
21 Dec 2021Tom Lanemaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #125 of 5 reviewers saw the final version

Takes only the missing-call component of #12, adding the omitted pltcl prefix alongside pltclu and explanatory guc.c comments. Tom Lane confirms that selection and correction in #15; neither proposed ERROR behavior change was included.

  • Daniel Gustafssonsaw the final versionLLM analysis, 92% sureReviewed the original call-site additions and identified the missing Tcl coverage in #2, with subsequent changes to this component extending the same calls and comments.
  • Bharath Rupireddysaw the final versionLLM analysis, 97% sureIdentified the missing delay_execution, ssl_passphrase_callback and worker_spi calls and requested usage guidance in #4, all reflected in the committed component.
  • Peter Eisentrautsaw the final versionLLM analysis, 91% sureRequested separating the uncontroversial call-site additions in #11, which the author did in #12 before that component was committed.
  • Kyotaro Horiguchisaw the final versionLLM analysis, 100% sureCaught the missing pltcl namespace alongside pltclu in #13, and Tom Lane explicitly incorporated that correction in #15.
  • Tom Lanesaw the final versioncommittercommitted the patch

Credits across 5 commits

Author
Tom LaneLLM2
Shinya KatoLLM1
Reviewer
Tom Lanecommitter1
Committer
Tom Lane5

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