Jun 2022

1 commit
Be more careful about GucSource for internally-driven GUC settings.
8 Jun 2022Tom Lanemaster15beta2REL_19_STABLE15beta2REL_18_STABLE15beta2REL_17_STABLE15beta2REL_16_STABLE15beta2REL_15_STABLE15beta2
LLM analysis, 100% surelanded as postedfrom patchset #651 of 3 reviewers saw the final version

The complete nine-file diff matches #65 after ignoring index metadata and hunk positions. Tom confirmed the push in #70.

  • Jonathan S. Katzsaw the final versionLLM analysis, 100% sureJonathan reviewed and tested #65, confirming generated-default suppression and correct standby visibility in #67, then endorsed the stack-limit behavior in #69.
  • Justin Pryzbysaw an earlier versionLLM analysis, 96% sureJustin reported unwanted computed shared-memory settings in #56, motivating this fix before either backend patch was posted.
  • Christoph Bergsaw an earlier versionLLM analysis, 96% sureChristoph argued in #63 that in_hot_standby should remain visible, a requirement implemented in #65 after his pre-patch feedback.

Apr 2022

4 commits
Fix case sensitivity in psql's tab completion for GUC names.
13 Apr 2022Tom Lanemaster15beta1REL_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

No attachment contains this change to the three GUC completion queries in src/bin/psql/tab-complete.c. It implements the case-insensitivity repair identified in #19 and explicitly deferred in #25, lowercasing the input pattern and flattening subqueries.

  • Jonathan S. Katzsaw an earlier versionLLM analysis, 88% sureJonathan raised mixed-case GUC handling in #18, prompting the completion investigation in #19, but did not review an attached version of this fix or the thread's final patchset.
Further tweak the default behavior of psql's \dconfig.
13 Apr 2022Tom Lanemaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surelanded as postedfrom patchset #511 of 1 reviewer saw the final version

The single describe.c hunk is identical to #51 modulo rebase metadata: it adds s.setting IS DISTINCT FROM s.boot_val to the existing source filter. The commit announcement is #55.

  • Jonathan S. Katzsaw the final versionLLM analysis, 100% sureJonathan reported the initdb-default clutter in #46, then tested #51 and explicitly said the patch looked good in #52.
Tweak the default behavior of psql's \dconfig.
11 Apr 2022Tom Lanemaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surewritten from the discussion, no matching patch0 of 4 reviewers saw the final version

No attached patch contains this initial default-only filtering change in describe.c and psql-ref.sgml. It implements Christoph's source-filter proposal in #36 and Tom's distinguishing table heading in #40, with completion announced in #45.

  • Christoph Bergsaw an earlier versionLLM analysis, 100% sureChristoph supplied the source-filter query and proposed non-default output with an explicit wildcard escape in #36, before the later filtering and backend revisions.
  • Jonathan S. Katzsaw an earlier versionLLM analysis, 99% sureJonathan challenged hiding defaults in #39 and accepted the clarified proposal in #44, before the later substantive revisions.
  • Julien Rouhaudsaw an earlier versionLLM analysis, 96% sureJulien supported default filtering while arguing against blanket exclusion of override settings in #38, feedback reflected in this initial change before subsequent revisions.
  • David G. Johnstonsaw an earlier versionLLM analysis, 88% sureDavid endorsed a default other than showing everything and examined how to explain the omitted settings in #41, before the thread's later implementation changes.
psql: add \dconfig command to show server's configuration parameters.
7 Apr 2022Tom Lanemaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% suremodified before commitlight changesfrom patchset #253 of 5 reviewers saw the final version

The commit retains #25's implementation, renames \dconf to \dconfig, adds a completion comment, and replaces planner-flag tests with work_mem tests. It explicitly credits Mark Dilger and Tom Lane; Mark identifies his original implementation in #9.

  • Jonathan S. Katzsaw the final versionLLM analysis, 100% sureJonathan reviewed output and mixed-case handling in #18, then ran the revised SQL and approved the documentation and code in #27.
  • Justin Pryzbysaw an earlier versionLLM analysis, 99% sureJustin proposed human-readable current_setting() output in #22, which #25 adopted alongside additional local logic changes.
  • Joe Conwaysaw an earlier versionLLM analysis, 87% sureJoe reviewed the draft's column priorities in #16, arguing for less prominence for unit and type before the revised display and qualified-name fixes in #25.
  • Pavel Stehulesaw the final versionLLM analysis, 94% surePavel identified the existing dconf project's naming collision in #34, and Tom adopted the \dconfig spelling in #35 with only light changes before landing.
  • Tom Lanesaw the final versioncommittercommitted the patch

Credits across 5 commits

Author
Tom LaneLLM2
Mark DilgerLLM1
Reviewer
Tom Lanecommitter1
Committer
Tom Lane5

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