Oct - Nov 2026

5 commits
doc PG 19 relnotes: add item about shared parallel vacuum cost
7 Nov 2026Bruce MomjianREL_19_STABLE19rc2
doc PG 19 relnotes: fix multi-xact wording
27 Oct 2026Bruce MomjianREL_19_STABLE19rc2
LLM analysis, 100% surelanded as postedfrom patchset #1271 of 1 reviewer saw the final version

Both multixact wording changes match #127 exactly, with only shifted line numbers in doc/src/sgml/release-19.sgml.

  • Bruce Momjiansaw the final versioncommittercommitted the patch
doc PG 19 relnotes: adjust for missing items
27 Oct 2026Bruce MomjianREL_19_STABLE19rc2
LLM analysis, 100% surelanded as postedfrom patchset #1321 of 1 reviewer saw the final version

The complete diff matches #132, including four replacement I/O variables, the public wait-event names, and the revised stemmer text.

  • shihao zhongsaw the final versionLLM analysis, 100% sureHe corrected three details in the initial patch #130 and explicitly said the revised patch looked good before commitment #133.
doc: adjust ALTER TABLE docs for PG 19
27 Oct 2026Bruce MomjianREL_19_STABLE19rc2
LLM analysis, 100% surelanded as postedfrom patchset #1301 of 1 reviewer saw the final version

The deletion of the obsolete DETACH PARTITION CONCURRENTLY CHECK-constraint claim in doc/src/sgml/ref/alter_table.sgml matches #130 exactly.

  • Bruce Momjiansaw the final versioncommittercommitted the patch
doc PG 19 relnotes: add mention of view pg_stat_progress_repack
27 Oct 2026Bruce MomjianREL_19_STABLE19rc2
LLM analysis, 99% surelanded as postedfrom patchset #129

The commit exactly matches Bruce's rewritten attachment #129, including regrouping the vacuum views; Pavel supplied the earlier proposal #126, but Bruce authored this replacement wording and layout.

Oct 2026

1 commit
doc PG 19 relnotes: remove reverted item
13 Oct 2026Bruce MomjianREL_19_STABLE19rc2
LLM analysis, 100% surelanded as postedfrom patchset #1191 of 1 reviewer saw the final version

The deletion of the reverted logical-decoding feature's entire release-note entry is identical to #119.

  • Bruce Momjiansaw the final versioncommittercommitted the patch

Sep 2026

1 commit
doc PG 19 relnotes: fix retain_dead_tuples details
15 Sep 2026Bruce MomjianREL_19_STABLE19beta4
LLM analysis, 100% surelanded as postedfrom patchset #1171 of 1 reviewer saw the final version

The subscription links and conflict-detection wording match #117 exactly despite a different base and hunk offset.

  • Bruce Momjiansaw the final versioncommittercommitted the patch

Aug 2026

1 commit
doc PG 19 relnotes: move replication items to logical replicat.
28 Aug 2026Bruce MomjianREL_19_STABLE19beta4
LLM analysis, 100% surelanded as postedfrom patchset #1151 of 1 reviewer saw the final version

The two complete item relocations into Logical Replication are identical to #115.

  • Bruce Momjiansaw the final versioncommittercommitted the patch

Jul 2026

3 commits
doc PG 19 relnote: move Oauth items
17 Jul 2026Bruce MomjianREL_19_STABLE19beta3
LLM analysis, 100% suremodified before commitmoderate changesfrom patchset #1081 of 1 reviewer saw the final version

Applying both patches from #108 to their base reproduces the commit except for an additional move of the complete validator failure-detail entry into Source Code, requested in the accompanying message and confirmed in #109.

  • Bruce Momjiansaw the final versioncommittercommitted the patch
doc PG 19 relnotes: swap OAUTH hook and validator items
15 Jul 2026Bruce MomjianREL_19_STABLE19beta3
LLM analysis, 97% suremodified before commitsignificant changesfrom patchset #1041 of 2 reviewers saw the final version

Bruce rebuilt #104 because of invalid markup, but swapped the hook into Source Code and failure details into Server Configuration instead of the proposed destinations; #106 and #109 explain the reconstruction and accidental reversal.

  • Jacob Championsaw an earlier versionLLM analysis, 98% sureJacob expressed no objection to Andreas's proposed moves #105, but the committed reconstruction changed their destinations substantially.
  • Bruce Momjiansaw the final versioncommittercommitted the patch
doc PG 19 relnotes: "OAUTH validators to supply failure item"
15 Jul 2026Bruce MomjianREL_19_STABLE19beta3
LLM analysis, 99% surepart of a posted patchsetlight changesfrom patchset #1042 of 2 reviewers saw the final version

This implements only the failure-detail entry's move to Source Code from #104, repairing the malformed item boundaries and carrying its explanatory paragraph along, as described in #106.

  • Jacob Championsaw the final versionLLM analysis, 98% sureJacob accepted the proposed section moves #105, and this extracted move preserves that intent with markup repairs.
  • Bruce Momjiansaw the final versioncommittercommitted the patch

May - Jun 2026

13 commits
doc PG 19 relnotes: fix autovacuum_vacuum_score_weight prefix
29 Jun 2026Bruce MomjianREL_19_STABLE19beta2
LLM analysis, 100% surelanded as postedfrom patchset #1011 of 1 reviewer saw the final version

The one-line autovacuum_vacuum_insert_score_weight prefix correction exactly matches #101.

  • Bruce Momjiansaw the final versioncommittercommitted the patch
doc PG 19 relnotes: remove VALIDATE CONSTRAINT lock item
18 Jun 2026Bruce Momjianmaster19beta2REL_19_STABLE19beta2
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

Fujii Masao requested removal of the reverted ALTER DOMAIN VALIDATE CONSTRAINT lock change #98, and Bruce confirmed its removal #99; no attachment contains this deletion.

  • Fujii Masaosaw the final versionLLM analysis, 100% sureHe identified the obsolete release-note entry and requested its deletion after the feature was reverted #98.
doc PG 19 relnotes: improve awkward or confusing wording
8 Jun 2026Bruce Momjianmaster19beta2REL_19_STABLE19beta2
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

The wording, autovacuum link, and background-worker explanation changes implement Álvaro Herrera's detailed inline review #91, with Bruce explaining the accepted and rejected suggestions in #92.

  • Alvaro Herrerasaw the final versionLLM analysis, 100% sureHe reviewed the published draft's migration, monitoring, configuration, and COPY entries and supplied the corrections implemented by this commit #91.
doc PG 19 relnotes: more fixes
26 May 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #831 of 1 reviewer saw the final version

All additional spelling, grammar, parameter-name, and terminology fixes match Thom's supplementary patch #83 exactly.

  • Bruce Momjiansaw the final versioncommittercommitted the patch
doc PG 19 relnotes: various corrections
26 May 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #811 of 1 reviewer saw the final version

The six corrections, including COPY input and dead-tuple retention wording, exactly match #81.

  • Bruce Momjiansaw the final versioncommittercommitted the patch
doc PG 19 relnotes: adjust item to mention pg_replication_slots
26 May 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 99% surelanded as postedfrom patchset #801 of 1 reviewer saw the final version

The diff exactly matches Bruce's attachment #80; the commit explicitly credits Chong Peng, whose correction Chao Li forwarded in #79, and Chong has no message of his own in the supplied thread.

  • Bruce Momjiansaw the final versioncommittercommitted the patch
doc PG 19 relnotes: remove "Add fake LSN support to hash index"
15 May 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

The fake-LSN entry is removed following Peter Geoghegan's explanation #77 and Bruce's agreement #78; the commit also adds a missing JSON_ARRAY commit link, without an attached patch.

  • Peter Geoghegansaw the final versionLLM analysis, 100% sureHe explained that hash-index fake LSNs had no independent user-visible benefit because index prefetching had not landed #77.
doc PG 19 relnotes: add two optimizer hooks
12 May 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #741 of 1 reviewer saw the final version

The added joinrel_setup_hook and join_path_setup_hook names and commit link exactly match #74.

  • jian hesaw the final versionLLM analysis, 100% sureHe identified the two missing optimizer hooks in the published draft, which Bruce added unchanged in substance #70.
doc PG 19 relnotes: remove "Optionally" for CPU optimizations
12 May 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #732 of 2 reviewers saw the final version

Both deletions of 'Optionally' are identical to #73.

  • John Naylorsaw the final versionLLM analysis, 100% sureHe pointed out that these CPU extensions have no user option and discussed removing the misleading qualification #66 #69.
  • Robert Treatsaw the final versionLLM analysis, 86% sureHe reviewed the misleading optionality wording and proposed availability-based phrasing before Bruce resolved the issue by deleting the qualifier #68 #73.
doc PG 19 relnotes: adjustments/removal of items
11 May 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

The ICU and int128 item removals, radix-sort wording, and stray incompatibility sentence deletion implement John Naylor's inline review #66, acknowledged in #67.

  • John Naylorsaw the final versionLLM analysis, 100% sureHe identified misleading performance and compatibility claims and proposed the wording corrections implemented here #66.
doc PG 19 relnotes: add UTF-8 case folding performance item
8 May 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #632 of 2 reviewers saw the final version

The new UTF-8 case-folding performance entry exactly matches #63.

  • Andreas Karlssonsaw the final versionLLM analysis, 100% sureHe identified the omitted ICU optimization and explained its substantial effect on case-conversion functions #55.
  • Phil Florentsaw the final versionLLM analysis, 94% sureHe explained the optimization's relevance to lower() expression indexes, a practical justification Bruce explicitly accepted #58 #63.
doc PG 19 relnotes: correct two items
8 May 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #621 of 1 reviewer saw the final version

Both the function-prototype and btree_gin corrections exactly match #62.

  • jian hesaw the final versionLLM analysis, 100% sureHe challenged the incorrect FDW qualification and partial-qualification description, prompting both corrections #52.
doc PG 19 relnotes: add missing commits and details
7 May 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #601 of 1 reviewer saw the final version

All three streaming-read and WAIT FOR changes exactly match #60.

  • Xuneng Zhousaw the final versionLLM analysis, 100% sureHe identified the omitted pgstattuple and GIN improvements and the missing write/flush WAIT FOR functionality #44.

Apr 2026

7 commits
doc PG 19 relnotes: fix typo, "date" -> "data"
20 Apr 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

The date-to-data correction follows Thomas Kellerer's inline typo report #41 and Bruce's confirmation #43; no attachment contains it.

  • Thomas Kellerersaw the final versionLLM analysis, 100% sureHe identified the bytea/uuid 'date types' typo and supplied the correction #41.
doc PG 19 relnotes: add author and move items
16 Apr 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #231 of 1 reviewer saw the final version

Richard's credit, wording, and item relocations match #23; the only additional change is deletion of a blank line, allowed by the whitespace exception.

  • Bruce Momjiansaw the final versioncommittercommitted the patch
doc PG 19 relnotes: update author
16 Apr 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #211 of 1 reviewer saw the final version

The one-line VACUUM memory-reporting attribution correction exactly matches #21.

  • Bruce Momjiansaw the final versioncommittercommitted the patch
doc PG 19 relnotes: add free space map all-visible item
15 Apr 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #191 of 1 reviewer saw the final version

The complete all-visible release-note addition is identical to #19, including its original free-space-map wording.

  • Melanie Plagemansaw the final versionLLM analysis, 100% sureShe identified the missing user-visible behavior of SELECT updating visibility information and explained why it deserved a release note #11.
doc PG 19 relnotes: remove "Lakshmi N" as author of checksums
15 Apr 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

Daniel Gustafsson requested the checksum credit correction inline #7, and Bruce explained the removal in #17; no matching patch was attached.

  • Daniel Gustafssonsaw the final versionLLM analysis, 100% sureHe reviewed the online-checksum author list and specified the corrected list adopted by Bruce #7.
doc PG 19 relnotes: fix "now targets"
15 Apr 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surelanded as postedfrom patchset #161 of 1 reviewer saw the final version

The vacuumdb 'now targets' correction exactly matches the applied patch subsequently attached in #16.

  • Laurenz Albesaw the final versionLLM analysis, 100% sureHe identified and corrected the vacuumdb 'now targets' typo before commitment #6.
doc PG 19 relnotes: adjust ShmemRequestStruct item
15 Apr 2026Bruce Momjianmaster19beta1REL_19_STABLE19beta1
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

The commit credits Ashutosh and adopts his inline ShmemRequestStruct description and author correction #3; Bruce omitted the proposed extra explanatory paragraph as stated in #9.

  • Bruce Momjiansaw the final versioncommittercommitted the patch

Credits across 31 commits

Author
Bruce MomjianLLM9
Masahiko Sawada4
Andreas KarlssonLLM2
Thom Brown2
Antonin Houska1
Ashutosh Bapat1
Chao Li1
Chong Peng1
Jacob Champion1
Laurenz Albe1
Richard Guo1
shihao zhong1
Reviewer
Bruce Momjiancommitter15
Reported by
Masahiko Sawada4
Andreas Karlsson3
jian he2
John Naylor2
Laurenz Albe2
shihao zhong2
Thom Brown2
Alvaro Herrera1
Antonin Houska1
Ashutosh Bapat1
Chao Li1
Chong Peng1
Daniel Gustafsson1
Fujii Masao1
Jacob Champion1
Melanie Plageman1
Pavel Luzanov1
Peter Geoghegan1
Richard Guo1
Thomas Kellerer1
Xuneng Zhou1
Co author
Bruce Momjian1
Committer
Bruce Momjian31

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