Sep 2020

4 commits
Remove precedence hacks no longer needed without postfix operators.
19 Sep 2020Tom Lanemaster14beta1REL_19_STABLE14beta1REL_18_STABLE14beta1REL_17_STABLE14beta1REL_16_STABLE14beta1REL_15_STABLE14beta1REL_14_STABLE14beta1
LLM analysis, 97% suremodified before commitmoderate changesfrom patchset #391 of 1 reviewer saw the final version

The gram.y cleanup follows John Naylor's attachment in #39, removing GENERATED, PRESERVE, and STRIP_P precedence, and also incorporates Peter Eisentraut's NULL_P removal from #29. Tom explicitly declined the proposed IDENT precedence movement and rewrote the explanation in #73.

  • Tom Lanesaw the final versioncommittercommitted the patch
Allow most keywords to be used as column labels without requiring AS.
18 Sep 2020Tom Lanemaster14beta1REL_19_STABLE14beta1REL_18_STABLE14beta1REL_17_STABLE14beta1REL_16_STABLE14beta1REL_15_STABLE14beta1REL_14_STABLE14beta1
LLM analysis, 100% surepart of a posted patchsetmoderate changesfrom patchset #541 of 2 reviewers saw the final version

Commits the keyword-label portion and row-estimate adjustment of #54, with BareColLabel terminology, separate barelabel/baredesc outputs, and ScanKeywordBareLabel storage replacing changes to ScanKeywordList and gen_keywordlist.pl. The estimate becomes 500 rather than the proposed 450; Tom describes the revisions in #65, #68, and #71.

  • John Naylorsaw an earlier versionLLM analysis, 100% sureJohn reviewed the keyword classification and API in #33, reported a pg_get_keywords crash in #35, and tested the corrected output in #39, before the committer changed its output and storage.
  • Tom Lanesaw the final versioncommittercommitted the patch
Remove support for postfix (right-unary) operators.
17 Sep 2020Tom Lanemaster14beta1REL_19_STABLE14beta1REL_18_STABLE14beta1REL_17_STABLE14beta1REL_16_STABLE14beta1REL_15_STABLE14beta1REL_14_STABLE14beta1
LLM analysis, 100% surepart of a posted patchsetmoderate changesfrom patchset #541 of 2 reviewers saw the final version

Commits the remaining postfix-support removal from the first patch of #54, after factorial removal was split out. Committer edits preserve psql compatibility with old servers, retain the ANY/ALL precedence-warning distinction in parse_expr.c, simplify deparsing, and revise errors and documentation; #65 confirms completion of this part.

  • John Naylorsaw an earlier versionLLM analysis, 99% sureJohn identified missing upgrade checks and dead code in #35, reviewed further cleanup in #39 and #42, and requested pg_dump warnings in #53, before subsequent compatibility and parser edits.
  • Tom Lanesaw the final versioncommittercommitted the patch
Remove factorial operators, leaving only the factorial() function.
17 Sep 2020Tom Lanemaster14beta1REL_19_STABLE14beta1REL_18_STABLE14beta1REL_17_STABLE14beta1REL_16_STABLE14beta1REL_15_STABLE14beta1REL_14_STABLE14beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #542 of 2 reviewers saw the final version

Extracts factorial-operator removal from the first patch of #54: matching deletions in pg_operator.dat, pg_proc.dat, and func.sgml, with adjusted regression tests and a catalog-version bump. Tom explicitly identifies the pushed change as a subset of that patch in #64.

  • John Naylorsaw the final versionLLM analysis, 100% sureJohn reviewed factorial tests in #35, proposed also removing the prefix factorial operator in #39, and caught obsolete factorial documentation and redundant tests in #53, followed by only light changes to this extracted portion.
  • Tom Lanesaw the final versioncommittercommitted the patch

Credits across 4 commits

Author
Mark DilgerLLM3
John NaylorLLM1
Reviewer
Tom Lanecommitter4
Committer
Tom Lane4

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