Repository navigation
Fix stale and inaccurate user docs - #129
Merged
jnasbyupgrade merged 1 commit intoOct 7, 2026
Merged
jnasbyupgrade merged 1 commit into
jnasbyupgrade merged 1 commit into
Conversation
README.asc: - PGXNTOOL_PGTLE_VERSION takes an exact range name (e.g. 1.5.0+), not a plain version mapped to a range; a plain version creates a bogus pg_tle/<value>/ directory. Fix the example and both descriptions. - make test: document the FS_INSTALL gating of install and the two safeguard checks (stale expected files, test/install ON_ERROR_STOP); test files run in byte-value order, not alphabetical; drop history narration from the exit-status note. - PGXNTOOL_ENABLE_FS_INSTALL=no also drops test-build's install prerequisite. - tag pushes to $(PGXN_REMOTE), not always origin. - test/install layout: there is no test/install/.gitignore. - testdeps/pgtap is a no-op with PGXNTOOL_ENABLE_PGXN_INSTALL=no; installcheck reaches all only through install. - Document the built-in pgxntool-sync-master/-local/-local-master sources, PGXNTOOL_POST_TAG_VERSION's allowed characters, and bin/version's optional history-file argument. - Fix the <<_Document_Handling>> cross-reference's case. CLAUDE.md: canonical repo URL, .asc doc extension, and the full make test chain with its gating and safeguard checks. pgtle.sh: comment/message-only changes (canonical repo URL, SEE ALSO pointing at README.asc's pg_tle Support section). README.html regenerated with asciidoctor. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
Important Review skippedAuto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the ⚙️ Run configuration
You can disable this status message by setting the Use the checkbox below for a quick retry:
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
jnasbyupgrade
merged commit Oct 7, 2026
c25cdf4
into
Postgres-Extensions:master
10 of 11 checks passed
This was referenced Oct 7, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Documentation fixes; no behavior changes. (Found while preparing the next release.)
README.asc
PGXNTOOL_PGTLE_VERSIONtakes an exact range name (1.0.0-1.4.0,1.4.0-1.5.0,1.5.0+), not a plain pg_tle version that gets mapped to a range. The value is passed straight through topgtle.sh --pgtle-version, so the old example (PGXNTOOL_PGTLE_VERSION=1.5.0) produced a boguspg_tle/1.5.0/directory. Fixed the example and both descriptions.make test: documents thatinstallis skipped withPGXNTOOL_ENABLE_FS_INSTALL=no, and that the stale-expected-file check and the test/installON_ERROR_STOPcheck both run as part of it.$(sort)).make testexit-status note.PGXNTOOL_ENABLE_FS_INSTALL=noalso drops test-build'sinstallprerequisite.make tagpushes to$(PGXN_REMOTE), not alwaysorigin.test/install/.gitignore; pgxntool never creates one (those ignores live in the top-level.gitignorefromsetup.sh).testdeps'pgtapprerequisite is a no-op withPGXNTOOL_ENABLE_PGXN_INSTALL=no;installcheckreachesallonly throughinstall.pgxntool-sync-master/-local/-local-mastersources, the charactersPGXNTOOL_POST_TAG_VERSIONallows ([A-Za-z0-9.-]+), andbin/version's optional<history-file>argument.<<_Document_Handling>>cross-reference, which didn't resolve.CLAUDE.md
Postgres-Extensions/pgxntoolsubtree URL;.asclisted as an accepted doc extension.make testchain now shows test-build, the FS_INSTALL/PGXN_INSTALL gating, and the safeguard checks.pgtle.sh
Comment and message text only: canonical repo URL (header and the delimiter-error message), and SEE ALSO now points at README.asc's "pg_tle Support" section instead of a nonexistent
README-pgtle.md. No logic changes. Because pgtle.sh isn't a doc file,check-test-prmay require a paired pgxntool-test PR or the maintainer-appliedcommit-with-no-testslabel.README.html
Regenerated from README.asc with asciidoctor. This also picks up README.asc changes already on master that the committed HTML lacked.
Full pgxntool-test suite run locally against this branch: 306/306 pass.
🤖 Generated with Claude Code