Repository navigation
CI: fix PG 19 RC1 build and lolor TAP tests - #653
Merged
Merged
Conversation
ensure_lolor() cloned lolor's main branch, so an unreleased commit there could break the lolor TAP tests for reasons unrelated to spock. lolor publishes no GitHub releases, only tags, so resolve the highest vN.N.N tag and clone that. Pre-release tags are excluded by the pattern, and a tag that cannot be resolved makes the caller skip rather than fall back to main.
Contributor
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info
📝 Walkthrough
Priority: ⬇️ Low Merge Risk: ⚪ Minimal · up to
|
| Check name | Status | Explanation |
|---|---|---|
| Docstring Coverage |  |
Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 1 functions across 1 files. (1 skipped: 1 … |
| Linked Issues check | |
Check skipped because no linked issues were found for this pull request. |
| Out of Scope Changes check | |
Check skipped because no linked issues were found for this pull request. |
| Title check | |
The title clearly summarizes the two main changes: the PostgreSQL 19 build fix and the lolor TAP test fixes. |
| Description check | |
The description directly explains the PostgreSQL 19 API change, lolor version pinning, preload configuration, and TAP test behavior. |
Full details: Docstring Coverage
-
Explanation
Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 1 functions across 1 files. (1 skipped: 1 unsupported.)
✨ Finishing Touches
-
🧪 Generate unit tests (beta)
-
- Commit to this branch
- Create a new PR
-
- Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts
A rabbit checks the tags at dawn,
Then sets the preload before moving on.
Three nodes test migration’s way,
While PostgreSQL versions call as they may.
I nibble greens and hop away.
Comment @coderabbitai help to get the list of available commands.
Up to standards ✅🟢 Issues
|
| Metric | Results |
|---|---|
| Duplication | 0 |
NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.
ibrarahmad
changed the title
PostgreSQL commit 2f7bf51b7b (in REL_19_RC1) added a skipduplicates argument to RelationFindReplTupleByIndex(): the caller now says whether the first index match may be taken without comparing it against the search slot, instead of the function re-deriving it from the index. Back branches keep the old signature and add RelationFindReplTupleByIndexExt() instead, so only PG 19 is affected. FindReplTupleInLocalRel() is only called with the replica identity index recorded in the relation map, so pass true. The -Werror and regression workflows build the newest REL_19_* tag and started failing once REL_19_RC1 appeared.
ensure_lolor() builds lolor from its master branch, which now refuses CREATE EXTENSION unless the library is in shared_preload_libraries. 032_lolor_largeobject_repset and 033_zodan_lolor_add_node therefore failed at their first CREATE EXTENSION lolor. Remember a successful ensure_lolor() and add lolor to shared_preload_libraries for every cluster created after it.
ibrarahmad
force-pushed
the
lolor-latest-release-tag
branch
from
October 16, 2026 14:23
4142a92 to
1fb2efd
Compare
ibrarahmad
force-pushed
the
lolor-latest-release-tag
branch
from
October 16, 2026 17:26
a8363fa to
1fb2efd
Compare
lolor's migrate_from_native(), and the drop that leaves the large objects in pg_catalog for it to pick up, arrived after v1.2.3, the newest release ensure_lolor() now builds. Skip that round-trip when the installed lolor lacks the function, and expect only the newly created large object afterwards. The drop, the recreate and the replication set checks around it run either way.
mason-sharp
approved these changes
Oct 17, 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
3 participants
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.
Supersedes #651, whose two commits are included here with one comment re-wrapped to satisfy
pgindent, plus one further change to how the lolor TAP tests obtain lolor.PG 19 RC1 picked up postgres/postgres@2f7bf51b7b, which added a
skipduplicatesargument toRelationFindReplTupleByIndex()and turned the-Werror, regression and random-delays workflows red.FindReplTupleInLocalRel()always passes the replica identity index from the relation map, so it passestrue. Back branches keep the old signature and are untouched.ensure_lolor()intests/tap/t/SpockTest.pmcloned lolor's main branch, so an unreleased commit there could fail tests 032 and 033 for reasons that have nothing to do with spock. It now clones the newest release tag. lolor publishes no GitHub releases —GET /repos/pgEdge/lolor/releasesreturns an empty list — so the release is the highestvN.N.Ntag, currentlyv1.2.3. Pre-release tags such asv1.2.3-rc.1carry a suffix and are excluded by the pattern, andsort -Vorders the rest by version rather than lexically. If no tag resolves,ensure_lolor()returns false and the caller skips, rather than silently falling back to main.Clusters created after a successful
ensure_lolor()also add lolor toshared_preload_libraries. v1.2.3 does not require that, but lolor main does, so the tests keep working once the next release is tagged.This pin takes
033_zodan_lolor_add_nodefrom 19 failing subtests to 9: theDROP EXTENSION lolorfailures reported in #651 go away, because pgEdge/lolor#62 landed after v1.2.3. The nine that remain are the large object migration assertions, which no released lolor can satisfy, so the last commit runs that round-trip only where lolor carriesmigrate_from_native()and expects just the newly created object otherwise. The drop, the recreate and the replication set checks around it run either way. Verified on PG 18: 79/79 against lolor 1.3.0 where the migration runs, and 79/79 against v1.2.3 where those six are skipped.