chore(release): sync main back into development - #1714
Closed
github-actions[bot] wants to merge 20 commits into
Closed
chore(release): sync main back into development#1714github-actions[bot] wants to merge 20 commits into
github-actions[bot] wants to merge 20 commits into
Conversation
OpenConnector contained several MySQL-specific SQL constructs that fail on
PostgreSQL (MySQL silently coerces; PostgreSQL errors). These broke job
creation, synchronization, log retention and the statistics dashboard on
PostgreSQL deployments.
- find(): drop `eq('id', $stringId)` from the non-numeric branch in Source,
Job, Mapping, Endpoint and Rule mappers. Comparing the bigint `id` column to
a uuid/slug threw SQLSTATE[22P02] "invalid input syntax for type bigint".
- findByConfiguration() (6 mappers): replace MySQL-only JSON_CONTAINS + backtick
table names with a cross-DB QueryBuilder LIKE, matching findByArgumentIds.
- RuleMapper::getMaxOrder(): replace COALESCE(MAX(`order`)) raw SQL with
func()->max('order'), which quotes the reserved word per platform.
- SourcesController: make the slow-requests JSON filter platform-aware
(Postgres ::jsonb->>, SQLite json_extract, MySQL JSON_EXTRACT).
- SettingsService::rebase(): extract a platform-aware setExpiryDates() helper
(Postgres/SQLite/MySQL date arithmetic; CAST(? AS bigint) so PDO's text-bound
param multiplies an interval on PG) and a portable columnExists() replacing
SHOW COLUMNS; emit empty-string/zero-date predicates only for MySQL.
- SettingsService::getStats(): drop backtick-quoted table names so COUNT(*)
runs on PostgreSQL instead of silently returning 0.
Validated end-to-end against PostgreSQL: job creation, find-by-slug (404 not
500), rebase (success, no errors) and stats (real counts).
…2 June 2026) + conduction.nl app link
…settings' (#37) from hotfix/postgresql-compatibility into main Reviewed-on: https://codeberg.org/Conduction/openconnector/pulls/37 Reviewed-by: WilcoLouwerse <wilcolouwerse@noreply.codeberg.org>
Inhoud exact Codeberg-main (tree ongewijzigd); banner-commit alleen als ancestor zodat de push fast-forward is. Migratie 16-07.
`code-quality.yml` on `main` had no `push:` trigger, only `pull_request`. So a
push to the release branch ran no jobs at all.
Measured 2026-08-09: openconnector `main` has ZERO Code Quality runs on
record. Not failing — absent. A branch with no runs shows no red, which is
exactly why this went unseen while every other measurement in the CI
programme was made on `development`.
`development` has carried the correct trigger for some time:
push:
branches: [main, development, feature/**, bugfix/**, hotfix/**]
It never reached `main` because the corrected workflow travels the release
train (development -> beta -> main) and no release has carried it across.
Deliberately workflow-only. This does not fix whatever `main` may be failing —
it makes `main`'s state VISIBLE, which has to come first. Expect the first run
to be red; that red is information the branch has not produced in months.
See ConductionNL/.github#285.
Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Deleting orphaned register/schema objects during sync cleanup was silently broken in two ways after OpenRegister's move to per-schema magic tables: - updateTargetOpenRegister() called the removed ObjectService::delete() method, fatally crashing the delete path whenever it actually ran. - findAllBySynchronizationAndSchema() joined against the retired, now-permanently-empty openregister_objects table, so it never found any contracts to clean up in the first place, regardless of what was orphaned. Switch to the current ObjectService::deleteObject() API, join against the schema's actual magic table (OCA\OpenRegister\Db\MagicMapper:: TABLE_PREFIX + registerId + schemaId), and stop crashing the whole cleanup loop when a single delete is blocked by a referential integrity constraint or the contract has already vanished. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
fix: repair deleteInvalidObjects for OpenRegister magic tables
* ci(release): drop inert beta-release.yaml (triggered on branch 'never') * ci(release): drop inert stable-release.yaml (triggered on branch 'never') * ci(release): drop inert unstable-release.yaml (triggered on branch 'never') * ci(release): drop inert push-beta-to-beta-release.yaml (triggered on branch 'never') * ci(release): drop inert push-development-to-development-release.yaml (triggered on branch 'never')
Dependabot reads its config from the DEFAULT branch (main here), whose copy carried no target-branch - so every PR was opened against main where branch-protection rejects it. This is development's file verbatim.
Release: merge development into beta
main held 16 commit(s) beta did not. Merged with -s ours: beta's tree is kept BYTE FOR BYTE and only the ancestry is recorded, so the beta -> main promotion stops conflicting on files where beta is simply newer. Not brought over -- beta is hundreds of commits ahead of main, so these are the OLDER copies, and several are dead Forgejo/Codeberg CI that development deliberately removed: .forgejo/workflows/documentation.yml .forgejo/workflows/release-beta.yml .forgejo/workflows/release-stable.yml .github/dependabot.yml .github/workflows/beta-release.yaml .github/workflows/code-quality.yml .github/workflows/push-beta-to-beta-release.yaml .github/workflows/push-development-to-development-release.yaml .github/workflows/stable-release.yaml .github/workflows/unstable-release.yaml appinfo/info.xml lib/Controller/SourcesController.php lib/Db/EndpointMapper.php lib/Db/JobMapper.php lib/Db/MappingMapper.php lib/Db/RuleMapper.php lib/Db/SourceMapper.php lib/Db/SynchronizationContractMapper.php lib/Db/SynchronizationMapper.php lib/Service/SettingsService.php lib/Service/SynchronizationService.php
Release: merge beta into main
The 0.3.9 release bumped the version on main. Without this, development stays behind main and the next development -> main promotion conflicts on the version file. Version files resolve to development's side, which is the higher line, so this never moves a version backwards.
Contributor
|
Superseded. fleet-back-merge.yml could not merge these while it waited for green on branches red for unrelated reasons, so one accumulated per run (107 open across the fleet). ConductionNL/.github#648 removed that wait, and the sweep that followed carried and merged the same ancestry immediately. This pull request changes zero files, so closing it loses nothing: the ancestry it recorded is already on the branch. |
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
4 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.
Opened by the release workflow after publishing 0.3.9 from
main.A release commits a version bump onto the branch it ran from. Without carrying
that back,
developmentfalls behindmain, and since both branches thenchanged the version file independently, the next
development -> mainpromotion conflicts on it — every time.
Version files are resolved to development's side, the higher line, so this
cannot move a version backwards. Any other conflict stops the workflow instead
of being resolved automatically.
Merge this, do not squash. A squash copies the files and discards the
relationship: the merge base would not move and the next promotion would
conflict exactly as before. A 0-file diff here is normal and is not a no-op —
recording the ancestry is the whole point.