ci: Add nightly model regeneration from the published OpenAPI spec - #983
Merged
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #983 +/- ##
==========================================
- Coverage 94.65% 94.63% -0.03%
==========================================
Files 58 58
Lines 5259 5258 -1
==========================================
- Hits 4978 4976 -2
- Misses 281 282 +1
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
vdusek
force-pushed
the
ci/nightly-openapi-spec-sync
branch
from
August 3, 2026 10:19
7539cdd to
fed06a3
Compare
vdusek
marked this pull request as ready for review
August 3, 2026 11:58
vdusek
force-pushed
the
ci/nightly-openapi-spec-sync
branch
from
August 3, 2026 11:58
dc06b74 to
85e396c
Compare
janbuchar
approved these changes
Aug 3, 2026
janbuchar
left a comment
Contributor
There was a problem hiding this comment.
Looks sound at a glance, let's battle test it
vdusek
added a commit
to apify/apify-docs
that referenced
this pull request
Aug 3, 2026
#2835) Removes the cross-repo dispatch that opened a companion model-regeneration PR in `apify-client-python` for every spec-changing docs PR, and rewrites the `AGENTS.md` section describing it. `apify-client-python` now pulls the spec on its own nightly schedule instead, so nothing on this side needs to push to it or be merged in lockstep with a client PR. Why the dispatch goes away: - The client can only generate models from the **published** spec, so it can never be genuinely ahead of a deployed docs change - the companion PR was opened against an unpublished bundle it would have to regenerate from anyway. - Two mechanisms writing the same generated files made them diverge. The dispatch reused one long-lived branch per docs PR and regenerated with the codegen tooling from that stale branch, which silently reverted fixes that had landed on the client's master in the meantime. - Every such PR opened with a placeholder `TODO` title that a human had to replace before it could be merged, so unmerged PRs and their stale branches piled up. Also drops what only the removed jobs needed: the `closed` pull request trigger, the `if:` guard skipping lint on close events, and the `pull-requests: read` permission. **Merge this before** apify/apify-client-python#983 - that PR deletes `manual_regenerate_models.yaml`, which the dispatch removed here still calls. *✍️ Drafted by Claude Code*
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.
Replaces the apify-docs-triggered model regeneration with a nightly job, and records which published OpenAPI specification the generated models came from.
What changes
on_schedule_regenerate_models.yaml- nightly at 02:00 UTC plus manual dispatch. Regenerates on master and opens a pull request when the models change.manual_regenerate_models.yamlis deleted.scripts/openapi_spec.py- downloads and validates the published specification into git-ignoredtmp/(both codegen passes read that one copy), then records itsinfo.versioninpyproject.tomlunder[tool.apify.openapi-spec]. The specification itself is not committed._models.pyloses thePlan.available_proxy_groupsdocstring - accumulated codegen drift from thedatamodel-code-generatorbumps (chore(deps): lock file maintenance #968, chore(deps): lock file maintenance #975), not a specification change.Why
The old workflow was dispatched by apify-docs and checked out a permanent, never-rebased branch before regenerating, so it generated with that branch's stale tooling. In #979 that produced
_literals.pyin the pre-#941 closedLiteral[...]form; merging it would have silently reverted the enum relaxation. It also opened with aTODOtitle that blockedpr-title-check- a design that needs human action to become mergeable, which is what let #979 rot for 20 days while the same spec changes were landed by hand (#923, #936, #947, #960, #974).The nightly job always generates on master and rebuilds its
ci/regenerate-modelsbranch from master instead of appending, so the diff is always "current spec vs current master" and can't resurrect a stale generated file. It opens with a mergeablechore:title; reviewers retitle tofix:/feat:when the diff is user-facing.Notes
SLACK_WEBHOOK_URLrepository secret for the failure alert.components/version.yamlin a follow-up[skip ci]commit, so a deploy can publish new content under the old stamp. A moved value proves the specification changed; an unchanged one proves nothing.✍️ Drafted by Claude Code