* fix(ci): give a Helm chart change an area label
deploy/ holds the Helm chart and nothing else. No AREA_RULES entry matched
that tree, so a PR touching only the chart -- and the backend/tests/test_helm_*.py
that comes with it -- arrived with no area label at all: #8123 reached merge
with just its size and risk labels, and #8161 only carried area:docs because
it happened to touch a README. Nothing routes those PRs to a reviewer who
owns the chart.
The chart is release/deploy tooling in the same sense as scripts/: bump_version.sh
rewrites its versions, verify_versions.sh and the chart workflow gate its
rendering, and area:ci already owns scripts/. Add deploy/ to that rule.
The offline replay in tests/test_label_declarations.py already proves the
labels the workflow would apply; it only covered the backend package trees.
Move its Node harness into a shared helper, reuse it, and add the chart cases
so this gap cannot reopen.
* Update CI area label matching for helm directory
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
---------
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* fix(ci): label every harness tree as backend
`area:backend` enumerated the harness subdirectories that map to it instead
of the trees that map elsewhere, so thirteen of the twenty-eight directories
under `deerflow/` matched no area rule at all: community/, authz/, scheduler/,
storage/, tui/ and eight more. A PR that only touched `deerflow/community/`
therefore came out unlabelled, or with `area:docs` alone when it also edited a
markdown file — #7482 (crawl4ai) and #7826 (jina) are merged examples.
Match the harness as a whole and exclude the six trees that carry a more
specific label (agents, subagents, reflection, sandbox, skills, mcp). Checked
against all 28 directories: the 13 unlabelled ones now match `area:backend`,
and the sandbox / skills / mcp / agents / frontend results are unchanged.
* fix(ci): cover backend package paths in PR triage
---------
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
* fix(ci): consolidate PR/issue labeling into one triage.yml; fix reviewing crash & label thrash
- Replace pr-labeler + pr-triage + issue-triage with a single triage.yml; drop actions/labeler.
Its sync-labels removed labels outside its config (clobbered size/risk/needs-validation and
could clobber maintainer labels). Area is now computed in-script and reconciled only within
owned namespaces (area:/size//risk:/needs-validation); first-time/reviewing are add-only.
- reviewing: gate on author_association in {OWNER,MEMBER,COLLABORATOR} + user.type==='User'
instead of getCollaboratorPermissionLevel, which 404'd on bot reviewers ('Copilot is not a
user') and crashed the job. Excludes all review bots with no denylist and no API call.
- Read live state (listFiles + listLabelsOnIssue) not the stale event payload, so rapid
synchronize events converge instead of thrashing. Size churn excludes lockfiles/snapshots.
* fix(ci): read labels live via paginate in reviewing & issue-triage jobs
Address review feedback on #3455:
- reviewing: listLabelsOnIssue now paginates (per_page:100) instead of the
default 30, matching pr-labels, so a 'reviewing' label is never missed on
PRs with many labels.
- issue-triage: read live labels via the API instead of the event payload,
consistent with the live-state reads documented in the header.