registrystack / registrystack/registry-stack
nightly coverage planner: detect drift between the workflow's flags map and SHARDS
- Dominant language
- Rust
- Stars
- 2
- Forks
- 0
- Avg merge
- 2h 57m
- Merged PRs (30d)
- 128
Description
## Context
PR #851 fixed #789 by defaulting the nightly coverage flag to the shard name (`.github/workflows/nightly-rust-coverage.yml`, `flags.get(name, name)`). The `flags` dict is keyed by shard name and lives in the workflow while `SHARDS` lives in `.github/scripts/ci_changes.py`; nothing checks that every key in `flags` still names a shard, so a renamed or removed shard leaves a stale mapping that is silently ignored.
## Acceptance
- `test_ci_changes.py` (or a workflow-structure test) asserts every `flags` key is a `SHARDS` key, and the planner refuses a stale one.
Contributor guide
Research direction
Compare the flags map in .github/workflows/nightly-rust-coverage.yml with SHARDS in .github/scripts/ci_changes.py. Start with test_ci_changes.py and inspect how the planner handles shard names. Done means stale flag keys are detected by a test or workflow-structure check and the planner refuses them.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, rust
- Domain
- build-system, ci-cd, testing
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Active
- Clarity
- Clearly specified
- Newbie friendliness
- 76/100