• Home
  • Features
  • Pricing
  • Docs
  • Announcements
  • Sign In

dangernoodle-io / breadboard / 29892069058
96%

Build:
DEFAULT BRANCH: main
Ran 22 Jul 2026 04:50AM UTC
Jobs 1
Files 190
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

22 Jul 2026 04:49AM UTC coverage: 96.375%. Remained the same
29892069058

push

github

web-flow
fix: hard-fail fence owner-fallback on a component-like directory gap (B1-1128) (#995)

- fence/_base.py: new is_component_like_gap() distinguishes a directory
  nested inside components/ or platform/<variant>/ (a real component
  dir discovery couldn't resolve, e.g. one missing a CMakeLists.txt
  anywhere on its branch) from a genuine loose file directly in that
  container (components/foo.c) or a path outside the convention
  entirely
- resolve_owner_fallback() raises UnresolvedComponentOwnerError for the
  former instead of silently returning the phantom "components"/
  "platform" owner name; the latter two shapes keep the existing
  WARN+counter fallback unchanged
- clamp/sat_sub/callback_slot's _component_of now delegate to
  resolve_owner_fallback instead of hand-rolling the same
  first-path-segment fallback three times
- production fence_cmd.run() does NOT clear discovery.build_index's cache
  (bbtool dispatches exactly one subcommand per process, so the cache is
  already empty on entry; a clear here would be a permanent production
  no-op). cache clearing belongs in tests only, where it is load-bearing:
  in-process multiple calls to run() against the SAME mutated root would
  otherwise see a stale pre-mutation index, surfaced by this change's
  hard-fail boundary turning two existing silent falsebacks into explicit
  UnresolvedComponentOwnerError
- consolidate six near-identical _run_fence_cli() copies from individual
  test files (test_fence.py, test_fence_callback_slot.py, etc.) into a
  shared fence_test_support.run_fence_cli() helper that defensively clears
  the cache on each invocation, per discovery.py's documented test
  convention
- verified against the real tree: fence --root . still reports 0 new
  markers / PASS across every family (no component-like gap exists in
  production today, matching the ticket's own assessment)


Claude-Session: https://claude.ai/code/session_01KzNnayrQ6MV929RF5c5GtZ

Co-authored-by: Claude Opus 4.8 <n... (continued)

7426 of 8016 branches covered (92.64%)

Branch coverage included in aggregate %.

12569 of 12731 relevant lines covered (98.73%)

2457.11 hits per line

Jobs
ID Job ID Ran Files Coverage
1 29892069058.1 22 Jul 2026 04:50AM UTC 190
96.38
GitHub Action Run
Source Files on build 29892069058
  • Tree
  • List 190
  • Changed 0
  • Source Changed 0
  • Coverage Changed 0
Coverage ∆ File Lines Relevant Covered Missed Hits/Line Branch Hits Branch Misses
  • Back to Repo
  • Github Actions Build #29892069058
  • 633d39c6 on github
  • Prev Build on main (#29891656205)
  • Next Build on main (#29893642544)
STATUS · Troubleshooting · Open an Issue · Sales · Support · CAREERS · ENTERPRISE · START FREE TRIAL · SCHEDULE DEMO
ANNOUNCEMENTS · TWITTER · TOS & SLA · Supported CI Services · What's a CI service? · Automated Testing

© 2026 Coveralls, Inc