shields.io can't fetch JSON from our self-hosted Gitea instance (not
publicly reachable), so badges showed "unknown". Switched to generating
self-contained SVG badge files that are served directly by Gitea's raw
file API — no external service needed.
Changes:
- generate_badges.py: Added render_svg() to produce shields.io-style
SVG badges with gradient, rounded corners, and Verdana font
- Replaced xml.sax.saxutils.escape with a simple _xml_escape() to
avoid bandit B406 warning (no defusedxml dependency needed)
- CI workflow: Push .svg files instead of .json to badges branch
- README.md and docs/index.md: Updated badge URLs to use raw SVG
from the badges branch instead of shields.io endpoint
Also fixed:
- Coverage regex now handles 100% without decimal (was 100.00%)
- doc_coverage.py: Removed redundant % in pct variable that caused
double-percent (100%%) in output
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-06-22 00:50:22 +02:00
Emil SimeonovandDevin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The badges job failed because git commit triggered pre-commit hooks
on the orphan branch where .pre-commit-config.yaml was removed by
git rm -rf . Added --no-verify and PRE_COMMIT_ALLOW_NO_CONFIG=1.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-06-22 00:38:27 +02:00
Emil SimeonovandDevin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The badges job failed because the checkout step didn't include the
REPO_TOKEN secret, so git push to the badges branch had no credentials.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Add /docs/ directory with user and technical documentation extracted from README, AGENTS.md, and source code. Add scripts/sync_wiki.py to sync docs to Gitea wiki via API. Add scripts/doc_coverage.py to check CLI commands, modules, and CI scripts are documented. Add sync-wiki.yml workflow for auto-sync on merge and release. Slim down README.md to lean entry point. 28 new unit tests, 100% coverage maintained.
Closes GRM-36
Gitea Actions does not implement fail-fast/max-parallel for matrix jobs,
so a failing runner does not stop the others. Added molecule_ci_guard.py
which polls the Gitea API in a background thread. If any other molecule
runner reports failure, the current runner kills its molecule subprocess
and exits early.
CI returns to a 3-runner matrix; each runner executes its assigned pairs
in parallel via run_molecule_parallel.py, guarded by molecule_ci_guard.py.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-06-20 23:37:52 +02:00
Emil SimeonovandDevin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Gitea Actions does not honor fail-fast/max-parallel for cancelling
other matrix runners when one fails. Use a single molecule job that
runs all (scenario, platform) pairs via run_molecule_parallel.py.
This gives true parallel execution + immediate termination on the
first failure, which is what we need to debug efficiently.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-06-20 23:25:49 +02:00
Emil SimeonovandDevin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Added scripts/run_molecule_parallel.py to run a runner's assigned
(scenario, platform) pairs in parallel. If any subprocess fails, the
remaining ones are terminated with SIGTERM/SIGKILL and the runner
exits immediately. This gives fast feedback without continuing to run
tests that are guaranteed to fail for the same reason.
CI workflow now calls this script per matrix runner. Added fail-fast and
max-parallel for best-effort cancellation across runners.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-06-20 23:19:25 +02:00
Emil SimeonovandDevin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Use a single molecule test job (no matrix) on ubuntu-2204 only. This
stops the workflow immediately when the first scenario fails instead of
wasting time running 3 parallel runners that all fail for the same
reason. Multi-platform will be restored once the base scenarios pass.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-06-20 23:08:10 +02:00
Emil SimeonovandDevin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
With fail-fast: true and max-parallel: 1, runner 0 must complete
before runner 1 starts. If runner 0 fails, runners 1 and 2 are
cancelled. This gives immediate feedback on the first failure
instead of waiting for all 3 to fail in parallel.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-06-20 22:42:58 +02:00
Emil SimeonovandDevin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Three changes:
1. CI: add set -e and fail-fast: true to stop on first molecule failure
instead of continuing (all pairs fail for same reason anyway)
2. Docker APT repo: use copy module to write sources.list directly
instead of apt_repository module which wasn't picking up the repo
3. Fix arch mapping: ansible_facts returns x86_64, Docker repo needs amd64
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-06-20 22:31:08 +02:00
Emil SimeonovandDevin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Three major improvements:
1. Rootless Docker refactor: Removes docker/binary modes, unifies to
rootless Docker with per-runner system users. Each runner gets its
own rootless Docker daemon, systemd user service, and isolated
environment. Simplifies CLI (removes --mode option), Ansible role
(single code path), and molecule scenarios (removes binary scenario).
2. Auto-merge fix: Fixes status check context mismatch in branch
protection (was requiring "lint", "unit-tests", "molecule-tests" but
actual contexts are "CI / quality", "CI / molecule-tests*"). Adds
retry/wait logic to auto_merge.py that polls commit statuses for up
to 15 minutes before attempting merge, eliminating the chicken-and-egg
problem where auto-merge would fail because CI hadn't completed yet.
3. Molecule platform matrix: Adds OS platform matrix to CI — all 6
scenarios now run on all 4 supported OSes (ubuntu-2204, ubuntu-2404,
debian-12, archlinux) = 24 test pairs distributed across 3 parallel
runners. Updates distribute_molecule.py to distribute (scenario,
platform) pairs. Updates Makefile with molecule-all target for
local multi-platform testing.
Generated with [Devin](https://devin.ai)
Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
emil
