Repository navigation
Conversation
scripts/wiring.py, MIP-0076 §5.1 without the gate: it reads .gitmodules, the devkit's tree under .devkit (else its own), every repo's workflows, pin files, and the scripts, justfiles, Dockerfiles, docker-compose*.yml and build.sbt naming a pin or image. It renders four tables (artifact, dispatch, pin bump, deploy) between a file's wiring:start/wiring:end markers. Beyond §5.1's sources, per the tasks file's decisions: branch artifacts (api-docs, site-data) from the `<branch>-push.sh` idiom, read by any file fetching the branch (Decision 1); compose `image:` lines as image readers (Decision 2); a workflow step opening a PR in or pushing to another marola-dev repo (Decision 3). A devkit reusable workflow's effects are attributed to each caller with the call's ref, and its inputs resolved from the caller's `with:`. Closes #25 Tested: wiring --self-test (13 cases, red against a stubbed scan, green after); a run over a fresh --recurse-submodules umbrella clone at 0145e12 Cost: ~$19.79 · diff-size estimate (cost-split --estimate-commit) Co-Authored-By: Claude <noreply@anthropic.com>
PyYAML joins the flake's python, which every tool wrapper and the self-tests check run with. Refs #25 Tested: just quality line by line without nix (ruff check/format, shellcheck, actionlint, tests/self-tests.sh, agents-check, docs_lint); nix flake check not run, no nix here Cost: ~$2.40 · diff-size estimate (cost-split --estimate-commit) Co-Authored-By: Claude <noreply@anthropic.com>
… missing submodules Review round 1 on #25. A call is now resolved with `git show <ref>:.github/workflows/<X>` in the devkit checkout (`--devkit DIR`, default `.devkit`, else this repo), so a caller still on an old tag shows that tag's dispatch type. An unknown ref falls back to the tree with a warning and "(resolved at <version>)" in the row; a callee missing at its ref is a warning, not a silent drop. An uninitialised submodule is an error instead of a near-empty block. The umbrella is named by `--name` (default `marola`), not the enclosing git config. Smaller: `event_type=` is matched after substitution; `-push.sh` on a `--self-test` line is ignored; a branch reader needs a `git … fetch` line; the self-exclusion is only the devkit's `scripts/wiring.py`; more than three push paths render as "touching N paths"; `git add` flags are not PR files; malformed YAML names its file; the README lists `wiring`. Closes #25 Tested: wiring --self-test (16 cases; the new ref and submodule cases fail with their fix stubbed out); ruff check/format, actionlint, shellcheck, tests/self-tests.sh, agents-check, docs_lint; a run over the fresh umbrella clone with --devkit at a tagged devkit checkout Cost: ~$8.59 · diff-size estimate (cost-split --estimate-commit) Co-Authored-By: Claude <noreply@anthropic.com>
h0ffmann
pushed a commit
that referenced
this pull request
Oct 8, 2026
One command, run twice around a human merge. On a main whose versions are not X.Y.Z it writes the version into plugin.json, flake.nix and every documented pin, adds a CHANGELOG heading with the commits since the last tag, and pushes chore/release-vX.Y.Z for a PR. On a main that already says X.Y.Z everywhere it tags vX.Y.Z and pushes the tag; release.yml checks the versions against the tag and publishes the GitHub release. `just quality` now fails when the versions disagree. The consumer pin-bump PRs from #47 wait on the token decision there. Part of #47. Tested: release.py --self-test; --check on main (29 locations at 0.6.0) and against 0.5.1 (fails); --dry-run of both paths in a fresh clone; ruff, actionlint, tests/self-tests.sh, docs-lint, agents-check; trial merge with #32 and #33 passes the same gates Cost: n/a (no measured figure in this cloud session) Co-Authored-By: Claude <noreply@anthropic.com>
h0ffmann
added a commit
that referenced
this pull request
Oct 8, 2026
One command, run twice around a human merge. On a main whose versions are not X.Y.Z it writes the version into plugin.json, flake.nix and every documented pin, adds a CHANGELOG heading with the commits since the last tag, and pushes chore/release-vX.Y.Z for a PR. On a main that already says X.Y.Z everywhere it tags vX.Y.Z and pushes the tag; release.yml checks the versions against the tag and publishes the GitHub release. `just quality` now fails when the versions disagree. The consumer pin-bump PRs from #47 wait on the token decision there. Part of #47. Tested: release.py --self-test; --check on main (29 locations at 0.6.0) and against 0.5.1 (fails); --dry-run of both paths in a fresh clone; ruff, actionlint, tests/self-tests.sh, docs-lint, agents-check; trial merge with #32 and #33 passes the same gates Cost: n/a (no measured figure in this cloud session) Co-authored-by: Claude <noreply@anthropic.com>
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
1 participant
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.
Size: 525 changed lines (
wiring.py512, ~110 of them self-test fixtures). Not split: rendering is ~35 lines, and the row's own last test is the generated block.Real run over a fresh
--recurse-submodulesumbrella clone with--devkitpointing at a devkit clone that has tags: every fact of today's hand-written REPOS.md block is produced, every reusable-workflow call resolves at its caller's own ref, and two runs are byte-identical. The block is in the task report; row 5 lands it.For rows 5, 11 and 15: nix's
.devkitis a store path, not a git checkout, so those rows pass--devkit <clone with tags>(otherwise each call warns and renders as "resolved at ").