No CI: 152 smoke tests, rigs and a fuzz harness exist but nothing runs them
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Newbie friendliness
- 78/100
Research direction
Start with homeboy.json and its scripts.lint and scripts.test recipes, then inspect the repository for the absent .github/workflows directory. Add a workflow that invokes those recipes for pull requests and pushes to main, and verify it reports status checks while a failing smoke test prevents a green result. Use tests/smoke-*.php as the suite being exercised.
Written by the indexing model from the issue text.
Description
Summary
This repository has no CI. There is no .github/workflows directory and no configuration for any other CI system. The only automation running against it is Dependabot's dependency graph updates.
Pull requests therefore report no checks reported, and nothing verifies a change on merge or catches a regression afterwards.
What already exists and is not being run
The repository is well equipped for verification. None of it is wired to anything:
| Asset | Count / detail |
|---|---|
| Smoke tests | 152 tests/smoke-*.php |
| Declared recipes | homeboy.json → scripts.lint, scripts.test |
| Rigs | rigs/mdi-native, rigs/mdi-primary, rigs/mdi-sqlite |
| Fuzz harness | fuzz/ |
| Benchmarks | bench/native-sqlite-differential.workload.json |
homeboy.json already declares exactly what a pipeline would need:
"scripts": {
"lint": ["php -l markdown-database-integration.php && for file in inc/*.php tests/*.php; do php -l \"$file\" || exit 1; done"],
"test": ["for test_file in tests/smoke-*.php; do php \"$test_file\" || exit 1; done"]
}
So the gap is wiring, not authoring. The full suite runs in seconds locally.
Why this matters more than usual here
MDI is not a peripheral plugin — when MARKDOWN_DB_MODE is primary it is the database. WP_Markdown_Native_WPDB replaces $wpdb and every option, post and table read on the site goes through it.
A silent read-side regression in this component does not surface as a failing feature. It surfaces as data that appears not to exist. #412 is a worked example: an indeterminate option read was reported as "this option is absent", which caused WordPress to regenerate auth_key/auth_salt and permanently destroy every credential encrypted under the previous salt. That happened twice in four days, and was not noticed for three days, because the only symptom was an unrelated remote HTTP 401.
That class of defect is precisely what a test suite catches and a human review does not.
The fix for it (#413) merged — or will merge — with zero automated verification. Its evidence is a local run and a manual mutation check recorded in the PR body. That is not a durable guarantee, and it will not stop the same read path regressing later.
Proposed
Add a workflow that runs the recipes this repository already declares:
scripts.lintandscripts.testfromhomeboy.jsonon pull requests and on pushes tomain.- Required status checks on
mainso a red suite blocks merge.
Worth considering once the basics are green:
- Run the
mdi-native/mdi-primary/mdi-sqliterigs, since backend parity is the property most likely to drift. - Schedule the differential fuzz harness rather than running it per-PR, to keep PR latency low while still exercising it.
152 smoke tests running on every PR is a large improvement over nothing, and needs no new test authoring.
Acceptance
- A pull request runs lint and the smoke suite and reports a status check.
- A failing smoke test blocks merge on
main. - The workflow invokes the recipes declared in
homeboy.jsonrather than duplicating them, so the local command and the CI command cannot drift.
AI assistance disclosure: Claude Sonnet 4.5 via Claude Code noticed this while opening #413 and finding no checks reported, inventoried the existing unused test assets, and drafted this issue at Chris Huber's request.
- Dominant language
- PHP
- Stars
- 7
- Forks
- 1
- Avg merge
- 3h 16m
- Merged PRs (30d)
- 44
Getting set up
This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from Automattic/markdown-database-integration
-
Difficulty 5/5 Over a week Newbie friendliness 25/100
Automattic/markdown-database-integration#424 ·
Maintainers usually reply within 1 day
-
Difficulty 3/5 1-2 days Newbie friendliness 78/100
Automattic/markdown-database-integration#420 ·
Maintainers usually reply within 1 day
-
Difficulty 5/5 Over a week Newbie friendliness 35/100
Automattic/markdown-database-integration#415 · 1 comment ·
Maintainers usually reply within 1 day
-
Difficulty 3/5 1-2 days Newbie friendliness 72/100
Automattic/markdown-database-integration#401 ·
Maintainers usually reply within 1 day
-
Difficulty 5/5 Over a week Newbie friendliness 20/100
Automattic/markdown-database-integration#377 ·
Maintainers usually reply within 1 day
All issues in Automattic/markdown-database-integration
Similar issues
-
sync-en
Difficulty 2/5 1-3 hours Newbie friendliness 75/100
Maintainers usually reply within 1 day
-
sync-en
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
Maintainers usually reply within 4 days
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 65/100
ProfessionalWiki/NeoWiki#1637 ·
Maintainers usually reply within 1 day
-
Перевод устарел
Difficulty 1/5 Under an hour Newbie friendliness 85/100
-
bug
Difficulty 2/5 Half a day Newbie friendliness 76/100
m3ue/m3u-editor#1604 ·
Maintainers usually reply within 1 day