You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: CONTRIBUTING.md
+3-1Lines changed: 3 additions & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -40,6 +40,7 @@ For a faster loop while you are iterating, use a narrower command on the area yo
40
40
- Lint: `./scripts/lint.ps1` or `bash ./scripts/lint.sh`
41
41
- Smoke checks: `./scripts/smoke-languages.ps1` or `bash ./scripts/smoke-languages.sh`
42
42
- Compiled-language builds: `./scripts/build-all.ps1` or `bash ./scripts/build-all.sh`
43
+
- Generated artifact cleanup: `./scripts/clean-artifacts.ps1` or `bash ./scripts/clean-artifacts.sh`
43
44
44
45
If you are changing one C++ file and want a local spot check before the full pass, you can still compile it directly with:
45
46
@@ -59,9 +60,10 @@ These smoke checks also compile standalone C# exercises by generating temporary
59
60
TypeScript checks restore Node dependencies from `package-lock.json`, compile with `tsc`, and execute the emitted JavaScript with `node`.
60
61
61
62
The public PowerShell and Bash scripts are thin wrappers over the shared Python automation core in `scripts/automation.py`. Curriculum validation and smoke target metadata live in `scripts/automation_manifest.json`.
63
+
The artifact cleanup command removes generated build outputs, reports, temporary binaries, and exercise report files while keeping restored dependencies such as `node_modules`.
62
64
63
65
Use [EDUCATIONAL_EXAMPLE_REVIEW_RUBRIC.md](EDUCATIONAL_EXAMPLE_REVIEW_RUBRIC.md) when reviewing `example/main.*` files for teaching clarity and parity.
64
-
During focused cleanup work, run `python scripts/automation.py audit-education-quality --fail-on-findings` to make learner-quality findings fail locally instead of only writing the advisory report.
66
+
`verify-repo` fails on blocking education-quality findings: low example-comment ratio, missing output explanation markers, or boilerplate comments. Oversized example findings remain advisory. During focused cleanup work, run `python scripts/automation.py audit-education-quality --fail-on-findings` to make every learner-quality finding fail locally.
65
67
66
68
4. Update related README files when behavior or structure changes.
67
69
5. Open a pull request with a clear description of what changed and why.
GitHub Actions validates links, README structure, module completeness, checkpoint completeness, documentation sync, compiled-language builds, multi-language smoke checks, and Linux lint checks for C++, Python, Go, C#, and TypeScript.
200
202
201
203
The public PowerShell and Bash scripts remain the supported entrypoints, but they now delegate to a shared Python automation core under `scripts/automation.py` backed by `scripts/automation_manifest.json`.
202
204
205
+
Use `clean-artifacts` when you want to remove generated build outputs, reports, temporary binaries, and exercise report files without removing dependencies such as `node_modules`.
206
+
203
207
The multi-language smoke scripts also compile standalone C# exercises by generating temporary validation projects during the check and compile TypeScript programs before executing their smoke targets.
204
208
205
209
Use [EDUCATIONAL_EXAMPLE_REVIEW_RUBRIC.md](EDUCATIONAL_EXAMPLE_REVIEW_RUBRIC.md) to keep entry examples pedagogically consistent during reviews. The education audit command is advisory and writes markdown/json findings without failing CI.
206
210
Use [EDUCATIONAL_ANTI_PATTERN_BACKLOG.md](EDUCATIONAL_ANTI_PATTERN_BACKLOG.md) for the prioritized anti-pattern vs corrected-example expansion plan.
207
211
208
-
When you want to enforce the education audit during focused cleanup work, run:
212
+
`verify-repo` now fails on blocking education-quality findings: low example-comment ratio, missing output explanation markers, or boilerplate comments. Oversized example findings remain advisory. When you want the stricter local cleanup mode that also fails on oversized examples, run:
0 commit comments