Classify the Public Surface Before an OSS Crash Fix Hits Review

Описание и применение

Пошаговый процесс классификации публичной поверхности перед созданием PR для исправления ошибки в OSS, включающий воспроизведение, патч, тестирование и заполнение Semver Impact Card.

Задача

A small crash fix in a TypeScript CLI caused a breaking change because an exported helper’s return value was altered, yet the PR was closed without proper versioning analysis.

Результат

The workflow ensures that crash fixes are reproducible, minimally invasive, test‑covered, and semantically classified before PR creation, preventing accidental breaking changes.

Рабочий процесс

• Reproduce the failure on the reported tag and record repro.md
• Patch only the owning module and store patch.diff and files.txt
• Run existing test runner, add a new assertion for the fixed behavior
• Fill the Semver Impact Card with public surface classification and bump proposal

Доказательства

Reproduce the reported failure on the issue tag, not on a dirty local main.

Источник

Patch the owning module without drive‑by refactors or formatting sweeps.

Источник

Run the project's test runner plus one new assertion that binds the contract.

Источник

Copy the template into .patch-packet/semver-impact.md and complete every field on the same day as the diff.

Источник