github-tools / github-tools/github-release-notes
Possible bug in version ranges for generating a release with --tags
- Vorherrschende Sprache
- JavaScript
- Sterne
- 897
- Forks
- 313
- PR-Merge-Kennzahlen
- Keine gemergten PRs in 30 T.
Beschreibung
## Background
> I originally started this issue for a feature request, but I see now that this might be a bug. I've updated the title to reflect this.
In our current setup, we utilize `master` only development, in which we deploy to different environments based on tags:
// Deploys to dev test
npm version 1.0.1-beta.x
// Deploys to acceptance test
npm version 1.0.1-rc.x
// Deploys to production
npm version 1.0.1
This results with the usage of `gren` that only the `1.0.1-beta.x` contains the full release notes since last release, and not `1.0.1`.
Is it possible to have a feature that enables `gren` to concatenate, or use the preferred release notes for a tag? For example:
gren release --prerelease --override --tags=v1.2.0..v1.2.0-beta.0
## Conclusion
I see that this might be a potential bug, as `gren` is probably not accepting version ranges in tags correctly, specifically, between prereleases?
## Another relevant scenario
Old production release is `0.58.1`, I have a `1.2.0-rc.0` version not in production yet.
gren release --override --tags=v1.2.0..v0.58.1
This will populate release notes with ALL the commits and tags created historically between those tags, including tags like `1.3.0-beta.0` 🤦♂️
Beitragsleitfaden
Rechercherichtung
No source files or tests are named. Start by reproducing the two reported gren release commands with the stated tag ranges, then trace the tag-range handling and existing release-generation tests. Done means prerelease and production ranges select the intended commits and exclude unrelated historical tags.
Vom Indexierungsmodell aus dem Issue-Text verfasst.
Bewertung
- Tech-Stack
- javascript
- Bereich
- cli, release
- Issue-Typ
- Bug
- Schwierigkeit
- 4/5
- Geschätzter Aufwand
- 3-5 Tage
- Aktivitätsstatus
- Veraltet
- Klarheit
- Größtenteils klar
- Anfängerfreundlichkeit
- 35/100