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
Currently, we'll run cargo semver-checks to prevent API changes without bumping the major version (or minor version under 0.x). Essentially, this requires each PR with breaking changes to bump the major version in place.
Imagine that we have 10 PRs with breaking changes merged one day and we intend to publish them together. Then we have to bump the version 10 times under current restrictions. 😕 Also, I don't think it's a good idea to bump the version in place. A separate PR (or commit) can be much clearer.
So I'm wondering whether we should only run semver test on release (publish) instead?
The text was updated successfully, but these errors were encountered:
IIRC semver compares with latest version on crates.io instead of main? Therefore only the first PR needs to bump version
Yes, since we don't have many PRs and release frequently, I also think it's OK to bump version in PR directly. It's a problem for projects that don't release frequently.
Currently, we'll run
cargo semver-checks
to prevent API changes without bumping the major version (or minor version under0.x
). Essentially, this requires each PR with breaking changes to bump the major version in place.Imagine that we have 10 PRs with breaking changes merged one day and we intend to publish them together. Then we have to bump the version 10 times under current restrictions. 😕 Also, I don't think it's a good idea to bump the version in place. A separate PR (or commit) can be much clearer.
So I'm wondering whether we should only run
semver
test on release (publish) instead?The text was updated successfully, but these errors were encountered: