Bower doesn't support `^` in semvers #148

Closed
wheresrhys opened this Issue Mar 10, 2014 · 2 comments

Comments

Projects
None yet
2 participants
@wheresrhys
Member

wheresrhys commented Mar 10, 2014

Just tried it out on a few modules. It treats ^ the same as ~, although I think this is a bug rather than a feature so will hopefully get fixed. In the meantime ^ should definitely be discouraged.

@wheresrhys wheresrhys reopened this Mar 20, 2014

@wheresrhys

This comment has been minimized.

Show comment
Hide comment
@wheresrhys

wheresrhys Mar 20, 2014

Member

Got to the bottom of this. When the major version is 0, ^ more or less treats the minor version as if it were a major version https://github.com/isaacs/node-semver#ranges

^0.1.3 := >=0.1.3-0 <0.2.0-0 "Compatible with 0.1.3". 0.x.x versions are special: the first non-zero component indicates potentially breaking changes, meaning the caret operator matches any version with the same first non-zero component starting at the specified version.

@dansearle-ft and I are wondering if we should just go ahead and bump every module that's pretty much ready (all the 'active' ones?) to v1 at least.

Member

wheresrhys commented Mar 20, 2014

Got to the bottom of this. When the major version is 0, ^ more or less treats the minor version as if it were a major version https://github.com/isaacs/node-semver#ranges

^0.1.3 := >=0.1.3-0 <0.2.0-0 "Compatible with 0.1.3". 0.x.x versions are special: the first non-zero component indicates potentially breaking changes, meaning the caret operator matches any version with the same first non-zero component starting at the specified version.

@dansearle-ft and I are wondering if we should just go ahead and bump every module that's pretty much ready (all the 'active' ones?) to v1 at least.

@triblondon

This comment has been minimized.

Show comment
Hide comment
@triblondon

triblondon Mar 20, 2014

Contributor

Agreed. Could you do me a favour and revert the changes to the spec then,
so we go back to recommending the ^ approach?

Contributor

triblondon commented Mar 20, 2014

Agreed. Could you do me a favour and revert the changes to the spec then,
so we go back to recommending the ^ approach?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment