I'd also recommend using a,b,c,d for the components instead of 0,1,2,3. Our tool `bumper` uses numbers and i always have to try to remember that the 1st component is 0...



On Wed, May 20, 2015 at 11:02 AM, Alois Cochard <alois.cochard@gmail.com> wrote:
Hi Oleg,

I was aware of similar tools thanks to Oliver Charles who pointed me to that great python tool: http://dzil.org/, seems like semantic-release have some interesting features as well!

Thanks for the pointers, especially `hackage-diff` that I was not aware of. It will help me a lot to get an even more automated tool :-)



On 19 May 2015 at 20:07, Oleg Grenrus <oleg.grenrus@iki.fi> wrote:
I don’t know if you are aware of https://www.npmjs.com/package/semantic-release project

It’s built for npm packages, but the same idea could be used for Haskell/cabal too.

The idea is to use SCM commit message to decide which version digit to bump.
- you still have to annotate changes
- and if you try to cheat, but semantic-release runs old test suite again new version of the library to detect obvious breaking changes

Seems there is https://github.com/blitzcode/hackage-diff which can be used to make the task of detecting API changes even easier.

Hopefully these pointers let you make even greater tool!

- Oleg

On 19 May 2015, at 20:02, Alois Cochard <alois.cochard@gmail.com> wrote:

Hi all,

I am currently writing a tool to automate the release and version bump of cabal projects.

Currently you can call the tool by giving as argument the rank of the branch you want to bump, for example given 0.0.0.0:

- `cabal-release 0` bump to `1`
- `cabal-release 1` bump to `0.1`
- `cabal-release 2` bump to `0.0.1`
- `cabal-release 3` bump to `0.0.0.1`

As we are not all robots (yet), I thought that in addition to this it would be great if the user could give a human friendly name for the rank.

For now, here is what I have for the different rank:
0 == --major
1 == --major-fix
2 == --minor
3 == --minor-fix

I'm curious to know what you think of those names? any suggestions or known established convention about that?

Thanks
_______________________________________________
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe




--

_______________________________________________
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe