On 6/13/07, Clint Adams <[EMAIL PROTECTED]> wrote:
On Wed, Jun 13, 2007 at 05:15:23PM +0100, martin f krafft wrote: > git reset HEAD Co<tab> > > does not complete. It should probably get its data from the git > status output: > > # On branch master > # Changes to be committed: > # (use "git reset HEAD <file>..." to unstage) > # > # new file: src/ConfigBroker.py > # > > Alternatively, normal file completion would work, might even be better.Probably __git_revisions should complete HEAD and HEAD^, though I don't know what that is. Index: Completion/Unix/Command/_git =================================================================== RCS file: /cvsroot/zsh/zsh/Completion/Unix/Command/_git,v retrieving revision 1.2 diff -u -r1.2 _git --- Completion/Unix/Command/_git 25 Nov 2005 12:52:23 -0000 1.2 +++ Completion/Unix/Command/_git 13 Jun 2007 16:35:30 -0000 @@ -865,7 +865,8 @@ '( --soft --hard)--mixed[like --soft but report what has not been updated (default)]' \ '(--mixed --hard)--soft[do not touch the index file nor the working tree]' \ '(--mixed --soft )--hard[match the working tree and index to the given tree]' \ - ':commit-ish:__git_revisions' && ret=0 + ':commit-ish:__git_revisions' \ + '*:path:_files' && ret=0 } _git-resolve () {
I have a vastly updated version of _git in the making. Sadly it's taken more than two months to put together, and it's still not done. I wrote the initial version, which matches version 1.3 quite well. The new one syncs _git to version 1.5. I'm about 90% there, but until I'm done I'm really not that interested in sending out patches - especially considering how much has changed, both in the way of completion and in way of git as such. More information here: http://git.bitwi.se/?p=dot-home.git;a=summary It's been somewhat tested, but there are probably loads of bugs to weed out. nikolai -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
