Git Support ready-ness

[Andreas Pakulat]

On 03.09.10 04:07:54, Aleix Pol wrote:
> On Fri, Sep 3, 2010 at 12:32 AM, Andreas Pakulat <apaku at gmx.de> wrote:
> 
> > On 02.09.10 22:46:15, David Nolden wrote:
> > > I think we don't have the ressources to maintain a git plugin that
> > > works with many versions, so we should simply go the version-check way
> > > (it could be checked when kdevelop decides whether a directory is
> > > controlled by git, it could always return false and do a warning on
> > > the command-line if the version is too old). Personally I can solve
> > > this for myself by doing a _local_ installation of a current git
> > > version.
> >
> > As I just discussed with Aleix on irc, thats not going to be a
> > good-enough solution as at least --porcelain for git status was only
> > implemented in 1.7 (which was first released end of February 2010). IMHO
> > we do need to support git versions that are 1.5-2 years old which means
> > everything since 1.6.0.1 (roughly) and hence we have to be able to parse
> > --porcelain as well as --short output for git status.
> >
> > Other git-parameters things will also need to be checked for being
> > available in 1.6.0.
> 
> I worked on that for a while, it quite a loss of time though. Before 1.7 git
> status was something like git commit without committing, so it had just that
> information.

But the code that was there 9 months ago must have worked somehow, after
all 1.7 was released at the end of February... 

> I think that for us it's quite important to use the machine-readable
> commands, so I'd vote for forcing git 1.7.

IMHO thats not an option, its simply too new.

Andreas

-- 
Are you making all this up as you go along?