Patch for preliminary Mercurial integration in kdevplatform

[Thomas Burdick]

Well I originally tried to do this myself, and moved on, but I have been
writting a nice set of dialogs for mercurial that may be relevant to this
discussion.

please see:

http://bitbucket.org/bfrog/cutehg-stable

I have no doubt that instead of making an extension to mercurial itself a
derivation of the dvcs plugin interface (done in python using kross, as
suggested) could call in to the nice little program I wrote to display some
fancy dialogs.

-Tom

On Sun, Mar 15, 2009 at 3:39 PM, Fabian Wiesel
<fabian.wiesel at fu-berlin.de>wrote:

> On Sun, 15 Mar 2009 00:50:30 +0100
> Andreas Pakulat <apaku at gmx.de> wrote:
>
> > On 14.03.09 10:43:25, Fabian Wiesel wrote:
> > Well, don't you have that problem right now too? i.e. what happens if
> > two of the hg-processes run at the same time on the same repository
> > doing different things?
>
> That is no problem. As any sane VCS (in other words all but CVS), the
> software is already written that two (or more processes) can access the
> same repository at the same time.
>
> > I don't know the python API (or even used
> > mercurial at all), but python itself can be made "single-threaded" by
> > locking the GIL so thats not necessarily an issue unless multiple
> > python interpreters are used.
>
> AFAIK, due to the GIL, Python is _just_ single a single thread of
> execution. There threads, but at no time there are more than two threads
> actually running in an interpreter. Still, you can have several threads
> in python, but the question stands: Is the code mercurial written to be
> reentrant? I don't think so. So, I'll expect to have to go for one
> interpreter <-> one thread.
> I will definitely give it a try, but I expect some difficulties.
>
> Fabian
>
>
> _______________________________________________
> KDevelop-devel mailing list
> KDevelop-devel at kdevelop.org
> https://barney.cs.uni-potsdam.de/mailman/listinfo/kdevelop-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kdevelop-devel/attachments/20090315/9eef2548/attachment.html>