On Fri, Nov 2, 2012 at 5:08 PM, Stefano Lattarini <stefano.lattarini@xxxxxxxxx> wrote: > On 11/02/2012 04:46 PM, Felipe Contreras wrote: >> >> In the end I liked this approach much better. >> >> If you have a solution for this that works in POSIX shell, I'll be >> glad to consider it, but for the moment, I think a simple, easy to >> understand and maintain code is more important, and if it needs bash, >> so be it. >> > If this is a deliberate decision, it's ok with me. I'm just a "casual" > reviewer here, not an active contributor, so I'll gladly accept > preferences and decisions of the "active crew", once it's clear that > they are deliberate and not the result of mistakes or confusion. > > In any case, I agree that having a clean, understandable code as a > starting point is better than having a more "portable" but trickier > one right away. If it will need converting to POSIX, that can be > done as a follow up (and as we've both noticed, this would be the > only point where such a conversion might be problematic -- the other > changes would be trivial, almost automatic). As things are the options are: 1) Remove this code and move to POSIX sh. People looking for reference might scratch their heads as to why 'git push' is not showing the update. 2) Keep this code and remain in bash. Until we have a: 3) Replace this code with a clean POSIX sh alternative I would rather vote for 2) Cheers. -- Felipe Contreras -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html