A Diumenge, 26 de juliol de 2009, Chani va escriure: > > > if something goes wrong on the pull, scripty will abort, which is bad. > > > in order to fix that I need to redesign findfiles... albert, is there a > > > reason it gathers all the .desktop files from all modules together and > > > then splits them back out again? or can I make it work on one module at > > > a time without pain? > > > > ¿Hmm? > > > > I have no idea why findfiles does that, coolo created it, but i don't see > > why you need to touch findfiles at all to make scripty not abort when a > > pull fails since there's no pulling at all in that file. > > when a pull fails, I'd like scripty to completely ignore that module for > the rest of that run. findfiles currently doesn't give me a way to do that. > I suppose I could still ignore the *results* findfiles gets for that > module, but it'd be nicer to not have it waste time on it at all.
On the other hand why don't you focus on pull not failing and ignore this? I mean why should pull ever fail? Albert > > > > also, what's the easiest way in bash to remove one module from the > > > module list? I guess it's really just one big space-separated string, > > > so some kind of string parsing could be used to remove it. or I could > > > be lazy and just build a second list of modules, consisting of only the > > > ones that updated ok. that'd probably be easier and less error-prone... > > > > > > oh, and afaik the scripty in branch does not yet support git. only > > > trunk/l10n- kde4 has my changes. the branch one needs to be updated > > > before amarok can have a second branch for scripty to check. > > > > Let's make sure trunk works first ;-) > > yeah. :) _______________________________________________ Kde-scm-interest mailing list [email protected] https://mail.kde.org/mailman/listinfo/kde-scm-interest
