Hi Jürgen On 06/02/2015 05:06 PM, Jürgen E. Fischer wrote: > > Huh? I didn't say breaks shouldn't be fixed. I said just the opposite. But > I > said the fix should not not generally be a simple revert.
Does that more translate to: * If the test fails for an unrelated reason: fix the test or disable it * If the test fails for a good reason and it's a simple fix: fix the code instead of reverting * If none of the above applies: revert the commit All that should be done in a reasonable time. We should still be able to decide based on particular situations and not just blindly stick to these rules. But some kind of reaction would be good because in the current situation it somebody else needed to step in to fix the code and documentation. And I think that this should be the original committers responsibility. Matthias
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Qgis-developer mailing list Qgis-developer@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/qgis-developer
