On 1 August 2011 09:40, Edgar Fuß <ef@math.uni-bonn.de> wrote:
Also, it's best sending patches using the git tools. I tried to understand the problem, but I didn't. I asked a GIT aware colleague to understand the problem, but he didn't either. I asked here to convert what I sent into what I should have sent (so I could understand the difference), but to no avail.
So, after finishing a proposed solution to Bug 6390, I went into the trouble of pulling the Koha GIT repository, applied my patch and looked what ``git diff'' would output. I can't spot any substantial difference from the original patch. Nevertheless, I re-sent my patches for both 5260 and 6390 in GIT format to hopefully make people happy.
That way, anyone can really easily apply them just from the bug, I just simply don't understand. Is it really that people are able to issue a git apply mailorder.git but are unable to issue a patch -p0 <mailorder.patch ? I can't believe that.
and they keep history and attribution correctly I don't see how this can be done more easily by git apply then by patch.
If you use git format-patch and then git send-email to send it to the koha-patches mailing list. And also attach that patch to the bug (that's not as nessecary the mailing list is the main place). The patches can then be applied by git am -iu3, keeping author and committer history intact. This is the way the other 147 developers do it, and it seems to work well for them. Chris