On Fri, Mar 20, 2015 at 2:02 PM, Richard Hipp <d...@sqlite.org> wrote: > I agree with Stephan, except to note that some repositories do not > store code. If you are checking in changes to text documentation, > then maybe testing is not as important and a partial commit would be > ok. >
I believe it can be evil even in this scenario. Isolation, in my opinion, must be achieved with branches. And "complete" commits, as opposed to partial ones, should be done with merges. I believe I'm totally misunderstanding, but used to do very small commits, it does not make much sense to have "partial" commits. Or at least, I cannot see how my workflow would be better using them. Moreover, I believe that partial commits, in the form I understand from this thread, will call for a continuos amend/rebase (are you sure you will never miss a line in a partial commit?). Luca _______________________________________________ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
