On Sun, Mar 18, 2012 at 8:31 AM, Keshav Kini <keshav.k...@gmail.com> wrote: > William Stein <wst...@gmail.com> writes: >> On Sun, Mar 18, 2012 at 7:59 AM, Keshav Kini <keshav.k...@gmail.com> wrote: >>> William Stein <wst...@gmail.com> writes: >>>> On Sun, Mar 18, 2012 at 5:45 AM, Jeroen Demeyer <jdeme...@cage.ugent.be> >>>> wrote: >>>>> On 2012-03-17 14:39, Keshav Kini wrote: >>>>>> If you instead tell people to base >>>>>> their patches on the stable release >>>>> I certainly don't want this. >>>>> >>>>>> Why do patches need to be based on the latest dev release? >>>> >>>> What do you mean by this question? Is it rhetorical? >>>> >>>> If I based http://trac.sagemath.org/sage_trac/ticket/10281 on Sage-4.8 >>>> instead of a recent beta, it would be totally impossible to apply my >>>> patch (or merge my branch) into Jereon's current development branch, >>>> because parts of it had to be totally rewritten to take into account >>>> that somebody added "slice functionality" to vector_integer_dense >>>> after sage-4.8 was released, which significantly impacts how my code >>>> has to be implemented. >>> >>> Yes, it would be totally impossible to apply your patch into Jeroen's >>> current development branch, because it would be conflicting with >>> someone's patch X. So you should rebase your patch on that patch X, not >>> on Jeroen's dev release. >>> >>> If Jeroen decides that patch X is broken and removes it from the next >>> dev release, you are currently (according to the requirement that you >>> rebase patches on the latest dev release) expected to undo your rebasing >>> of your own patch. Why should you? The patch still conflicts with X, and >>> either you must rebase on X or X must rebase on you, and that doesn't >>> change no matter what Jeroen does with his dev releases. (Assuming that >>> patch X, or your patch, isn't going to languish as needs_work for a long >>> time.) >> >> Not that it matters, and maybe you're not really asking, but what I >> actually did 2 days ago was post two patches, one based on X and one >> not based on X. > > Great, but now you have the disadvantage that any changes you make need > to be done twice and kept in synchronization between the two patches (or > branches).
Does git magically solve that problem? > Though in practice there's probably a stage beyond which you > stop making changes - say, when the ticket is getting close to positive > review. > > -Keshav > > ---- > Join us in #sagemath on irc.freenode.net ! > > -- > To post to this group, send an email to sage-devel@googlegroups.com > To unsubscribe from this group, send an email to > sage-devel+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/sage-devel > URL: http://www.sagemath.org -- William Stein Professor of Mathematics University of Washington http://wstein.org -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
