On Wed, Jun 24, 2009 at 01:46:29PM -0700, Robert Miller wrote:
>
> On Jun 24, 10:34 pm, Robert Miller <rlmills...@gmail.com> wrote:
> > > I really like having the ticket number first, it makes it easy to see
> > > (given an ordered list of patches) what patches belong as part of a
> > > single ticket. E.g.
> >
> > > 6201-heegner.patch
> > > 6201-referee-fixes.patch
> > > ...
> >
> > +1
If just one could tolerate also a description field before the number,
imposing it to be identical for all patches in the ticket, that would
do as well.
Oh, and if you really want a prefix, sage_... would be more specific
than trac_ (many projects use trac)
> Something occurs to me, which is that the sage-combinat group spends a
> lot of time working on patches outside of trac. This is probably why
> Nicolas prefers to have them organized by concept. But for me, I'm
> always on trac-- everything I work on, even experimental, has a trac
> ticket. I generally tend to have several tickets open in Firefox so I
> know which issues I'm tracking. Then I just tab over to the relevant
> ticket, and that's why trac_### is so useful. Since we are discussing
> naming conventions for patches which are on trac, maybe we should keep
> that in mind.
>
> Maybe the combinat group can discuss a related scheme, and have
> conversion tools which make it effortless to swap between the two
> schemes (I'm not volunteering here, just so nobody gives the usual
> response :-) ).
Oh well, I guess I'll just end up doing this.
Btw: would it be easy to extract from the current automated release
tools a python function that given a ticket number would return the
url's of the corresponding patches on trac?
> I'd also like to point out that one of the biggest reasons projects
> fail is due to trying to change their tools midstream, which is one
> argument for keeping the convention the way it is. However, that is
> precisely what the combinat group would have to do to adapt to this
> scheme. This is why I recommend automated conversion tools.
> Another point to make is a video I saw recently by the SVN guys, who
> claimed that the amount of discussion which centers around a decision
> is inversely proportional to how important it is. They related a story
> where the project nearly forked and many feelings were hurt over
> something ten times more trivial than this topic. Just for perspective.
Definitely. Speak of pickling for the category stuff ...
But yeah I am being grumpy, and mainly needed to express my
disagreement a last time, in preparation for the future complaints of
our developers about the yet another little grain of complexity of
having different patch names on trac and on sage-combinat server.
Best,
Nicolas
--
Nicolas M. Thiéry "Isil" <nthi...@users.sf.net>
http://Nicolas.Thiery.name/
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
