Peter N. Lundblad wrote:
> Ben Collins-Sussman writes:
> > On 3/8/06, Alan Barrett <email@example.com> wrote:
> > That was my intuition too, yes.
> > It just gets complex when you have a working copy switched to reflect
> > a bunch of disjoint areas of the repository. I'm sure it's just a
> > SMOP. :-)
> If the base URL for all commit items isn't easily available, we could
> just print the URL of the target and then relative paths for commit
> items under the target. Other paths could be printed using their
> complete URL. This would give us a common case where the base URL
> isn't repeated for each path and it would also make switched paths
> stand out for the user.
The entire commit process is driven by commit URLs, so they *better* all
Further, we already display URLs for remote commits (svn mkdir URL, svn
delete URL, svn copy URL URL, svn move URL URL, ...), so I think it
would only add consistency to switch to URLs for regular commits.
The relative path in a log message is useful only to the user whose
working copy is doing the commit. So in the complex switched-up working
copy case, and presuming that no sane log message best-practices or
policy would encourage expressing commit paths in any form other than as
paths relative to the root of the repository, those paths would have to
touched up by hand anyway.
+1 on switching to URLs of committables instead of local paths.
-1 (no veto, yet) on using *both* URLs and local paths.
C. Michael Pilato <firstname.lastname@example.org>
CollabNet <> www.collab.net <> Distributed Development On Demand
Received on Thu Mar 9 18:08:25 2006