Just a reminder to folks wishing to paint the shed that whatever naming
scheme we have will need to be extensable with clash-protection, and
very likely we'll want to ensure that we can reuse the unique filename
code we already have in Subversion which first tries to create a file
named PATH + SUFFIX, and failing that (because such a file already
exists) tries PATH + "." + INTEGER + SUFFIX. So, for example, if
someone had a mine-foo.doc file already in their directory, Subversion
would want to create something like mine-foo.2.doc.
I don't see how that fact would negatively affect any of these proposals
-- just wanted to make sure folks with particular aesthetic desires were
planning for both the common and less common cases.
On Thu, 2006-01-05 at 20:22 +0100, Martin Furter wrote:
> On Thu, 5 Jan 2006, [ISO-8859-1] Nicol√°s Lichtmaier wrote:
> >>>>> Is it possible to tell the Subversion client to use other namings for
> >>>>> the .mine and before/after-revision files, e.g.:
> >>>>> foo.mine.doc
> >>>>> foo.r63.doc
> >>>>> foo.r64.doc
> > Now that this is being analyzed, wouldn't it be better to use something like:
> > mine-foo.doc (or my-foo.doc)
> > r63-foo.doc
> > r64-foo.doc
> > ... so that files with double extensions still work (tar.gz).
> > Perhaps, it would be better to keep these files together in a "sort by name"
> > list:
> > svn-mine-foo.doc (or my-foo.doc)
> > svn-r63-foo.doc
> > svn-r64-foo.doc
> +1 I like this idea, it makes the conflict a bit more visible.
> To unsubscribe, e-mail: email@example.com
> For additional commands, e-mail: firstname.lastname@example.org
C. Michael Pilato <email@example.com>
CollabNet <> www.collab.net <> Distributed Development On Demand
Received on Thu Jan 5 22:45:56 2006