path module

[holger krekel]

Cliff Wells wrote:
> On Tue, 2003-07-08 at 11:01, Ian Bicking wrote:
> 
> > Interesting, but I think a bad idea.  I don't believe Python has been
> > ported to Tops-20, and I'm not sure if there's a viable VMS port
> > either.  Most filesystems don't have the complexity that the Lisp
> > pathname encapsulates.  If someone was using VMS paths, I would assume
> > they would subclass path for that OS, adding the portions that applied. 
> > I think it's unreasonable to expect people programming on normal
> > platforms to pay attention to components like version, so even including
> > it in a structured manner is asking for trouble.
> 
> There is talk that Windows will have versioning in its next filesystem
> (WinFS).  It would surprise me if there weren't similar plans on the
> Linux side.

Right.  Reiserfs plans this and Subversion has it (speaking about URLs here, not 
only local pathes).  But i think the way you specify versions will be 
vastly different so the best bet probably is to pass an additional argument to
a path-like object, e.g. for subversion 

    fn = svnpath('py.py', rev=1050)

or remotely

    fn = svnpath('http://.../py.py', rev=7362)

where 'rev' specifies a revision number (which identifies exactly
one state of a subversion-repository).  Other than that, a 'svnpath'
could probably behave like a regular local 'path' object, i guess. 
Hmmm, the above wouldn't be hard to do because svn has python bindings
on all levels ... but enough advertisement :-)

Either way, i believe that path/file versioning deserves some thoughts 
as it might be a "next big thing" (besides java and .NET, of course :-)

cheers,
    
    holger