Brock Pytlik wrote:
Danek Duvall wrote:
Brock Pytlik wrote:
I'd argue it is useful as it provides a way to restrict search to
particular publishers (as opposed to depots). I could replace it with
fmri_obj.get_pkg_stem()
and have pkg:// required as part of the search term if you'd prefer.
Yes, I'd rather see the explicit scheme be required. At some point, I
guess I wouldn't mind seeing rooted package names, where a leading
single
slash indicates that it's the full package name without a publisher,
and a
leading double slash indicates that it's the full package name with a
publisher, but that likely will take a bit more plumbing than we can
do for
this release.
Danek
Ok, I'll make that change as well in this wad (removing the
scheme=False from the function call, not the larger bit).
Brock
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
Actually, I can't make that change. The problem is that the ':' in
pkg://, clashes with the ':' used as a field separator for search. If
it's critical, I can look into putting a special case into the
tokenizing/parsing logic so that pkg:// is always treated as a unit, but
my inclination is to just leave this alone for now and file a bug to
address it in the future.
Brock
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
