On Tue, 2011-02-08 at 16:01 -0800, Danek Duvall wrote:
> This will help the userland folks do the right thing:
>
> http://cr.opensolaris.org/~dduvall/pkg-pkglint-desc/
>
> Most of the bulk here is making sure the new check doesn't bust a ton of
> the existing tests.
That looks good - and will hopefully land before the pkglint ON wad, so
I'll make sure the ON manifests are doing the right thing by this check
- thanks for fixing that bug.
One question, you're adding:
146 def bogus_description(self, manifest, engine, pkglint_id="002"):
I'm not sure how strict we want to be about pkglint_id values, but
"opensolaris.manifest002" used to be assigned to the print_fmri check,
since removed by the fix for 17072.
Should we be reusing pkglint_id values or assigning new ones? Perhaps
17072 should have left a comment in the code saying that ID was no
longer in use.
cheers,
tim
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
