[EMAIL PROTECTED] wrote: > Shawn, > >> webrev: >> http://cr.opensolaris.org/~swalker/pkg-1324/ > > Most of this looks good to me. I just have a question about feed.py. > > On line 103-107 you deal with setting the Atom ID. This needs to be > unique and unchanging. Does your approach satisfy this constraint?
In reading further, no. Thanks for spotting this. > > Dan blogged about this a bit here: > > http://blogs.sun.com/dp/date/20080324 > > He also references this article: > > http://diveintomark.org/archives/2004/05/28/howto-atom-id That is actually *very* helpful -- certainly more so than the material I've read so far (RFC, W3C spec, etc.). It looks like my plan of action should be to: * Ensure that a repository has a uuid on initialisation. * Use that uuid as the Atom ID * Use a <link rel="alternate" href="..."/> to specify the URI to the location of the feed. * Use a tag: uri (which is now IANA registered) for the Atom ID of entries with the pkg fmri as the value and then use a <link rel="alternate"> for the actual info link. I'll have to experiment with this a bit and run everything through the validator again. Thanks, -- Shawn Walker _______________________________________________ pkg-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
