On Tue, Oct 21, 2008 at 02:35:53PM -0700, Rich Burridge wrote: > Danek Duvall wrote: >> The code looks fine, but I'm wondering about two things: >> >> - Is the double colon really what you meant, or a typo (on Stephen's >> part >> -- I saw the email he sent with it)? >> > > It could be. Stephen's original email is here: > > http://mail.opensolaris.org/pipermail/pkg-discuss/2008-August/005556.html > > and that only uses a single colon. >> - Why is the importer keyword "info.classification" rather than just >> "classification"? >> > My guess is there are going to be more info.<wherever> things.
Remember that the importer keyword doesn't have to end up as the name of the package attribute. You could have the keyword be "classification" and publish a set action with "name=info.classification". As it stands, there's little point in modifying the parser, since you'd get exactly the same effect by prefixing each line in the importer files with "add set " (you wouldn't get the print out during the import, but I don't think it's that important). The real benefit of having the keyword is that if you need to change the way the classifications are done (the name of the attribute, or the prefix of the value), you do it once, in solaris.py. Of course, if we're going to have multiple classification schemes, this is perhaps a bit less useful. And so, we wait for Stephen to find out what he had in mind going forward here. Danek _______________________________________________ pkg-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
