On Thu 03 Apr 2008 at 04:43PM, Danek Duvall wrote: > Dan handed this off to me to brush up so he could focus on the > info/list/status/contents wad y'all saw earlier. All kudos to him, all > complaints to me. > > http://cr.opensolaris.org/~dduvall/pkg-search/ > > The search database doubles in size, but it's still not all that big, in > the scheme of things.
Someone other than me should review this. But to me it looks fine. Thanks again for cleaning this up and taking it on. > It also seems to have uncovered a bug: > > $ pkg search -s http://localhost:10001 gmake > basename link usr/sfw/bin/gmake pkg:/[EMAIL > PROTECTED],5.11-0.84:20080403T075050Z > > even though I believe it should also return > > basename file usr/bin/gmake pkg:/[EMAIL > PROTECTED],5.11-0.84:20080403T075050Z > > For now, I've just put in an XXX comment. Is it also broken on the local search for gmake? File a bug perhaps? > It also turns out that import now isn't completely successful. After > wrapping the database store in a try block, I get this: > > I've no idea why these are happening. DBM just doesn't give any useful > information back, and it's not obvious (like if the data were bigger than > would fit in a DBM page, which was my first thought before I got this > printed out). > > Given the apparently limited scope of this breakage, and the fact that it > doesn't seem to be happening on any interesting files, I'm considering just > leaving it as-is for now. It's OK with me as long as we get a bug filed. -dp -- Daniel Price - Solaris Kernel Engineering - [EMAIL PROTECTED] - blogs.sun.com/dp _______________________________________________ pkg-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
