On Mon, Jun 02, 2008 at 01:46:51PM -0500, Shawn Walker wrote: > > Okay (I skipped most of that subthread, since I was away). It's not clear > > to me that this is useful -- on a system that hasn't been hacked to bits, > > it'll be there (otherwise SUNWipkg won't even install). I'll want this to > > go away, though, certainly once cherrypy gets properly integrated. > > I have no objections to removing the check completely if so desired. > > Just let me know.
No, leave it in. > In the case of transaction.py, the open() function is returning a > particular status to the calling function. > > The calling function is explicitly evaluating the return value and > then performing exceptions or various other bits of logic based on > that return value. > > As such, it wasn't appropriate for me to raise an exception there. Okay, sorry I didn't pick up on that. > So, in short, it's just doing what it used to. My personal preference > is to let the caller handle abandon() and return the status since that > would make it match open()'s behaviour. > > Would that be your desired approach? I don't really care too much; if you think that'd be cleaner ... > >> > SUNWpython-cherrypy/Makefile: > >> > > >> > - I wonder if perhaps doing the tarball download higher up, in the main > >> > "make install" wouldn't be better -- so that it'll work out of the > >> > box, > >> > and you could integrate it into "make link" and so on. > >> > >> Eventually cherrypy will go away and not be part of the build, so I'm > >> not certain. > > > > Yeah, but we don't have a timeframe on that yet, do we? Seems like we > > could be in the current situation for a while. > > True. So what are your expectations then in regards to this wad? Don't worry about it for now. If it bugs me enough, I'll do something about it. Thanks, Danek _______________________________________________ pkg-discuss mailing list pkg-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
