Shawn, Thanks for the review. I've fixed up the newline issues. On Mon, Aug 03, 2009 at 04:12:40PM -0500, Shawn Walker wrote: > [email protected] wrote: >> http://cr.opensolaris.org/~johansen/webrev-10411/ > > lines 185-194: I don't think the ProgressTracker class should be > calling emsg(). That will cause the Null and Quiet classes to output > when there's an error (which I don't think should happen). I think this > should be in an override in FancyUNIXProgressTracker and > CommandLineProgressTracker. Note that the GUI client inherits from > NullProgressTracker as well, and the emsg() would typically never be > seen.
I'm honestly surprised that this is contraversial. The assertion lives in the base class, and the message is here just to provide ancillary debugging information if the assertion fails. The stack traces always go to stderr, so I'm confused as to why we shouldn't print this message there too. > src/modules/client/transport/engine.py: > line 644: I'm feeling a bit dumb; why does having completed as a > engine level property work given that commit() is called on a > per-handled basis? You're probably just confused by the diff format. Completed is an instance variable of a FileProgress class. There's one of these for each easy handle enqueued in the multi interface. (See lines 524-528). -j _______________________________________________ pkg-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
