abhinandan.ekande@oracle.com [Fri, 19 Oct 2012 10:27:26 +0530] rev 2817
7139743 license actions without a path can cause a stacktrace
Dan Price <daniel.price@oracle.com> [Wed, 17 Oct 2012 13:37:11 -0700] rev 2816
7192183 nasty depot testing should be part of core testing
Dan Price <daniel.price@oracle.com> [Wed, 17 Oct 2012 13:37:07 -0700] rev 2815
7195083 p5i.parse should discriminate "" versus None
Dan Price <daniel.price@oracle.com> [Wed, 17 Oct 2012 13:37:04 -0700] rev 2814
7193416 client can traceback if manifest response is empty
Dan Price <daniel.price@oracle.com> [Wed, 17 Oct 2012 13:37:02 -0700] rev 2813
7193408 variable name incorrect in error path when getting a p5i
Dan Price <daniel.price@oracle.com> [Wed, 17 Oct 2012 13:37:00 -0700] rev 2812
7192182 pkgrecv traceback trying to setup request headers on a refetch
Dan Price <daniel.price@oracle.com> [Wed, 17 Oct 2012 13:36:58 -0700] rev 2811
7192447 transport.py has bad exception handling in do_search
Dan Price <daniel.price@oracle.com> [Wed, 17 Oct 2012 13:36:57 -0700] rev 2810
7195051 catalog.attrs code needs improved armor against bogus responses
Dan Price <daniel.price@oracle.com> [Wed, 17 Oct 2012 13:36:54 -0700] rev 2809
7201323 catalog.extract_matching_fmris() could go away
Dan Price <daniel.price@oracle.com> [Wed, 17 Oct 2012 13:36:50 -0700] rev 2808
7195369 corrupt manifests can end up on disk when -g is used