Hey, There's a webrev at http://cr.opensolaris.org/~richlowe/scm_474 that attempts to always use hg-active when finding the parent revision, as described in the bug (as also described, Ethan's problem could be fixed via less invasive means, but this way should hopefully remove the problems we may have with branches in that area).
I've tested it with the slim_install gate at the changesets Ethan mentioned, and it works in those cases in the cases) - No file list specified - plain file list without HG_PARENT specified - wx filelist without HG_PARENT specified - wx filelist with HG_PARENT specified - plain filelist with HG_PARENT specified However, I'm not particularly happy with it, so I'm soliciting opinion. If you feel like doing a real code review, that's fine, but general comment would be equally appreciated. Thanks, -- Rich
