Danek Duvall wrote: > On Wed, Nov 12, 2008 at 09:02:43PM -0800, Bart Smaalders wrote: > >>> - line 278: I'm not sure I see the point of these two tests. What >>> happens if you have two chattrs, the first one changes an attribute >>> of the original to something different, then the second one comes in >>> and changes it to the original. With this test, the second chattr >>> would be ignored, but I can't see why you'd want to ignore it more >>> than you'd want to ignore any other chattr. >>> >> The purpose of these two tests are accumulate only changes from each >> chattr* in f.changed_attributes... since each chattr is (confusingly) >> applied to the original file specification, unless we note which portions >> are changed by each chattr, subsequent changes by other >> chattrs will undo our previous modifications. > > I thought we already were potentially overwriting previous chattrs -- > that's what the test on line 280 indicates. Can you give an example? >
Ah... the changed_attr list before got the entire modified action; each chattr was applied to the original action.. So each change would completely replace the previous change, which was fine if they affected the same attribute but not if they affected different ones, since the previous change would be overwritten. -- Bart Smaalders Solaris Kernel Performance [EMAIL PROTECTED] http://blogs.sun.com/barts "You will contribute more with mercurial than with thunderbird." _______________________________________________ pkg-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
