On Mon, Oct 06, 2008 at 04:22:11PM -0700, Brad Hall wrote:
> http://cr.opensolaris.org/~bhall/bug-388-5/
client.py:
- line 343: "in case"
- line 345, 348: you don't do anything with "be"?
- line 341: I'd remove this line, and on line 349 simply return 1 if
img.repair() fails, and then keep your return 0 where it is.o
image.py:
- line 691: you can say "for fmri, actions in repairs:"
- line 709: You should catch KeyboardInterrupt first and re-raise it.
- line 701: I'd remove this, and return False directly from the exception
handler (and return True where you currently have your return
statement).
imageplan.py:
- line 528: maybe combine this with line 532? But where do you ever call
preexecute with an argument? I don't see it in the patch.
pkgplan.py:
- So the way I read this, all the actions in the manifest are going to be
added on line 161, since origin is empty. That doesn't seem right. Am
I missing something? Should origin and destination be the same?
file.py:
- line 131: this will be True if final_path is a symlink to a directory,
too. You could check explicitly for ENOTDIR, but you might be better
off doing an lstat() on the path and removing it if it's anything but a
regular file -- salvage it if it's a directory and trash it otherwise.
This raises the idea that perhaps fix will eventually need a "keep"
flag that preserves files somewhere when they've had their contents
replaced.
t_fix.py:
- the hardlink test should make sure that /etc/amber.hardlink and
/etc/amber1 have the same inode number, not that /etc/amber.hardlink
has a different inode than it had before the fix.
Danek
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
