It looks like --parent works.

Not sure what I did before, but when I reran a test based on a series of
changes to a single test file, the diffs come out as expected. I didn't try
file deletions/additions just changes within a single file:

- Created a file with three lines
- commited and pushed it (say mercurial revision 93)
- changed line2 to line2A
- commited and pushed it; (say hg revision 94)
- changed line3 to line3B
- commited and pushed it; (say hg revision 96)
- added a new line4
- commited and pushed it; (say hg revision 98)
- created a code review:
   hg postreview   93  (start at the oldest revision first)
   (creates code review id 50)
- added the other revisions
  hg postreview -e 50   --parent 93  94
  hg postreview -e 50   --parent 93  96
  hg postreview -e 50   --parent 93  98

- checked out the diffs in RB and the left hand side for all revisions is
the file at hg revision 93. The right hand side for the revisions matches up
for 94, 96 and 98 as expected.

Very nice! The command line is kind of clunky, kind of calls out for a GUI
version doesn't it?!  We're going to try it as part of our process for a
while to see if there's any hidden glitches in more realistic use cases but
definitely promising.

Thanks for your help,

On Mon, Nov 1, 2010 at 11:30 PM, J Arrizza <> wrote:

> 1. interesting. I tried "parent" but it didn't seem to work the way I
> expected. It should compare the current revision with the parent revision,
> but didn't seem to do that. I suspect I probably did not set up the test
> correctly.  In any case, I'll give it another try.

Want to help the Review Board project? Donate today at
Happy user? Let us know at
To unsubscribe from this group, send email to
For more options, visit this group at

Reply via email to