On Sat, May 19, 2012 at 5:58 AM, Jens Lehmann [off-list ref] wrote:
This version of the patch does break some existing tests, but your
follow up suggests you already found that out yourself ;-)
Actually, that was just by reading the patch and noticing odd-ness. I
realise now that I snuck in that bogus edit after I had run the tests
which were otherwise clean - d'oh.
Me thinks up_path should be set in the case below, which is the only
place where it is used.
Ok.
Isn't sm_path already set correctly here? I think this line should
be dropped.
Yes, good point. Thanks.
Same here.
Thanks.
Other than that the patch looks fine.
Thanks for the review - I'll add some tests for the new behaviour.
jon.