Frank Kingswood <frank@kingswood-consulting.co.uk> [Fri, 26 Jul 2013 14:44:13 +0100] rev 19505
convert: handle changeset sorting errors without traceback (issue3961)
Siddharth Agarwal <sid0@fb.com> [Thu, 25 Jul 2013 14:43:15 -0700] rev 19504
ancestor.deepest: ignore ninteresting while building result (issue3984)
ninteresting indicates the number of non-zero elements in the interesting
array, not the number of elements in the final list. Since elements in
interesting can stand for more than one gca, limiting the number of results to
ninteresting is an error.
Tests for issue3984 are included.
Wei, Elson <elson.wei@gmail.com> [Thu, 25 Jul 2013 17:35:53 +0800] rev 19503
ancestor.deepest: decrement ninteresting correctly (issue3984)
The invariant this code tries to hold is that ninteresting is the number of
non-zero elements in the interesting array. interesting[nsp] is incremented at
the same time as interesting[sp] is decremented. So if interesting[nsp] was
previously 0, ninteresting shouldn't be decremented.
Siddharth Agarwal <sid0@fb.com> [Thu, 25 Jul 2013 14:20:37 -0700] rev 19502
ancestor.deepest: sort revs in C version
This isn't strictly necessary, but it makes the code more consistent with the
Python version.
Augie Fackler <raf@durin42.com> [Thu, 25 Jul 2013 10:44:51 -0400] rev 19501
check-code: add a check for the next() builtin, which was new in 2.6
Augie Fackler <raf@durin42.com> [Thu, 25 Jul 2013 10:42:36 -0400] rev 19500
mq: rename next() to nextpatch() to avoid confusing a future check-code patch
next() is a builtin starting in 2.6, so it's also nice to avoid
shadowing the builtin.
Alexander Plavin <me@aplavin.ru> [Thu, 25 Jul 2013 15:27:41 +0400] rev 19499
hgweb: replace next(revs) to revs.next() to fix compatibility with Python 2.5-
Matt Mackall <mpm@selenic.com> [Thu, 25 Jul 2013 02:44:27 -0500] rev 19498
tests: glob out exception type for directory collision
Got three different results on buildbot.
Matt Mackall <mpm@selenic.com> [Thu, 25 Jul 2013 02:34:09 -0500] rev 19497
record: add checkunfinished support (issue3955)
Matt Mackall <mpm@selenic.com> [Thu, 25 Jul 2013 02:17:52 -0500] rev 19496
checkunfinished: accommodate histedit quirk
Turns out histedit actually intends for commits (but not other
operations like update) to be possible during its operation.