Sean Farley <sean@farley.io> [Tue, 15 Dec 2015 16:00:06 -0800] rev 27530
crecord: ensure that curses is False if not imported
This provides no functional change but makes the next two patches easier to
review.
Sean Farley <sean@farley.io> [Tue, 15 Dec 2015 15:56:10 -0800] rev 27529
crecord: add helper function to determine if we should use curses
To fix issue5008 properly, we need a helper function to determine if curses is
imported and also if the user has enabled the experimental flag.
Sean Farley <sean@farley.io> [Wed, 16 Dec 2015 10:33:19 -0800] rev 27528
crecord: use try/except for import of curses
Not only does this improve fragility with 'if os.name == ...' it will help
future patches enable the behavior to fallback to use plain record when curses
is unavailable (e.g. python compiled without curses support).
Bryan O'Sullivan <bos@serpentine.com> [Sun, 27 Dec 2015 23:55:54 +0900] rev 27527
histedit: only use pickle if not using the modern save format
This avoids a case where PyPy's cPickle module throws a more confusing
error than CPython's.
Bryan O'Sullivan <bos@serpentine.com> [Sun, 27 Dec 2015 23:55:54 +0900] rev 27526
perf: close transaction in perffncachewrite
This fixes a bug, and brings CPython behaviour on this test into
line with PyPy.
Bryan O'Sullivan <bos@serpentine.com> [Sun, 27 Dec 2015 23:55:54 +0900] rev 27525
hbisect: use tryreadlines to load state
This closes the file handle after reading, which stops PyPy from
leaking open file handles and thus failing test-bisect3.t.
Bryan O'Sullivan <bos@serpentine.com> [Sun, 27 Dec 2015 23:55:54 +0900] rev 27524
eol: make output stable
This eliminates a divergence in behaviour between PyPy and Python.
Gregory Szorc <gregory.szorc@gmail.com> [Wed, 23 Dec 2015 12:32:08 -0800] rev 27523
exchange: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> [Wed, 23 Dec 2015 12:30:14 -0800] rev 27522
localrepo: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com> [Mon, 21 Dec 2015 21:52:58 -0800] rev 27521
httpconnection: use absolute_import