Fri, 23 Jun 2017 17:19:29 +0200 configitems: register 'ui.interactive'
Pierre-Yves David <pierre-yves.david@octobus.net> [Fri, 23 Jun 2017 17:19:29 +0200] rev 33061
configitems: register 'ui.interactive' That item default value is a bit special (None) so this adds a second proof that everything is still working fine.
Sun, 25 Jun 2017 14:41:12 +0200 config: use '_config' within 'configbytes'
Pierre-Yves David <pierre-yves.david@octobus.net> [Sun, 25 Jun 2017 14:41:12 +0200] rev 33060
config: use '_config' within 'configbytes' This will prevent bugs from using None as the sentinel value (eg: 'ui.interactive')
Sun, 25 Jun 2017 14:38:56 +0200 config: use '_config' within 'configbool'
Pierre-Yves David <pierre-yves.david@octobus.net> [Sun, 25 Jun 2017 14:38:56 +0200] rev 33059
config: use '_config' within 'configbool' This will prevent bugs from using None as the sentinel value (eg: 'ui.interactive')
Sun, 25 Jun 2017 14:34:34 +0200 config: extract the core config logic into a private method
Pierre-Yves David <pierre-yves.david@octobus.net> [Sun, 25 Jun 2017 14:34:34 +0200] rev 33058
config: extract the core config logic into a private method This will make it easier for the other 'configxxx' function to detect unset value.
Fri, 23 Jun 2017 01:38:10 +0200 debugrevlog: also display the largest delta chain span
Pierre-Yves David <pierre-yves.david@octobus.net> [Fri, 23 Jun 2017 01:38:10 +0200] rev 33057
debugrevlog: also display the largest delta chain span Mercurial read all data between the base of the chain and the last delta when restoring content (including unrelated delta). To monitor this, we add data about the size of the "delta chain span" to debugrevlog.
Sat, 24 Jun 2017 21:13:48 -0700 rebase: clean up rebasestate from active transaction
Jun Wu <quark@fb.com> [Sat, 24 Jun 2017 21:13:48 -0700] rev 33056
rebase: clean up rebasestate from active transaction Previously, rebase assumes the following pattern: rebase: with transaction as tr: # top-level ... tr.__close__ writes rebasestate unlink('rebasestate') However it's possible that "rebase" was called inside a transaction: with transaction as tr1: rebase: with transaction as tr2: # not top-level ... tr2.__close__ does not write rebasestate unlink('rebasestate') tr1.__close__ writes rebasestate That leaves a rebasestate on disk incorrectly. This patch adds "removefilegenerator" to notify transaction code that the state file is no longer needed therefore fixes the issue.
Sat, 24 Jun 2017 21:01:28 -0700 test-rebase: add a test showing rebasestate left behind
Jun Wu <quark@fb.com> [Sat, 24 Jun 2017 21:01:28 -0700] rev 33055
test-rebase: add a test showing rebasestate left behind The test demonstrates that .hg/rebasestate is left behind if "rebase" was called inside an existing transaction.
(0) -30000 -10000 -3000 -1000 -300 -100 -30 -10 -7 +7 +10 +30 +100 +300 +1000 +3000 +10000 tip