Wed, 01 Aug 2018 23:25:35 -0400 linelog: port to Python 3
Augie Fackler <augie@google.com> [Wed, 01 Aug 2018 23:25:35 -0400] rev 38849
linelog: port to Python 3 Differential Revision: https://phab.mercurial-scm.org/D4051
Sat, 28 Jul 2018 17:42:36 -0700 exchange: move simple narrow changegroup generation from extension
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 28 Jul 2018 17:42:36 -0700] rev 38848
exchange: move simple narrow changegroup generation from extension The narrow extension completely replaced the function generating the changegroup part when a narrow changegroup was requested. Previous commits have taught the in-core changegroup code how to filter files based on a matcher. This commit teaches the in-core bundle2 part generation code to construct a matcher based on arguments. It will also emit a bundle2 part describing the narrow spec. I believe the changegroup part generation code in the narrow extension is now limited to ellipsis serving mode. i.e. core is now capable of narrow changegroup generation when ellipsis mode is disabled. Differential Revision: https://phab.mercurial-scm.org/D4014
Sat, 28 Jul 2018 17:33:20 -0700 exchange: make narrow ACL presence imply narrow=True
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 28 Jul 2018 17:33:20 -0700] rev 38847
exchange: make narrow ACL presence imply narrow=True And refactor the logic for determining when to invoke our custom changegroup part function so it is only conditional on narrow being set. This makes it more obvious under which conditions we should invoke the custom implementation. Also use raw strings so **kwargs works on Python 3. Differential Revision: https://phab.mercurial-scm.org/D4013
Sat, 28 Jul 2018 14:52:46 -0700 changegroup: inline prune() logic from narrow
Gregory Szorc <gregory.szorc@gmail.com> [Sat, 28 Jul 2018 14:52:46 -0700] rev 38846
changegroup: inline prune() logic from narrow prune() needs to ellide manifests that aren't part of the narrow matcher. The code is violating storage abstractions, so a comment has been added. Keep in mind the impetus for moving this code to core is so changegroup code can be refactored to be storage agnostic. Differential Revision: https://phab.mercurial-scm.org/D4012
Sun, 22 Jul 2018 15:50:45 +0900 fileset: extract language processing part to new module (API)
Yuya Nishihara <yuya@tcha.org> [Sun, 22 Jul 2018 15:50:45 +0900] rev 38845
fileset: extract language processing part to new module (API) I'll add a couple more functions that work on parsed tree. % wc -l mercurial/fileset*.py 559 mercurial/fileset.py 135 mercurial/filesetlang.py 694 total
Sat, 21 Jul 2018 15:23:56 +0900 fileset: flatten 'or' nodes to unnest unionmatchers
Yuya Nishihara <yuya@tcha.org> [Sat, 21 Jul 2018 15:23:56 +0900] rev 38844
fileset: flatten 'or' nodes to unnest unionmatchers This also makes it easier to compile a union of basic patterns into a single regexp pattern.
Sat, 21 Jul 2018 15:14:38 +0900 fileset: flatten arguments list
Yuya Nishihara <yuya@tcha.org> [Sat, 21 Jul 2018 15:14:38 +0900] rev 38843
fileset: flatten arguments list Just prepares for flattening 'or' nodes. This change would have no impact on performance.
Sat, 21 Jul 2018 15:05:40 +0900 debugfileset: add option to show matcher representation
Yuya Nishihara <yuya@tcha.org> [Sat, 21 Jul 2018 15:05:40 +0900] rev 38842
debugfileset: add option to show matcher representation
(0) -30000 -10000 -3000 -1000 -300 -100 -30 -10 -8 +8 +10 +30 +100 +300 +1000 +3000 +10000 tip