Sat, 25 Feb 2017 21:44:34 -0500 wix: include the help for pager
Matt Harbison <matt_harbison@yahoo.com> [Sat, 25 Feb 2017 21:44:34 -0500] rev 31115
wix: include the help for pager Similar (I assume) to 65d2538ac993.
Sun, 26 Feb 2017 10:16:47 -0800 manifest: check 'if x is None' instead of 'if not x'
Durham Goode <durham@fb.com> [Sun, 26 Feb 2017 10:16:47 -0800] rev 31114
manifest: check 'if x is None' instead of 'if not x' The old code here would end up executing __len__ on a tree manifest to determine if 'not _data' was true or not. This was very expensive on large repos. Since this function just cares about memoization, we can just check 'if _data is None' instead and save a bunch of time.
Sat, 25 Feb 2017 03:42:43 +0530 pager: add support to --patch, --list and --stat options of hg shelve
Pulkit Goyal <7895pulkit@gmail.com> [Sat, 25 Feb 2017 03:42:43 +0530] rev 31113
pager: add support to --patch, --list and --stat options of hg shelve
Mon, 20 Feb 2017 12:42:35 +0100 color: drop the 'colorui' class
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Mon, 20 Feb 2017 12:42:35 +0100] rev 31112
color: drop the 'colorui' class Now that all logics formally bared by 'colorui' have been moved to the main ui class, that class is empty and can be dropped. As a nice side effect we can get rid of the baroque Initialization associated to it. There was much rejoicing.
Mon, 20 Feb 2017 12:42:23 +0100 color: move 'write-err' logic to the core ui class
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Mon, 20 Feb 2017 12:42:23 +0100] rev 31111
color: move 'write-err' logic to the core ui class This is similar to what we needed for 'write', we move the logic from the extension to the core class. Beside the dispatch to 'win32print', we just apply label to the argument.
Fri, 24 Feb 2017 19:53:41 +0100 ui: extract low level part of 'write_err' in its own method
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Fri, 24 Feb 2017 19:53:41 +0100] rev 31110
ui: extract low level part of 'write_err' in its own method Same as for 'write' and '_write' we needs it available for some of the color logic.
Fri, 24 Feb 2017 19:28:45 +0100 ui: extract buffer write from protect and timed 'write_err' output
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Fri, 24 Feb 2017 19:28:45 +0100] rev 31109
ui: extract buffer write from protect and timed 'write_err' output That subcall to 'self.write' is never doing actual write but only store things in buffers. So we do not need to protect it for exception not to time its execution. This will make it easier to extract a '_write_err' function as we did for 'write'.
Mon, 20 Feb 2017 12:31:39 +0100 color: move 'write' logic to the core ui class
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Mon, 20 Feb 2017 12:31:39 +0100] rev 31108
color: move 'write' logic to the core ui class One more step, the support for writing color is not directly in core. No behavior change for the default case ('_colormode' = None). Here are the details of what we have to change to the core method: * apply to 'self.label' to input in the buffered case * dispatch to 'win32print' when applicable * apply to 'self.label' to input when applicable
Fri, 24 Feb 2017 19:27:00 +0100 ui: extract the low level part of 'write' in a dedicated function
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Fri, 24 Feb 2017 19:27:00 +0100] rev 31107
ui: extract the low level part of 'write' in a dedicated function We are about to add some extra logic related to color. That logic will need to access the low level layer of ui doing the actual write to a stream. (eg: 'win32print'). We extract this logic into a private method for this purpose.
Fri, 24 Feb 2017 21:34:07 +0100 color: add multiple messages input support to 'win32print'
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Fri, 24 Feb 2017 21:34:07 +0100] rev 31106
color: add multiple messages input support to 'win32print' All other function doing writes support any number of input message. For simplicity, we make 'win32print' able to do the same.
Fri, 24 Feb 2017 21:31:47 +0100 color: clarify name of an argument of 'win32print'
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Fri, 24 Feb 2017 21:31:47 +0100] rev 31105
color: clarify name of an argument of 'win32print' In the current code, the function called to write happens to the 'orig' version of the method calling 'win32print' (obtained with a 'super' call). However, the variable could have a better name. That will be useful when we'll stop having inheritance in play.
Thu, 23 Feb 2017 19:45:32 +0100 color: move the 'colorlabel' call to the core 'ui' class
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Thu, 23 Feb 2017 19:45:32 +0100] rev 31104
color: move the 'colorlabel' call to the core 'ui' class This bring us closer to supporting color in core natively. Core already have a 'label' method that was a no-op. We update its to call the new 'colorlabel' function. Behavior is unchanged when colormode = None.
Thu, 23 Feb 2017 19:10:24 +0100 color: move the 'colorlabel' function in the core module
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Thu, 23 Feb 2017 19:10:24 +0100] rev 31103
color: move the 'colorlabel' function in the core module The extract code is relocated in core.
Thu, 23 Feb 2017 19:00:26 +0100 color: extract the label code into its own function
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Thu, 23 Feb 2017 19:00:26 +0100] rev 31102
color: extract the label code into its own function We extract the logic into a function. This will allow us to move the logic into the core 'color' module and later call it directly from core.
Mon, 20 Feb 2017 12:13:23 +0100 color: inline the 'showlabel' method
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Mon, 20 Feb 2017 12:13:23 +0100] rev 31101
color: inline the 'showlabel' method This function is quite simple and only have one call siteā€¦ located a handful of line under the definition. We inline the function for the sake of simplicity. One of the motivation to do that now is that it reduce the amount of new method we will have to add to the core 'ui' class for color support.
Mon, 20 Feb 2017 12:12:08 +0100 color: move '_colormode' to the core 'ui' class
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Mon, 20 Feb 2017 12:12:08 +0100] rev 31100
color: move '_colormode' to the core 'ui' class Having all 'ui' objects aware of 'color' allows us to update the core code to handle color. The mode will stay 'None' in the default case so that will not introduce any changes.
Tue, 17 Jan 2017 17:25:48 +0100 hgweb: explictly pass basectx in webutil.diffs
Denis Laxalde <denis.laxalde@logilab.fr> [Tue, 17 Jan 2017 17:25:48 +0100] rev 31099
hgweb: explictly pass basectx in webutil.diffs There's only one case where `basectx` parameter is None (over two usages), so it's probably not worth handling the special case as it makes code-reading harder. Along the way, use ctx.p1() instead of checking for ctx.parents() being empty which should not occur.
Wed, 22 Feb 2017 18:26:12 -0800 serve: mark --stdio and --cmdserver as "(ADVANCED)" flags
Jun Wu <quark@fb.com> [Wed, 22 Feb 2017 18:26:12 -0800] rev 31098
serve: mark --stdio and --cmdserver as "(ADVANCED)" flags This will hide them from the default --help output.
Tue, 01 Nov 2016 14:50:45 +0000 help: hide command line options marked as "advanced"
Jun Wu <quark@fb.com> [Tue, 01 Nov 2016 14:50:45 +0000] rev 31097
help: hide command line options marked as "advanced" Previously, we have keywords like "(DEPRECATED)" and "(EXPERIMENTAL)" to hide command line options in non-verbose help output. However, sometimes an option is neither deprecated nor experimental. It's well-tested and working, but just not designed to average users. This patch adds a keyword "(ADVANCED)" to fit in such use cases. Thanks rom1dep for the suggestion of the word "advanced".
Thu, 23 Feb 2017 21:27:25 +0900 pager: do not try to run an empty pager command
Yuya Nishihara <yuya@tcha.org> [Thu, 23 Feb 2017 21:27:25 +0900] rev 31096
pager: do not try to run an empty pager command If pagercmd is explicitly set to '', the pager process would exit silently and the output would be lost. We'd better disable the pager in such case.
Thu, 23 Feb 2017 21:20:26 +0900 graphlog: restore pager lost at 1cec1d863008
Yuya Nishihara <yuya@tcha.org> [Thu, 23 Feb 2017 21:20:26 +0900] rev 31095
graphlog: restore pager lost at 1cec1d863008
Wed, 22 Feb 2017 12:09:17 +0100 gitweb: add "more" and "less" navigation links in filelog template
Denis Laxalde <denis.laxalde@logilab.fr> [Wed, 22 Feb 2017 12:09:17 +0100] rev 31094
gitweb: add "more" and "less" navigation links in filelog template
Mon, 16 Jan 2017 17:14:36 +0100 context: also return ancestor's line range in blockancestors
Denis Laxalde <denis.laxalde@logilab.fr> [Mon, 16 Jan 2017 17:14:36 +0100] rev 31093
context: also return ancestor's line range in blockancestors
Mon, 16 Jan 2017 17:08:25 +0100 context: add a followfirst flag to blockancestors
Denis Laxalde <denis.laxalde@logilab.fr> [Mon, 16 Jan 2017 17:08:25 +0100] rev 31092
context: add a followfirst flag to blockancestors
Mon, 20 Feb 2017 18:40:42 +0530 py3: use pycompat.fsencode() to convert __file__ to bytes
Pulkit Goyal <7895pulkit@gmail.com> [Mon, 20 Feb 2017 18:40:42 +0530] rev 31091
py3: use pycompat.fsencode() to convert __file__ to bytes __file__ returns unicodes on Python 3. This patch uses pycompat.fsencode() to convert them to bytes.
Wed, 22 Feb 2017 10:14:18 -0800 wireproto: remove unused code
Jun Wu <quark@fb.com> [Wed, 22 Feb 2017 10:14:18 -0800] rev 31090
wireproto: remove unused code Removed an unused line introduced by f3807a135e43.
Thu, 22 Dec 2016 06:18:45 +0100 color: set initial default value for 'colormode' to None
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Thu, 22 Dec 2016 06:18:45 +0100] rev 31089
color: set initial default value for 'colormode' to None This should not introduce any behavior changes when using the color extension. In practive, the colormode will be setup at early at run time to the proper value (from config and environment). We do this change as this gets us closer of a state were we can have all the mechanisms associated to color in core with the feature disabled by default.
Thu, 22 Dec 2016 13:19:12 +0100 color: minor reversal of two conditional clause for clarity
Pierre-Yves David <pierre-yves.david@ens-lyon.org> [Thu, 22 Dec 2016 13:19:12 +0100] rev 31088
color: minor reversal of two conditional clause for clarity Another minor cleanup while reading the code. The two branches of the conditional have similar complexity so we go for the order that give us the simplest condition (we drop the negation).
(0) -30000 -10000 -3000 -1000 -300 -100 -50 -28 +28 +50 +100 +300 +1000 +3000 +10000 tip