Convert USERSELECT_ALL to a runtime enabled feature.

Convert USERSELECT_ALL to a runtime enabled feature.

We'd like to turn this on again once yahoo mail fixes their site.
In the meantime, it can be runtime enabled so we keep it compiling
at least.

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=155893

[adamk, 2013-08-09 21:42:32]

https://codereview.chromium.org/22751005/diff/4001/Source/core/dom/Node.cpp
File Source/core/dom/Node.cpp (right):

https://codereview.chromium.org/22751005/diff/4001/Source/core/dom/Node.cpp#n...
Source/core/dom/Node.cpp:657: if (RuntimeEnabledFeatures::userSelectAllEnabled()
&& node->renderer()->style()->userSelect() == SELECT_ALL && treatment ==
UserSelectAllIsAlwaysNonEditable)
Rather than sprinkling these RuntimeEnabledFeatures checks around, is there some
way to keep style()->userSelect() from ever returning SELECT_ALL if the flag is
off? That seems like a more maintainable approach to runtime flags like this.

https://codereview.chromium.org/22751005/diff/4001/Source/core/editing/FrameS...
File Source/core/editing/FrameSelection.cpp (right):

https://codereview.chromium.org/22751005/diff/4001/Source/core/editing/FrameS...
Source/core/editing/FrameSelection.cpp:647: adjustPositionForUserSelectAll(pos,
directionOfEnclosingBlock() == LTR);
Indentation is off-by-one here.

[ojan, 2013-08-09 22:10:10]

https://codereview.chromium.org/22751005/diff/4001/Source/core/dom/Node.cpp
File Source/core/dom/Node.cpp (right):

https://codereview.chromium.org/22751005/diff/4001/Source/core/dom/Node.cpp#n...
Source/core/dom/Node.cpp:657: if (RuntimeEnabledFeatures::userSelectAllEnabled()
&& node->renderer()->style()->userSelect() == SELECT_ALL && treatment ==
UserSelectAllIsAlwaysNonEditable)
On 2013/08/09 21:42:32, adamk wrote:
> Rather than sprinkling these RuntimeEnabledFeatures checks around, is there
some
> way to keep style()->userSelect() from ever returning SELECT_ALL if the flag
is
> off? That seems like a more maintainable approach to runtime flags like this.

This code doesn't add a new css value, it changes the behavior of the existing
one. That said, I was able to get rid of a bunch of the sprinkled checks by
using Position::nodeIsUserSelectAll in more places.

[adamk, 2013-08-09 22:28:34]

https://codereview.chromium.org/22751005/diff/9001/Source/core/editing/FrameS...
File Source/core/editing/FrameSelection.cpp (right):

https://codereview.chromium.org/22751005/diff/9001/Source/core/editing/FrameS...
Source/core/editing/FrameSelection.cpp:562: if
(!RuntimeEnabledFeatures::userSelectAllEnabled())
You can avoid this since rootUserSelectAlLForNode() will always return
null...any reason to leave it here?

https://codereview.chromium.org/22751005/diff/9001/Source/core/page/EventHand...
File Source/core/page/EventHandler.cpp (right):

https://codereview.chromium.org/22751005/diff/9001/Source/core/page/EventHand...
Source/core/page/EventHandler.cpp:386: if
(!RuntimeEnabledFeatures::userSelectAllEnabled())
Same here as before, the null check will do the right thing...

[ojan, 2013-08-09 22:45:09]

Good points. Done. Also reordered the if-statements I had arbitrarily reordered.

[adamk, 2013-08-09 22:51:45]

lgtm

[commit-bot: I haz the power, 2013-08-09 22:51:55]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ojan@chromium.org/22751005/17001

[commit-bot: I haz the power, 2013-08-10 01:34:22]

Change committed as 155893