Bug 121559 – Make "autoscroll for drag'n" drop a setting

• Component: WebCore Misc.
• Status: RESOLVED FIXED

[Antonio Gomes]

That way it can be testable/toggled via Internals::Settings from Layout tests, and tested independently from a specific port decision about enabling it or not.

[Antonio Gomes]

Created attachment 212001 [details]
patch

[Antonio Gomes]

Created attachment 212002 [details]
patch

[WebKit Commit Bot]

Attachment 212002 [details] did not pass style-queue:

Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'LayoutTests/ChangeLog', u'LayoutTests/fast/events/drag-and-drop-autoscroll.html', u'Source/WebCore/ChangeLog', u'Source/WebCore/page/AutoscrollController.cpp', u'Source/WebCore/page/ChromeClient.h', u'Source/WebCore/page/Settings.in', u'Source/WebCore/testing/InternalSettings.cpp', u'Source/WebCore/testing/InternalSettings.h', u'Source/WebCore/testing/InternalSettings.idl']" exit_code: 1
Source/WebCore/page/AutoscrollController.cpp:38:  Alphabetical sorting problem.  [build/include_order] [4]
Total errors found: 1 in 8 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Antonio Gomes]

Created attachment 212013 [details]
patch

[Antonio Gomes]

(In reply to comment #4)
> Created an attachment (id=212013) [details]
> patch

QtWebKit and WebKitGtk+ are interested in this feature (see https://bugs.webkit.org/show_bug.cgi?id=40981#c12 ), but Apple's Mac port has the best DRT/WTR drag'n drop support.

This patch makes the feature more easily testable by ports without necessarily enabling it for their products.

Please review.

[Allan Sandfeld Jensen]

Anyone have any idea what kind of heuristics Chromium could be using to guess the settings on RenderBox level?

[Antonio Gomes]

(In reply to comment #6)
> Anyone have any idea what kind of heuristics Chromium could be using to guess the settings on RenderBox level?

Yes. Chromium used no heuristic. Basically chromium was returning 'true' unconditionally.

See the commit that removed this ChromeClient hook in Blink:


____
commit 0599e64dc309c94902f9d5b5de245dc6f2016ea1
Author: [e-mail] <[e-mail]@bbb929c8-8fbe-4397-9dbb-9b2b20218538>
Date:   Tue Apr 30 19:00:03 2013 +0000

    Remove ChromeClient cruft
    
    Many ChromeClientImpl overrides simply return true, false, or do
    nothing. This patch plumbs that information back up to WebCore and
    thus simplifies a bunch of code.

Review URL: https://codereview.chromium.org/13851023

[Antonio Gomes]

Fixed by http://trac.webkit.org/changeset/156158