[Webkit-unassigned] [Bug 46761] Remove calls to Document::updateStyleForAllDocuments()

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=46761





--- Comment #16 from James Robinson <jamesr at chromium.org>  2012-01-19 18:51:34 PST ---
(In reply to comment #13)
> Have we tested to see what Firefox's behavior here is?  Do they create an empty renderer?
> 
> I'm not sure this "regression" matters much in practice.  If this is the worse consequence of removing updateStyleForAllDocuments() I would just remove it. :)   However I also support further investigation.

The actual DOM is the same here with or without this patch. For example on LayoutTests/fast/canvas/resize-while-save-active.html the last element of document.body.childNodes is a #text node containing three linebreaks. This is the same as Firefox.

I don't think this renderobject has any impact on the actual rendering of the page or on any APIs exposed to authors. It might have a small perf impact, but certainly the impact is a whole lot smaller than synchronously updating styles on all documents all the time! Here's a list of all tests that show this issue on a ToT chromium build:

editing/selection/caret-ltr-2-left.html
editing/selection/caret-ltr-2.html
editing/selection/caret-ltr-right.html
editing/selection/caret-ltr.html
editing/selection/caret-rtl-2-left.html
editing/selection/caret-rtl-2.html
editing/selection/caret-rtl-right.html
editing/selection/caret-rtl.html
fast/canvas/resize-while-save-active.html
fast/dom/title-directionality.html
fast/xsl/import-non-document-node.xhtml
http/tests/cache/subresource-expiration-1.html
http/tests/cache/subresource-expiration-2.html
http/tests/xmlhttprequest/send-array-buffer.html
http/tests/xmlhttprequest/send-undefined-and-null.html

I think we should move forward with this, rebaseline those 15 tests, and then if anyone feels motivated to optimize this so there's no 0x0 RenderText node they can feel free.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.