[Webkit-unassigned] [Bug 57731] [CSS 3] missing cursor support for 'none' on Windows

Wed May 25 18:10:44 PDT 2011

https://bugs.webkit.org/show_bug.cgi?id=57731

--- Comment #11 from Ryan <aikavanak at gmail.com>  2011-05-25 18:10:44 PST ---
(In reply to comment #9)
> (From update of attachment 88773 [details])
> Have you tested that this change does not reintroduce bug 45692 or bug 52024?

Yes, I cannot reproduce those bugs with the patch.

> I'm surprised that you were able to fix this for WebKit2 without touching WebView::onSetCursor.

Ah, I may still have to change it - I noticed some flickering of the cursor while it is set to 'none' with my patch. Not sure if it is related, I will investigate further.

> Is there a way we can assert that if we're calling ::SetCursor(0) it's because Cursor::None is being used? That would help us catch mistakes where we're calling ::SetCursor(0) for some incorrect reason.

This should be straightforward for WebKit, but less so for WebKit2 for at least two reasons:

1. The WebView class stores only HCURSORs, so it doesn't have any information about the intended type of the cursor. Would be acceptable to change the HCURSOR members to WebCore::Cursors?

2. The overrideCursor only seems to be set with an HCURSOR, so it also has no information about its intended cursor type.

> I don't believe you're a committer, so you should probably mark your patches cq?. That will get them landed once they've been reviewed.

That's correct, I'm not a committer, but I am new to WebKit development! I'll add the flag from now on. Thanks!

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