Bug 99862 – [chromium] Restrict link highlights to targets that display a hand-cursor

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[W. James MacLean]

[chromium] Restrict link highlights to targets that display a hand-cursor

[W. James MacLean]

Created attachment 169671 [details]
Patch

[W. James MacLean]

Created attachment 169676 [details]
Patch

[W. James MacLean]

Add fixed to rotated div layout test in this revised patch.

[James Robinson]

Comment on attachment 169676 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=169676&action=review

> Source/WebKit/chromium/src/WebViewImpl.cpp:1179
> +static bool invokesHandCursor(Node* node, bool shiftKey)

This seems like a strange heuristic to be applying in the WebKit layer.  Normally this is the sort of decision that WebCore would make.

[W. James MacLean]

(In reply to comment #4)
> (From update of attachment 169676 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=169676&action=review
> 
> > Source/WebKit/chromium/src/WebViewImpl.cpp:1179
> > +static bool invokesHandCursor(Node* node, bool shiftKey)
> 
> This seems like a strange heuristic to be applying in the WebKit layer.  Normally this is the sort of decision that WebCore would make.

Agreed, but there doesn't seem to be a simple way to ask WebCore if a particular node *would* get a hand cursor (I looked at re-factoring EventHandler::selectCursor() to see if I could separate the logic used for auto-cursor selection in a way that would allow it to be used externally, but it seemed  difficult to get a clean re-factoring). I can present a CL based on this if you prefer.

[James Robinson]

(In reply to comment #5)
> (In reply to comment #4)
> > (From update of attachment 169676 [details] [details])
> > View in context: https://bugs.webkit.org/attachment.cgi?id=169676&action=review
> > 
> > > Source/WebKit/chromium/src/WebViewImpl.cpp:1179
> > > +static bool invokesHandCursor(Node* node, bool shiftKey)
> > 
> > This seems like a strange heuristic to be applying in the WebKit layer.  Normally this is the sort of decision that WebCore would make.
> 
> Agreed, but there doesn't seem to be a simple way to ask WebCore if a particular node *would* get a hand cursor (I looked at re-factoring EventHandler::selectCursor() to see if I could separate the logic used for auto-cursor selection in a way that would allow it to be used externally, but it seemed  difficult to get a clean re-factoring). I can present a CL based on this if you prefer.

I'd like to explore that direction.  This case seems very similar.  We're doing a hit-test and want to see what the resulting cursor is.

[W. James MacLean]

Created attachment 171481 [details]
Patch

[W. James MacLean]

I've re-factored this to keep all the logic for determining when a handCursor() is appropriate in EventHandler, short of consulting node->style()-cursor().

PTAL.

[James Robinson]

Comment on attachment 171481 [details]
Patch

R=me

[WebKit Review Bot]

Comment on attachment 171481 [details]
Patch

Clearing flags on attachment: 171481

Committed r132945: <http://trac.webkit.org/changeset/132945>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.