[Webkit-unassigned] [Bug 35432] New: [Gtk] Make DRT EventSender::keyDown to consider 'hardware_keycode' field when synthesizing an event

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=35432

           Summary: [Gtk] Make DRT EventSender::keyDown to consider
                    'hardware_keycode' field when synthesizing an event
           Product: WebKit
           Version: 528+ (Nightly build)
          Platform: PC
        OS/Version: Mac OS X 10.5
            Status: NEW
          Keywords: Gtk
          Severity: Normal
          Priority: P2
         Component: Tools / Tests
        AssignedTo: webkit-unassigned at lists.webkit.org
        ReportedBy: tonikitoo at webkit.org
                CC: xan.lopez at gmail.com, gns at gnome.org,
                    joanmarie.diggs at gmail.com
            Blocks: 18662


When synthesizing an event in Gtk's DRT EventSender keydownCallback
implementation, an invalid 'hardware_keycode' value
can cause it to be badly processed by Gtk+.

Scenarios:

1) (most of current layout tests fall in this category)
* Focused node an editor (e.g. <input type=text> , <textarea>, etc).
* EventSender::keyDown("arrowDown") is called from JS.
* webkit_web_view_key_press is called and the event is processed by WebCore.

bool webkit_web_view_key_press_event (...)
(...)
  if (frame->eventHandler()->keyEvent(keyboardEvent))
    return TRUE;
(...)


2) On the other hand, in spatial navigation tests in bug 18662 ...

* Focused node is a link (i.e. <a href="xxx">)
* EventSender::keyDown("arrowDown") is called from JS.
* webkit_web_view_key_press does not proccess the event and fallback to GTK+

bool webkit_web_view_key_press_event(...)
(...)
  /* Chain up to our parent class for binding activation */
  return
GTK_WIDGET_CLASS(webkit_web_view_parent_class)->key_press_event(widget, event);


|gtk_bindings_activate_event| fails at some point because hardware_keycode is
not provided.


patch coming ...

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