While working on bug #52878 I noticed that PlatformKeyboardEvent::getCurrentModifierState() is currently unimplemented in the gtk port. I don't know if any test is failing because of this, but it could be easily implemented.
Created attachment 79719 [details] Patch implementing getCurrentModifierState()
I grepped the entire source tree an only found implementations with no callers. Perhaps it makes more sense just to remove this method, as it appears to be dead code.
Comment on attachment 79719 [details] Patch implementing getCurrentModifierState() This does the current global event, not the actual object in question, right?
(In reply to comment #3) > (From update of attachment 79719 [details]) > This does the current global event, not the actual object in question, right? Yes, it takes the event from the global event list
(In reply to comment #2) > I grepped the entire source tree an only found implementations with no callers. Perhaps it makes more sense just to remove this method, as it appears to be dead code. Sounds like a good idea to me. About this patch.. is the MOD1 key guaranteed to be Alt?
(In reply to comment #5) > (In reply to comment #2) > > I grepped the entire source tree an only found implementations with no callers. Perhaps it makes more sense just to remove this method, as it appears to be dead code. > > Sounds like a good idea to me. > > About this patch.. is the MOD1 key guaranteed to be Alt? GTK docs says: "It depends on the modifier mapping of the X server which key is interpreted as this modifier, but normally it is the Alt key" we are currently using it everywhere in gtk port, in MouseEventGtk.cpp we have a FIXME comment though: // FIXME: Would be even better to figure out which modifier is Alt instead of always using GDK_MOD1_MASK.
Comment on attachment 79719 [details] Patch implementing getCurrentModifierState() Seems this is used in DragController.cpp for createMouseEvent.
Committed r79151: <http://trac.webkit.org/changeset/79151>