Robert Roessler:
It looks like the SendMessage on line 1175 of ScintillaWin.cxx needs a
bit more work to bring it into conformance with the added state info
changes... shouldn't ctrl and alt be encoded similarly to shift?
Also, why is the only platform-level change here for Win32 - doesn't
GTK need a similar update, or are these state buggers handled
differently enough that it doesn't make sense / isn't needed?
The specialisation for Win32 is not for the Scintilla double click
notification but to also produce a standard Win32 WM_LBUTTONDBLCLK
message since some containers may be expecting these messages.
WM_LBUTTONDBLCLK allows the key modifier flags (MK_CONTROL,
MK_LBUTTON, MK_MBUTTON, MK_RBUTTON, MK_SHIFT) and there may be some
containers that would like more of these set. There is no MK_ALT so
the only one that is easy to add is MK_CONTROL. IIRC the
WM_LBUTTONDBLCLK code was added for a particular project and I doubt
it is widely used so any improvements can come from its users.
Part of the reason for this change was to be more similar to the
Ru-Board version but it adds in the extra bits to WM_LBUTTONDBLCLK
incorrectly using SCI_SHIFT=1, SCI_CTRL=2, and SCI_ALT=4 flags which
are different from the MK_SHIFT=4 and MK_CONTROL=8 expected for
WM_LBUTTONDBLCLK.
Neil
_______________________________________________
Scintilla-interest mailing list
[email protected]
http://mailman.lyra.org/mailman/listinfo/scintilla-interest
