[FIX][BUG] Hidden text cursor ([et] [sm] no effect with the WorldMenu)

Hernan Tylim htylim at yahoo.com.ar
Sun Apr 11 15:17:17 UTC 2004


Hi Ken, 
	Try hitting shift-escape while on a text morph. It should pop up
the world menu. After you dismiss the world menu there is no text
cursor. Don't forget to enable menuKeyboardControl first.

Regards,
Hernán

> -----Original Message-----
> From: squeak-dev-bounces at lists.squeakfoundation.org
[mailto:squeak-dev-
> bounces at lists.squeakfoundation.org] On Behalf Of Hernan Tylim
> Sent: Sunday, April 11, 2004 12:07 PM
> To: ken at kencausey.com; squeak-dev at lists.squeakfoundation.org
> Subject: RE: [FIX][BUG] Hidden text cursor ([et] [sm])
> 
> Hi Ken,
> 
> 	I have no doubt that a MenuMorph should restore the keyboard
> focus to the original owner and for that I think your change set
should
> be approved.  But I am not sure if what I was reporting was happening
> only when invoking menus.
> 
> 	I would like to ask someone who knows about Morphic if would
> make sense to enforce an invariant that says that "if a Morph receives
a
> KeyboardEvent then it must have keyboard focus". To implement such a
> thing I would make that when a KeyboardEvent is sent to a Morph which
> isn't the keyboard focus (right now this only happens if keyboardFocus
> is nil) make it be the keyboard focus.
> 
> 	Returning to your changeset I think it should be approved. In
> the meantime I will use it and check if I found any other scenario
where
> the keyboard focus could be set to nil.
> 
> Regards,
> Hernán
> 
> > -----Original Message-----
> > From: ken at kencausey.com [mailto:ken at kencausey.com]
> > Sent: Sunday, April 11, 2004 11:11 AM
> > To: squeak-dev at lists.squeakfoundation.org; htylim at yahoo.com.ar
> > Subject: [FIX][BUG] Hidden text cursor
> >
> >
> > This is a possible fix for the problem reported in BFAV ID 21797
> 'Hidden
> > text cursor'.  Basically I diagnose the problem as a case of the
menu
> > stealing the keyboard focus and then not properly returning it.
This
> > seems to fix the specific case that Hernan reported but may miss
other
> > cases.  I also wonder if the simple check for non-nil for the
original
> > focus holder is enough in all cases.




More information about the Squeak-dev mailing list