Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!uwm.edu!linac!att!ucbvax!bloom-beacon!dont-send-mail-to-path-lines
From: leei@thunder.mcrcim.mcgill.EDU
Newsgroups: comp.windows.x
Subject: Re: Two Athena widget questions.
Message-ID: <9104101639.AA06412@thunder.McRCIM.McGill.EDU>
Date: 10 Apr 91 16:39:30 GMT
References: <9103281952.AA15167@excess.lcs.mit.edu>
Sender: daemon@athena.mit.edu (Mr Background)
Organization: The Internet
Lines: 41

In message <9103281952.AA15167@excess.lcs.mit.edu> you write:
>
>
>
>	2) Is the promise that an asciiTextWidget will call a callback when
>	the data is modified a lie?  I can't seem to register the callback.
>	The promise that I'm referring to is in Section 5.5.1 of the Athena
>	Widget Set docs.  Under a description of the AsciiText resources, the
>	entry for callback is:
>
>	callbacks  The callbacks registered on this resource will be called
>		   every time the text buffer changes.  This is an AsciiSrc
>		   resource.
>
>
>The Ascii Text widget does not have a callback resource.
>Add the callback to the Ascii Source, not the Ascii Text.
>You can do GetValues on the AsciiText's textSource resource to get the
>source object, and add the callback to the source object.
>

Thanks for the help.  This seems to work fine, but as I use it I become unclear
on the intended semantics of the callback.  At what time is the callback
generated?  From a quick scan through the AsciiSrc source code, it seems to be
sent before the replace operation has actually been completed, so if I do a
XtVaGetValues( w, XtNstring, &name, NULL) in my callback routine, I will get
the unchanged value!  Well, that is supposing that it works.  In fact, if the
useStringInPlace property is False, this call causes a core dump since the
initialization performed before the XtCallCalbacks call in ReplaceText() is
invalidated by the XawAsciiSave() call via GetValuesHook().  I'd call it a bug,
but since the semantics of the callback seem undefined (at least anywhere I can
see in section 5 of the Athena Widget Set docs) it is not clear to me what I'm
allowed to do in this callback and what not.  If you decide this is a bug, let
me know and I'll file a real bug report.  If the semantics are intended to be
such that by the time the callback is invoked, the string resource should
actually be changed, then let me know that too.

------------------------------------------------------------------------------
Lee Iverson			McGill Research Centre for Intelligent Machines
leei@mcrcim.mcgill.edu		Computer Vision and Robotics Lab
				McGill University, Montreal