Re: mouse cursor and iso-accents-mode

[Jan D.]

Jan D. wrote:
> >     No, running with "-q" and renaming .emacs do not solve it.  I also 
> > tried
> >     cleaning out my X resources with "xrdb </dev/null" and got the 
> > same result.
> >
> > I hope that someone who can observe this problem will decide to
> > debug it and track down the actual cause.
> >
> > ======================================================================
> >     Start emacs as an X window and enter iso-accents-mode.  Whenever you
> >     type a composite character, for example `" SPC' for the `"' 
> > character,
> >     the mouse cursor becomes the wristwatch shape and the usual mouse
> >     functions cease to work.  Everything reverts to normal at the next 
> > RET
> >     or C-SPC or C-p or certain other actions.  Normal typing of text,
> >     however, does not cause the mouse to resume functioning.
>
>
> This happens every time on the RC branch and 21.2, but not at all in CVS
> HEAD.  I haven't found yet how to fix this in RC, but I'm looking.

In CVS head there is this:


#if 0

#ifdef HAVE_WINDOW_SYSTEM
  if (display_hourglass_p)
    start_hourglass ();
#endif

#endif

i.e. this makes the hourglass never to be shown.  In RC, the code is not 
surrounded by the same #if 0/#endif.

The changes occured here:

2002-03-04  Pavel Janík  <address@hidden>

        * lread.c (read_filtered_event): Do not call start_hourglass
        before returning.

Do you remember the reason for this change?  Should we delete the code?  What 
are the risks of putting the same fix in RC?

I see unwanted hourglass in RC in other situations as well, for example opening 
a file, modify it but don't save, open the file with emacsclient and when Emacs 
asks, "Revert buffer from file ...", then say yes.  After that the hourglass 
appears, and has to be removed by C-g.  I have a fix for that, but I am not 
sure this is RC material (although at work, we have this program that uses 
Eamcs and emacsclient so I see this bug *alot* :-), I need some opinion on 
that.  I am checking the fix into the trunk.


Thanks,

        Jan D.