GNU bug report logs - #11134
24.0.94; Mac's Text to Speech stopped working in emacs buffers when I moved from Snow Leopard to Lion

Previous Next

Full log

Message #26 received at 11134 <at> debbugs.gnu.org (full text, mbox):

From: YAMAMOTO Mitsuharu <mituharu <at> math.s.chiba-u.ac.jp>
To: Jan Djärv <jan.h.d <at> swipnet.se>
Cc: Charles Magid <cmmagid <at> gmail.com>, 11134 <at> debbugs.gnu.org
Subject: Re: bug#11134: 24.0.94;
	Mac's Text to Speech stopped working in emacs buffers when I
	moved	from Snow Leopard to Lion
Date: Mon, 02 Apr 2012 07:50:47 +0900

>>>>> On Sun, 1 Apr 2012 11:10:48 +0200, Jan Djärv <jan.h.d <at> swipnet.se> said:

>> Is it safe to call Fbuffer_string (in particular, move the gap)
>> inside read_socket_hook?  It is apparently unsafe without
>> SYNC_INPUT, but I'm not sure if it is supposed to be OK for the
>> SYNC_INPUT case.

> Can you show where move_gap runs Lisp code?

What I was concerned was not about running Lisp code.  If
read_socket_hook is called while the Lisp interpreter is operating on
some buffer contents through a pointer, then you shouldn't move the
gap in the middle of such an operation.  At least without SYNC_INPUT,
this actually happens.

> Updating the properties might run hooks, so we probably don't want
> to use Fbuffer_string for that reason.

And another reason would be what you are interested in is not always
the same as `current_buffer', but probably `XBUFFER (XWINDOW
(f->selected_window)->buffer)'.

				     YAMAMOTO Mitsuharu
				mituharu <at> math.s.chiba-u.ac.jp

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.