hi Tobias,
ive just had a look at your recent version of ncruses ruby (which fixes the
funcall issue), preparing a new debian package. As it stands there is also some
debugging code left, creating some file in /tmp/.
Was this your Intention? Will there be a 1.2.4 soon, without this code?
bye,
- michael
--
+++ ncurses-ruby-1.2.3/ncurses_wrap.c 2009-06-26 00:01:24.000000000 +0200
@@ -18,7 +18,7 @@
* License along with this module; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
- * $Id: ncurses_wrap.c,v 1.16 2009/05/03 20:37:26 t-peters Exp $
+ * $Id: ncurses_wrap.c,v 1.17 2009/06/25 21:53:57 t-peters Exp $
*
* This file was adapted from the original ncurses header file which
* has the following copyright statements:
@@ -210,7 +210,7 @@
/// Portable (1.8,1.9) determination of array length (calling #length)
long rbncurs_array_length(VALUE array)
{
- return NUM2LONG(funcall(array, rb_intern("length"), 0));
+ return NUM2LONG(rb_funcall(array, rb_intern("length"), 0));
}
VALUE wrap_window(WINDOW* window)
@@ -848,6 +848,9 @@
/* sleep on infd until input is available or tv reaches timeout */
FD_ZERO(&in_fds);
FD_SET(infd, &in_fds);
+ FILE * f = fopen("/tmp/xlog","a");
+ fprintf(f, "Callig select\n");
+ fclose(f);
rb_thread_select(infd + 1, &in_fds, NULL, NULL, &tv);
}
#ifdef NCURSES_VERSION
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
