Re: [vile] Clearing errors?

[Thomas Dickey]

On Sat, Mar 12, 2016 at 08:05:24AM -0800, Marc Simpson wrote:
> On Fri, Mar 11, 2016 at 6:19 PM, Thomas Dickey <address@hidden> wrote:
> > no... it propagates the error up until it's either ignored with a ~force,
> > or to the topmost level.  You can see most of that by turning on the
> > trace, e.g., with the command-line -D option, or in a macro using
> >
> >    ~trace on
> 
> thanks for clarifying.

I've checked this in... will try to cleanup my 9.8r to-do's...

===================================================================
RCS file: RCS/main.c,v
retrieving revision 1.728
diff -u -r1.728 main.c
--- main.c      2015/11/09 00:07:57     1.728
+++ main.c      2016/03/13 23:21:58
@@ -22,7 +22,7 @@
  */
 
 /*
- * $Header: /usr/build/vile/vile/RCS/main.c,v 1.728 2015/11/09 00:07:57 tom 
Exp $
+ * $Header: /usr/build/vile/vile/RCS/main.c,v 1.729 2016/03/13 23:21:58 tom 
Exp $
  */
 
 #define realdef                        /* Make global definitions not external 
*/
@@ -1128,6 +1128,12 @@
        kbd_mac_check();
        c = kbd_seq();
 
+       /*
+        * Now that we have started a command, reset "$_".  If we did this at a
+        * lower level, we could not test for it in macros, etc.
+        */
+       tb_scopy(&last_macro_result, status2s(TRUE));
+
        /* reset the contents of the command/status line */
        if (kbd_length() > 0) {
            mlerase();

-- 
Thomas E. Dickey <address@hidden>
http://invisible-island.net
ftp://invisible-island.net

signature.asc
Description: Digital signature