getzlequery

getzlequery

[Peter Stephenson]

I'm getting a `test is always zero' message from getlzlequery() in
zle_utils.c due to testing whether a char is EOF.  This is something
to do with:

        * Src/utils.c: getquery always returned n on cray-unicos. From
          Richard D. Slater (2136)

Either the test shouldn't be there, or the variable c should be an
int, or the variable should be a signed char and we need to check
whether the compiler supports it (and we presumably lose the ability
to recognise character 255, so that's probably wrong).  I don't know
which avoids reintroducing the UNICOS problem.

-- 
Peter Stephenson <pws@ifh.de>       Tel: +49 33762 77366
WWW:  http://www.ifh.de/~pws/       Fax: +49 33762 77413
Deutches Electronen-Synchrotron --- Institut fuer Hochenergiephysik Zeuthen
DESY-IfH, 15735 Zeuthen, Germany.

Re: getzlequery

[Zoltan Hidvegi]

Peter Stephenson wrote:
> I'm getting a `test is always zero' message from getlzlequery() in
> zle_utils.c due to testing whether a char is EOF.  This is something
> to do with:
> 
>         * Src/utils.c: getquery always returned n on cray-unicos. From
>           Richard D. Slater (2136)

I do not see the relation between these.  However the patch below should fix
that.

Zoltan


--- Src/zle_utils.c	1996/10/15 20:16:35	2.14
+++ Src/zle_utils.c	1996/10/15 20:59:31
@@ -309,7 +309,7 @@
 int
 getzlequery(void)
 {
-    char c;
+    int c;
 #ifdef FIONREAD
     int val;
 
@@ -332,5 +332,5 @@
 
     /* echo response and return */
     putc(c, shout);
-    return (int)c;
+    return c;
 }