Re: signed < unsigned

[Akim Demaille]

| I'm really sorry, but that wasn't the good patch, the following patch
| is ok, sorry again.
| 
| --- src/bison.simple.origin   Wed Jul 18 18:05:52 2001
| +++ src/bison.simple  Wed Jul 18 18:06:16 2001
| @@ -672,8 +672,8 @@
|         count = 0;
|         /* Start X at -yyn if nec to avoid negative indexes in yycheck.  */
|         for (x = (yyn < 0 ? -yyn : 0);
| -            x < (int) (sizeof (yytname) / sizeof (char *)); x++)
| -         if (yycheck[x + yyn] == x)
| +            x < (sizeof (yytname) / sizeof (char *)); x++)
| +         if ((int) yycheck[x + yyn] == x)
|             size += strlen (yytname[x]) + 15, count++;
|         size += strlen ("parse error, unexpected `") + 1;
|         size += strlen (yytname[YYTRANSLATE (yychar)]);
| @@ -688,8 +688,8 @@
|               {
|                 count = 0;
|                 for (x = (yyn < 0 ? -yyn : 0);
| -                    x < (int) (sizeof (yytname) / sizeof (char *)); x++)
| -                 if (yycheck[x + yyn] == x)
| +                    x < (sizeof (yytname) / sizeof (char *)); x++)
| +                 if ((int) yycheck[x + yyn] == x)
|                     {
|                       strcat (msg, count == 0 ? ", expecting `" : " or `");
|                       strcat (msg, yytname[x]);

Huh?  yycheck is composed of short, and x is int, so everything is
fine here (shorts are promoted to int).  I don't understand the point
of your patch.  I believe the current implementation is the right one.