http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56081
--- Comment #2 from Steve Kargl <sgk at troutmask dot apl.washington.edu>
2013-01-23 04:28:01 UTC ---
On Wed, Jan 23, 2013 at 01:27:11AM +0000, kargl at gcc dot gnu.org wrote:
> Index: match.c
> ===================================================================
> --- match.c (revision 195396)
> +++ match.c (working copy)
> @@ -5128,6 +5128,17 @@ gfc_match_select (void)
> if (m != MATCH_YES)
> return m;
>
> + /* R812 case-expr is scalar-int-expr
> + or scalar-char-expr
> + or scalar-logical-expr. */
> +
> + if (expr->rank != 0)
> + {
> + gfc_error ("case-expr of SELECT construct at %C should be a "
> + "scalar-int-expr, scalar-char-expr, or scalar-logical-expr");
> + return MATCH_ERROR;
> + }
> +
> new_st.op = EXEC_SELECT;
> new_st.expr1 = expr;
Patch didn't work. Time to scratch head. :-/
