[patch] Deal with larger anti-ranages in compare_range_with_value

[James A. Morrison]

Tom Truscott <trt@unx.sas.com> writes:

> Interesting.  But don't EQ/NE work the same way?
> E.g. ~[0,1] != 0  should be considered true, I think.
> Something like:

 Yes they do.  I realized this this morning.

>      /* VAL_1 <= VAL <= VAL_2 implies ~[VAL1, VAL2] != VAL  */
>       if ((min_comp == 0 || min_comp == -1) && (max_comp == 0 || max_comp == 1))
>         return (comp == NE_EXPR) ? boolean_true_node : boolean_false_node;
>
> Tom Truscott
>


-- 
Thanks,
Jim

http://www.csclub.uwaterloo.ca/~ja2morri/
http://phython.blogspot.com
http://open.nit.ca/wiki/?page=jim