Re: [PATCH] Re: 'make' warning

On Tue, Jun 8, 2010 at 05:39, Nicholas Clark <[email protected]> wrote:
> On Mon, Jun 07, 2010 at 05:11:36PM -0400, Jerry D. Hedden wrote:
>> Attached patch fixes this.
>>
>> On Mon, Jun 7, 2010 at 14:42, Jerry D. Hedden <[email protected]> wrote:
>> > Warning during 'make':
>> >
>> > `sh  cflags "optimize='-Os -pipe -funit-at-a-time -march=pentium4
>> > -mfpmath=sse -mieee-fp -mmmx -msse -msse2'" utf8.o`  utf8.c
>> >          CCCMD =  gcc -DPERL_CORE -c -DPERL_USE_SAFE_PUTENV
>> > -U__STRICT_ANSI__ -DNO_MATHOMS -fno-strict-aliasing -pipe  -Os -pipe
>> > -funit-at-a-time -march=pentium4 -mfpmath=sse -mieee-fp -mmmx -msse
>> > -msse2 -Wall -W -Wextra -Wdeclaration-after-statement -Wendif-labels
>> > utf8.c: In function `Perl_foldEQ_utf8':
>> > utf8.c:2591: warning: empty body in an else-statement
>> > utf8.c:2598: warning: empty body in an else-statement
>
> Good catch. Thanks for the proposed patch, but I think the code would be
> clearer if the warnings are silenced by using a { } block rather than the
> pre-processor. Does the appended patch also silence the warnings on your
> system?
>
> Nicholas Clark
>
> diff --git a/utf8.c b/utf8.c
> index 1f4192f..53085e6 100644
> --- a/utf8.c
> +++ b/utf8.c
> @@ -2588,14 +2588,18 @@ Perl_foldEQ_utf8(pTHX_ const char *s1, char **pe1, register UV l1, bool u1, cons
>         * only go as far as the goal */
>         e1 = g1;
>     }
> -    else assert(e1);    /* Must have an end for looking at s1 */
> +    else {
> +       assert(e1);    /* Must have an end for looking at s1 */
> +    }
>
>     /* Same for goal for s2 */
>     if (g2) {
>         assert(! e2  || e2 >= g2);
>         e2 = g2;
>     }
> -    else assert(e2);
> +    else {
> +       assert(e2);
> +    }
>
>     /* Look through both strings, a character at a time */
>     while (p1 < e1 && p2 < e2) {

Yes.  The patch works.