You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@apr.apache.org by "Roy T. Fielding" <fi...@apache.org> on 2002/07/19 19:55:52 UTC
Re: cvs commit: apr-util/include apr_xlate.h apu.h.in apu.hnw apu.hw
Ugh. Is there some reason we don't use normal symbol wrappers
instead of this macro argument name replacement stuff?
....Roy
On Friday, July 19, 2002, at 01:32 AM, wrowe@apache.org wrote:
> +#if APR_HAS_APR_ICONV
> +#define HAVE_ICONV
> +#define iconv_(n) apr_iconv_##n
> +#else
> +#define iconv_(n) iconv_##n
> +#endif
> +
> #ifndef min
> #define min(x,y) ((x) <= (y) ? (x) : (y))
> #endif
> @@ -95,7 +102,7 @@
> char *topage;
> char *sbcs_table;
> #ifdef HAVE_ICONV
> - iconv_t ich;
> + iconv_(_t) ich;
> #endif
> };
Re: cvs commit: apr-util/include apr_xlate.h apu.h.in apu.hnw
apu.hw
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
At 12:55 PM 7/19/2002, Roy T. Fielding wrote:
>Ugh. Is there some reason we don't use normal symbol wrappers
>instead of this macro argument name replacement stuff?
Actually, this is borked anyways, I'll be reverting.
We have a basic problem with the results, apr_iconv() fn's return
apr_status_t, while iconv returns status via errno.
So I needed to rethink this anyways.
Bill
>On Friday, July 19, 2002, at 01:32 AM, wrowe@apache.org wrote:
>> +#if APR_HAS_APR_ICONV
>> +#define HAVE_ICONV
>> +#define iconv_(n) apr_iconv_##n
>> +#else
>> +#define iconv_(n) iconv_##n
>> +#endif
>> +
>> #ifndef min
>> #define min(x,y) ((x) <= (y) ? (x) : (y))
>> #endif
>> @@ -95,7 +102,7 @@
>> char *topage;
>> char *sbcs_table;
>> #ifdef HAVE_ICONV
>> - iconv_t ich;
>> + iconv_(_t) ich;
>> #endif
>> };
>