Hi Bill,
I think this patch is another candidate for 0.9.2 -- or at least for
httpd-2.0.45. It turns out that, when I was moving the native charset
detection from apr-util to apr, I forgot to move the configury, too. So
now, apr-xlate doean't detect the native charset on Unix. :-(
Here's the patch:
Index: configure.in
===================================================================
RCS file: /home/cvs/apr/configure.in,v
retrieving revision 1.516
diff -u -u -r1.516 configure.in
--- configure.in 19 Feb 2003 12:48:48 -0000 1.516
+++ configure.in 6 Mar 2003 22:07:03 -0000
@@ -1833,6 +1833,12 @@
AC_SUBST(have_ipv6)
+dnl Check for langinfo support
+
+APR_FLAG_HEADERS(langinfo.h)
+APR_FLAG_FUNCS(nl_langinfo)
+APR_CHECK_DEFINE(CODESET, langinfo.h, [CODESET defined in langinfo.h])
+
dnl ----------------------------- Finalize the variables
echo "${nl}Restore user-defined environment settings..."
Index: misc/unix/charset.c
===================================================================
RCS file: /home/cvs/apr/misc/unix/charset.c,v
retrieving revision 1.2
diff -u -u -r1.2 charset.c
--- misc/unix/charset.c 1 Jan 2003 00:01:52 -0000 1.2
+++ misc/unix/charset.c 6 Mar 2003 22:07:03 -0000
@@ -53,6 +53,7 @@
*/
#include "apr.h"
+#include "apr_private.h"
#include "apr_strings.h"
#include "apr_portable.h"
Unfortunately, I don't have a Unix box available to test this at the
moment. IMHO a sufficient test would be to apply this patch and run
configure, then check that HAVE_LANGINFO_H, HAVE_NL_LANGINFO and
HAVE_CODESET are mentioned in apr_private.h.
Sorry I didn't notice this before.
Branko Äibej wrote:
>Vladimir Prus wrote:
>
>
>
>>Branko Äibej wrote:
>>
>>
>>
>>
>>>Can you look into your apr.h and apr_private.h and see if those
>>>constants are defined in one of them? It's possible that charset.c
>>>should include apr_private.h, too.
>>>
>>>
>>>
>>>
>>No, neither symbol in neither of files.
>>
>>
>>
>>
>>
>>>If you hurry (i.e., get me that info
>>>today), I can push a fix into the upcoming apr-0.9.2 (and httpd-2.0.45).
>>>
>>>
>>>
>>>
>>I guess there's something deeper? BTW, I note that apu_config.h does declare
>>both of those symbols.
>>
>>
>>
>>
>
>O.K., then I guess I screwed up when I moved this code from apr-util to
>apr. I'll have to hack the configury, too. I'll try to do this tonight.
>
>Thanks for looking!
>
>
--
Brane Äibej <[EMAIL PROTECTED]> http://www.xbc.nu/brane/
