Hi,

On Thu, Jul 22, 2010 at 8:29 AM, Hendrik Sattler
<p...@hendrik-sattler.de> wrote:
> Am Mittwoch 21 Juli 2010, 22:41:20 schrieb Luiz Augusto von Dentz:
>> Well it used work before, so Im not really sure what you have changed
>> that affect this. In the other hand it doesn't seems that obex.c  has
>> any direct include that has the declaration of obex_t and other
>> symbols from openobex/obex.h, apparently you have change it here:
>>
>> http://gitorious.org/openobex/mainline/commit/404cbcc46f8945f539b1f9c5c767c
>> de70a641dee
>
> No, those changes were initiated by Marcel and for the purpose of actually
> using our own headers! This was actually a good move.

I didn't mean we shouldn't have our own headers, but it must compile.

>> Anyway having different build system is (IMO) asking for trouble but I
>> guess it is too late, the inner forking already started with windows
>> building with with cmake and linux with autotools.
>
> First, you can _always_ use cmake as build system, even on Linux. And that
> works. Marcel wants to keep the autotools build system and I didn't object.

Let's leave this discussion for other time, I don't want us to loose
the focus in fixing the build for autotools.

> Try the following patch.
>
> HS
>
> ---
> diff --git a/acinclude.m4 b/acinclude.m4
> index d825aba..6eb84a7 100644
> --- a/acinclude.m4
> +++ b/acinclude.m4
> @@ -23,7 +23,6 @@ AC_DEFUN([COMPILER_FLAGS], [
>                CFLAGS+=" -Wredundant-decls"
>                CFLAGS+=" -Wcast-align"
>        fi
> -       CPPFLAGS+=" -I${srcdir}/include"
>  ])
>
>  AC_DEFUN([AC_INIT_OPENOBEX], [
> @@ -66,14 +65,19 @@ AC_DEFUN([AC_PATH_WIN32], [
>  ])
>
>  AC_DEFUN([AC_PATH_IRDA_LINUX], [
> -       AC_CACHE_CHECK([for IrDA support], irda_found, [
> -               AC_TRY_COMPILE([
> -                               #include <sys/socket.h>
> -                               #include <linux/irda.h>
> -                       ], [
> -                               struct irda_device_list l;
> -                       ], irda_found=yes, irda_found=no)
> -       ])
> +dnl We ship linux/irda.h in $srcdir/include, so it makes no
> +dnl sense to test for it.
> +dnl    AC_CACHE_CHECK([for IrDA support], irda_found, [
> +dnl            AC_TRY_COMPILE([
> +dnl                            #include <sys/socket.h>
> +dnl                            #include <linux/irda.h>
> +dnl                    ], [
> +dnl                            struct irda_device_list l;
> +dnl                    ], irda_found=yes, irda_found=no)
> +dnl    ])
> +       irda_found=yes
> +       AC_MSG_CHECKING([for IrDA support])
> +       AC_MSG_RESULT([$irda_found])
>        irda_linux=$irda_found
>  ])
>
> diff --git a/lib/Makefile.am b/lib/Makefile.am
> index cd48264..897450d 100644
> --- a/lib/Makefile.am
> +++ b/lib/Makefile.am
> @@ -22,7 +22,7 @@ libopenobex_la_SOURCES = \
>        bluez_compat.h \
>        visibility.h defines.h debug.h cloexec.h
>
> -libopenobex_la_CPPFLAGS = @CPPFLAGS@
> +libopenobex_la_CPPFLAGS = -I$(top_srcdir)/include
>  libopenobex_la_CFLAGS = -D_GNU_SOURCE @CFLAG_VISIBILITY@ @USB1_CFLAGS@
>
>  libopenobex_la_LDFLAGS = \

This one did the trick, it finally compiles.

Btw, how have you been testing this changes?

-- 
Luiz Augusto von Dentz
Computer Engineer

------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Openobex-users mailing list
Openobex-users@lists.sourceforge.net
http://lists.sourceforge.net/lists/listinfo/openobex-users

Reply via email to