Re: RRIDXP 0.2 compile prob on RH 7.3 and quick and dirty workaround

Hi Sandro,

I can confirm this behavior on redhat 7.3
I meant to research and submit a patch but I never got around to it.
I found that on my system the errors were caused by rr_debug statements 
which only had a format parameter and not any variables to be formated.
In example rr_idxp.c:156 :
  rr_debug3 ("idxp::set_greeting_active: Mutex unlocked\n");

after running the preprocessor the rr_debug statement above got expanded to:
  g_log ("RR", RR_LOG_LEVEL_DEBUG3, "idxp::set_greeting_active: Mutex 
unlocked\n",);

Instead of commenting the lines out I found that adding a NULL to the 
end seemed to work:
   rr_debug3 ("idxp::set_greeting_active: Mutex unlocked\n", NULL);

I'm sure there is a way to have something like this done automatically 
in rr-debug.  There are also some statements like this in rrtls too.

Ryan


Sandro Poppi wrote:

>Hi there,
>
>when compiling rridxp 0.2 on a RedHat 7.3 system make stops with the
>following error:
>
>gcc -DHAVE_CONFIG_H -I. -I. -I.. -I.. -I/usr/include/glib-2.0
>-I/usr/lib/glib-2.0/include -I/usr/include/libxml2 -g -O2 -c rridxp.c  -fPIC -DPIC -o
>.libs/rridxp.lo
>rridxp.c: In function `set_greeting_active':
>rridxp.c:156: parse error before `)'
>rridxp.c: In function `frame_available':
>rridxp.c:240: parse error before `)'
>rridxp.c:294: parse error before `)'
>rridxp.c:339: parse error before `)'
>rridxp.c:346: parse error before `)'
>rridxp.c:360: parse error before `)'
>rridxp.c:365: parse error before `)'
>rridxp.c: In function `handle_greeting':
>rridxp.c:503: parse error before `)'
>rridxp.c:723: parse error before `)'
>rridxp.c: In function `rr_idxp_start':
>rridxp.c:821: parse error before `)'
>make[2]: *** [rridxp.lo] Error 1
>
>This happens because both variables G_HAVE_ISO_VARARGS and
>G_HAVE_GNUC_VARARGS are defined and in librr/rr-debug.h if G_HAVE_ISO_VARARGS is defined
>G_HAVE_GNUC_VARARGS is not checked resulting in the mentioned error (checked cpp
>output).
>
>Since I'm not very familiar with autoconf/automake my quick and dirty
>workaround is to comment out the part in librr/rr-debug.h regarding
>G_HAVE_ISO_VARARGS, loosing portability.
>
>You guys for sure will find a much better (and of course portable) solution
>on that issue.
>
>Regards,
>Sandro
>
>  
>