Created attachment 45438 [details] [review]
proposed patch

Kjartan, didn't you have a patch for this already?

Attaching patches for both gst-plugins and gstreamer. Not sure they are in an
acceptable shape yet...

Created attachment 45535 [details] [review]
warning fixes for gst-plugins

Created attachment 45536 [details] [review]
fixes for gst core

Comment on attachment 45536 [details] [review]
fixes for gst core

-  specs = g_object_class_list_properties (G_OBJECT_GET_CLASS (object),
&nspecs);
+  specs = g_object_class_list_properties (G_OBJECT_GET_CLASS (object), (guint
*)&nspecs);

Can that be done differently? E.g. by making nspecs a guint already? As for the
rest, I see libxml insanity, which apparently requires this fix. The rest is
sane.

The core patch is committed with the above one slightly modified. Looking at
plugins patch now.

Just tried compiling core CVS and got the following:
cc1: warnings being treated as errors
gstpad.c: In function 'gst_pad_event_default':
gstpad.c:4124: warning: pointer targets in passing argument 3 of
'gst_event_discont_get_value' differ in signedness
gstpad.c:4134: warning: pointer targets in passing argument 5 of
'gst_pad_convert' differ in signedness
make[4]: *** [libgstreamer_0.8_la-gstpad.lo] Error 1

Above fixed by Ronalds Guint to gint change,
Here is the next one
cc1: warnings being treated as errors
gsttrace.c: In function 'gst_trace_flush':
gsttrace.c:114: warning: ignoring return value of 'write', declared with
attribute warn_unused_result
gsttrace.c: In function 'gst_trace_text_flush':
gsttrace.c:136: warning: ignoring return value of 'write', declared with
attribute warn_unused_result

Created attachment 45543 [details] [review]
fixes the rest for gstreamer core

Git/did this when applying this patch 

patching file gst/indexers/gstfileindex.c
can't find file to patch at input line 413
Perhaps you used the wrong -p or --strip option?
The text leading up to this was:
--------------------------
|Index: gst/schedulers/gstoptimalscheduler.c
|===================================================================
|RCS file:
/cvs/gstreamer/gstreamer/gst/schedulers/Attic/gstoptimalscheduler.c,v|retrieving
revision 1.109
|diff -u -p -r1.109 gstoptimalscheduler.c
|--- gst/schedulers/gstoptimalscheduler.c       2 Feb 2005 16:15:18 -0000      1.109
|+++ gst/schedulers/gstoptimalscheduler.c       22 Apr 2005 11:09:05 -0000
--------------------------
File to patch:
Skip this patch? [y] y
Skipping patch.
1 out of 1 hunk ignored
patching file libs/gst/dataprotocol/dataprotocol.c
Hunk #2 succeeded at 379 (offset -14 lines).

nm, patched 0.9 by mistake

Still have this one after Kjartans patch:

cc1: warnings being treated as errors
gsttrace.c: In function 'gst_trace_flush':
gsttrace.c:114: warning: ignoring return value of 'write', declared with
attribute warn_unused_result
gsttrace.c: In function 'gst_trace_text_flush':
gsttrace.c:136: warning: ignoring return value of 'write', declared with
attribute warn_unused_result

Been building RPMS up to now, just running make fails with this:
cc1: warnings being treated as errors
runxml.c: In function ‘main’:
runxml.c:47: warning: pointer targets in passing argument 2 of
‘gst_xml_parse_file’ differ in signedness
make[3]: *** [runxml.o] Error 1

Kjartan, I applied half of your patch, I did the other half differently (but
thanks to your patch for making me notice them). Can you both please test
current CVS again and give me a clear list of outstanding issues?

Christian, for the write, change:

[..]
write (..);
[..]

to:

int unused;
[..]
unused = write (..);
[..]

and let me know if that fixes it.

Getting this with current CVS

 gcc -DHAVE_CONFIG_H -I. -I. -I../../.. -pthread -I/usr/include/glib-2.0
-I/usr/lib/glib-2.0/include -I/usr/include/libxml2 -D_LARGEFILE_SOURCE
-D_FILE_OFFSET_BITS=64 -fno-common -g -Wall -Werror -DGST_DISABLE_DEPRECATED
-I../../../libs -I../../.. -I../../../libs -DGST_ENABLE_NEW -g -O2 -MT
libgstdataprotocol_la-dataprotocol.lo -MD -MP -MF
.deps/libgstdataprotocol_la-dataprotocol.Tpo -c dataprotocol.c  -fPIC -DPIC -o
.libs/libgstdataprotocol_la-dataprotocol.o
dataprotocol.c: In function 'gst_dp_packet_from_caps':
dataprotocol.c:279: error: syntax error before '<<' token
dataprotocol.c:283:9: error: too many decimal points in number
dataprotocol.c:290: error: syntax error before '<<' token
dataprotocol.c:294:9: error: too many decimal points in number
dataprotocol.c: In function 'gst_dp_packet_from_event':
dataprotocol.c:404: error: syntax error before '<<' token
dataprotocol.c:408:9: error: too many decimal points in number
dataprotocol.c: In function 'gst_dp_caps_from_packet':
dataprotocol.c:472: error: syntax error before '<<' token
dataprotocol.c:476:9: error: too many decimal points in number
cc1: warnings being treated as errors
dataprotocol.c:465: warning: unused variable 'string'
make[4]: *** [libgstdataprotocol_la-dataprotocol.lo] Error 1
make[4]: Leaving directory `/home/cschalle/gst/08/gstreamer/libs/gst/dataprotocol'
make[3]: *** [all-recursive] Error 1
make[3]: Leaving directory `/home/cschalle/gst/08/gstreamer/libs/gst'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/home/cschalle/gst/08/gstreamer/libs'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/cschalle/gst/08/gstreamer'
make: *** [all] Error 2

That's because you are trying to compile with CVS conflicts in your tree :)

Whoops :) ok, still fails with the same message and my #14 entry after a clean
checkout.

Fixed, pelase try again.

make[3]: Entering directory `/home/cschalle/gst/08/gstreamer/tests/seeking'
if gcc -DHAVE_CONFIG_H -I. -I. -I../..    -pthread -I/usr/include/glib-2.0
-I/usr/lib/glib-2.0/include -I/usr/include/libxml2   -D_LARGEFILE_SOURCE
-D_FILE_OFFSET_BITS=64   -fno-common -g -Wall -Werror -DGST_DISABLE_DEPRECATED
-I../../libs -I../.. -g -O2 -MT seeking1.o -MD -MP -MF ".deps/seeking1.Tpo" -c
-o seeking1.o seeking1.c; \
then mv -f ".deps/seeking1.Tpo" ".deps/seeking1.Po"; else rm -f
".deps/seeking1.Tpo"; exit 1; fi

verified fixed; next stop is this one:
cc1: warnings being treated as errors
seeking1.c: In function ‘main’:
seeking1.c:58: warning: pointer targets in passing argument 4 of ‘gst_pad_query’
differ in signedness
seeking1.c:60: warning: pointer targets in passing argument 4 of ‘gst_pad_query’
differ in signedness
seeking1.c:83: warning: pointer targets in passing argument 4 of ‘gst_pad_query’
differ in signedness
seeking1.c:85: warning: pointer targets in passing argument 4 of ‘gst_pad_query’
differ in signedness
make[3]: *** [seeking1.o] Error 1

Kjartan, it seems like all testsuite items fail, can you run make check and fix
them? That'll go faster than me fixing each single item reported by Christian.

 gcc -DHAVE_CONFIG_H -I. -I. -I../.. -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64
-pthread -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include
-I/usr/include/libxml2 -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -fno-common -g
-Wall -Werror -DGST_DISABLE_DEPRECATED -mpreferred-stack-boundary=2
-I/usr/include/valgrind -I/usr/include/valgrind/x86
-I/usr/include/valgrind/linux -I/usr/include/valgrind/x86-linux -I../.. -O2 -g
-pipe -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m32 -march=i386 -mtune=pentium4
-fasynchronous-unwind-tables -MT libgstparse_la-lex._gst_parse_yy.lo -MD -MP -MF
.deps/libgstparse_la-lex._gst_parse_yy.Tpo -c lex._gst_parse_yy.c  -fPIC -DPIC
-o .libs/libgstparse_la-lex._gst_parse_yy.o
cc1: warnings being treated as errors
lex._gst_parse_yy.c: In function '_gst_parse_yylex':
parse.l:144: warning: ignoring return value of 'fwrite', declared with attribute
warn_unused_result

Got this trying to build an RPM (without the tests)

Christian, please follow my instructions as given in comment 15 and see if that
fixes it.

Only reference I find to fwrite is this, and I am unsure how to apply the
comment 15 to that.

#define ECHO (void) fwrite( yytext, yyleng, 1, yyout )
#endif

Fixed. Closing.

*** Bug 315078 has been marked as a duplicate of this bug. ***