On 5/6/2012 7:56 AM, Hanspeter Niederstrasser wrote:
> On 05/06/2012 12:30 AM, Alexander Hansen wrote:
>> On 5/5/12 8:27 PM, Jack Howarth wrote:
>>> Is anyone else seeing this build failure for the new
>>> libgnomecups-shlibs-0.2.3-8
>>> packaging on 10.7 fink?
>>>
>>> gcc -dynamiclib -o .libs/libgnomecups-1.0.1.0.0.dylib
>>> .libs/gnome-cups-i18n.o .libs/gnome-cups-init.o .libs/gnome-cups-printer.o
>>> .libs/gnome-cups-queue.o .libs/gnome-cups-request.o .libs/gnome-cups-util.o
>>> .libs/util.o -L/sw/lib/system-openssl/lib -L/sw/lib
>>> /sw/lib/libgobject-2.0.dylib /sw/lib/libgthread-2.0.dylib
>>> /sw/lib/libglib-2.0.dylib /sw/lib/libintl.dylib -install_name
>>> /sw/lib/libgnomecups-1.0.1.dylib -compatibility_version 2 -current_version
>>> 2.0
>>> Undefined symbols for architecture x86_64:
>>> "_ippDelete", referenced from:
>>> _update_printers in gnome-cups-printer.o
>>> _gnome_cups_printer_set_description in gnome-cups-printer.o
>>> _gnome_cups_printer_set_location in gnome-cups-printer.o
>>> _gnome_cups_printer_pause in gnome-cups-printer.o
>>> _gnome_cups_printer_resume in gnome-cups-printer.o
>>> _gnome_cups_printer_delete in gnome-cups-printer.o
>>> _update_default in gnome-cups-printer.o
>>> ...
>>> "_ippFindAttribute", referenced from:
>>> _update_printers in gnome-cups-printer.o
>
>>
>> Same here, which I guess is good because I've got a similar setup.
>
> Ditto. This is due to I presume a bug in the latest revision that
> attempts to remove the -liconv flag from "cups-config --libs" output,
> but instead wipes the entire line if there.
>
> The recent gtk+2 checkin has the same underlying issue, though different
> result, because 'grep -v -- -liconv' doesn't remove just that flag, it
> wipes the entire line if present.
As a follow up, replacing 'grep -v -- -liconv' in the patch with
sed 's/-liconv//g'
fixes the libgnomecups-shlibs build.
For gtk+2, the patch needs to be applied to configure (not configure.in
as currently happens) or the fix doesn't get applied to the actual
build. The patch below takes care of 1) changing the bad grep call to
sed, and 2) moving the chunk from configure.in to configure.
diff -u -r1.3 gtk+2.patch
--- gtk+2.patch 4 May 2012 22:48:26 -0000 1.3
+++ gtk+2.patch 6 May 2012 14:50:54 -0000
@@ -30,15 +30,12 @@
$as_echo "#define HAVE_RANDR 1" >>confdefs.h
-diff -Nurd -x'*~' gtk+-2.18.9.orig/configure.in gtk+-2.18.9/configure.in
---- gtk+-2.18.9.orig/configure.in 2010-03-17 09:31:39.000000000 -0400
-+++ gtk+-2.18.9/configure.in 2012-05-03 22:16:14.000000000 -0400
-@@ -1860,7 +1860,7 @@
+@@ -24713,7 +24713,7 @@
fi
else
- CUPS_CFLAGS=`$CUPS_CONFIG --cflags | sed 's/-O[0-9]*//' | sed
's/-m[^\t]*//g'`
+ CUPS_CFLAGS=`$CUPS_CONFIG --cflags | sed 's/-O0-9*//' | sed
's/-m^\t*//g'`
- CUPS_LIBS=`$CUPS_CONFIG --libs`
-+ CUPS_LIBS=`$CUPS_CONFIG --libs | grep -v -- -liconv`
++ CUPS_LIBS=`$CUPS_CONFIG --libs | sed 's/-liconv//g'`
CUPS_API_VERSION=`$CUPS_CONFIG --api-version`
CUPS_API_MAJOR=`echo $ECHO_N $CUPS_API_VERSION | awk -F. '{print
$1}'`
Hanspeter
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
List archive:
http://news.gmane.org/gmane.os.apple.fink.devel
Subscription management:
https://lists.sourceforge.net/lists/listinfo/fink-devel
