devel/gobject-introspection lässt sich nicht kompilieren

[grisu]

Hallo zusammen,
ich hänge gerade beim allwöchentlichen update lauf am port devel/gobject-introspection. Diese lässt auch meinem FreeBSD 8/i386 nicht bauen, da er während des buildvorgangs mit einem fehler der durch einen Python skript verursacht wird, abbricht.
Als Fehler kommt

Making all in gir
gmake[2]: Entering directory `/usr/ports/devel/gobject-introspection/work/gobject-introspection-0.6.10/gir'
  GISCAN GLib-2.0.gir
Traceback (most recent call last):
  File "../tools/g-ir-scanner", line 38, in <module>
    sys.exit(scanner_main(sys.argv))
  File "/usr/ports/devel/gobject-introspection/work/gobject-introspection-0.6.10/giscanner/scannermain.py", line 296, in scanner_main
    ss.parse_files(filenames)
  File "/usr/ports/devel/gobject-introspection/work/gobject-introspection-0.6.10/giscanner/sourcescanner.py", line 225, in parse_files
    self._parse(headers)
  File "/usr/ports/devel/gobject-introspection/work/gobject-introspection-0.6.10/giscanner/sourcescanner.py", line 258, in _parse
    stdout=subprocess.PIPE)
  File "/usr/local/lib/python2.6/subprocess.py", line 633, in __init__
    errread, errwrite)
  File "/usr/local/lib/python2.6/subprocess.py", line 1139, in _execute_child
    raise child_exception
OSError: [Errno 2] No such file or directory
gmake[2]: *** [GLib-2.0.gir] Fehler 1
gmake[2]: Leaving directory `/usr/ports/devel/gobject-introspection/work/gobject-introspection-0.6.10/gir'
gmake[1]: *** [all-recursive] Fehler 1
gmake[1]: Leaving directory `/usr/ports/devel/gobject-introspection/work/gobject-introspection-0.6.10'
gmake: *** [all] Fehler 2
*** Error code 1

Ich hab python 2.6 noch mal neu installiert, mit standard optionen, doch auch das hat keinen erfolg gebracht. In /usr/ports/UPDATING ist auch kein wort zu diesem port erwähnt.

Ich wäre dankbar für schnelle hilfe.

 

[unmatrix]

versuch python mal ohne pth zu bauen - hat bei mir geholfen

 

[grisu]

Da python ist ohne pth installiert, kann es daran nicht liegen.

 

[morromett]

Hast Du gamin installiert?

Removing GAMIN allows gobject-introspection to build.
This problem has been bothering me for some time and I filed a bug but my system was blamed and "tinderbox works for me" stuff.
Lots of people cant built gobject-introspection its easy to see on google.
Anyway, gamin is causing this and things like g-ir-scanner to fail on a number of systems and removing gamin allowed things to start working again.
reference below.

Newsgroups: mailing.freebsd.ports
From: g...@lynge.org (Gert Lynge)
Date: Tue, 27 Apr 2010 16:29:37 +0200
Local: Tues, Apr 27 2010 7:29 am
Subject: Make of devel/gobject-introspection prevented by devel/gamin
Reply | Reply to author | Forward | Print | Individual message | Show original | Report this message | Find messages by this author
Hi list

This one has been annoing me for a while. I have not been able to  
google a solution, but it seems several people are having this  
problem, so I would like to share my findings.

Both on FreeBSD 7.2, 7.3 and 8.0 I have experienced problems making  
devel/gobject-introspection. It simply freezes after outputting:
---
libtool: link: (cd ".libs" && rm -f  
"libgirepository-everything-1.0.so" && ln -s  
"libgirepository-everything-1.0.so.1"  
"libgirepository-everything-1.0.so")
libtool: link: (cd ".libs" && rm -f  
"libgirepository-everything-1.0.so" && ln -s  
"libgirepository-everything-1.0.so.1"  
"libgirepository-everything-1.0.so")
libtool: link: ar cru .libs/libgirepository-everything-1.0.a  
libgirepository_everything_1_0_la-everything.o
libtool: link: ranlib .libs/libgirepository-everything-1.0.a
libtool: link: ( cd ".libs" && rm -f  
"libgirepository-everything-1.0.la" && ln -s  
"../libgirepository-everything-1.0.la"  
"libgirepository-everything-1.0.la" )
env LPATH=.libs  env PYTHONPATH=..:..:YTHONPATH  
UNINSTALLED_INTROSPECTION_SRCDIR=..  
UNINSTALLED_INTROSPECTION_BUILDDIR=.. ../tools/g-ir-scanner -v  
--add-include-path=../gir --add-include-path=. -v  
--add-include-path=../gir --add-include-path=. --namespace=GLib  
--nsversion=2.0 --libtool="/bin/sh /usr/local/bin/libtool"  
--library=glib-2.0 --pkg=glib-2.0  --noclosure --strip-prefix=g  
--c-include="glib.h" -I/usr/local/include  
-I/usr/local/include/glib-2.0 -DGETTEXT_PACKAGE=Dummy  
-D__G_I18N_LIB_H__ /usr/local/include/glib-2.0/glibconfig.h  
/usr/local/include/glib-2.0/glib/*.h ./glib-2.0.c --output GLib-2.0.gir
In file included from <stdin>:23:
/usr/local/include/glib-2.0/glib/ghostutils.h:21:2: error: #error  
"Only <glib.h> can be included directly."
/usr/include/machine/endian.h:107: syntax error, unexpected '{' in '  
return (__extension__ ({ register __uint32_t __X = (_x); __asm ("bswap  
%0" : "+r" (__X)); __X; }));' at '{'
/usr/include/machine/endian.h:107: syntax error, unexpected ';' in '  
return (__extension__ ({ register __uint32_t __X = (_x); __asm ("bswap  
%0" : "+r" (__X)); __X; }));' at ';'
---

Finaly I setup a cloned system in a Sun Virtual Box and made a script  
which removes my 250 ports one by one while trying to portupgrade  
devel/gobject-introspection. A separate script killed the process  
named "[GLib-2.0]" (which seems to be the process hanging the make) a  
couple of seconds after it was found.
The script was run on:
bsd# uname -a
FreeBSD bsd.xxxxxxx 8.0-RELEASE-p2 FreeBSD 8.0-RELEASE-p2 #0: Tue Jan  
5 16:02:27 UTC 2010      
r...@i386-builder.daemonology.net:/usr/obj/usr/src/sys/GENERIC  i386

My findings is that if devel/gamin is installed, a make of  
devel/gobject-introspection will hang. If I deinstall devel/gamin  
first, I'm able to make devel/gobject-introspection without any  
problems. But as soon as I install devel/gamin again a make in  
devel/gobject-introspection will hang.

I hope this will help others having this problem.
Please let me know if anyone is interested in looking into fixing the  
problem and I can provide more info. (I've been using FreeBSD for  
several years but are not familiar with debugging or bug-reporting :-)).

Regards
     Gert Lynge

 

[grisu]

Als es geht weder mit gamin noch ohne also daran kann es nicht liegen, ich hab jetzt acuh noch mal Python 2.5 probiert. das führte zu keiner änderung.