Trade Me - a message from maori_with_guns
Trade Me Logo Security Note: Trade Me will never ask you for your password via email This is an automated email regarding listing reference: 64307169 A trademe.co.nz member has asked a question on your auction(s). Hi there, can I have your phone number ? Or let me know if you received my other email. bye posted by:maori_with_guns (23 [IMAGE]) Address Verified 06/05/06 8:04 am, Fri 04 Aug Answer this question The question will only be displayed on your auction if you answer it. We recommend you answer all reasonable questions on your auctions as it helps buyers to make informed decisions. Happy trading! The Trade Me Team www.trademe.co.nz To change to plain-text emails click here oldfriends.co.nz Ever wondered what happened to the people from your primary school? Look them up now!
Re: error building unrar 3.54p0
On Wed, Aug 02, 2006 at 11:02:08AM +0200, the unit calling itself Marc Espie wrote: 3) both ftp sites *and* local packages can be specified in PKG_PATH Of course ;-) So perhaps if I had followed this sequence I could have avoided the problems I encountered: # setenv PKG_PATH ftp://ftp3.usa.openbsd.org/pub/OpenBSD/3.9/packages/i386/;/usr/ports/packages/i386/all Nope. What is that ; ? Use : to separate elements, as usual. pkg_add is smart enough to desambiguate ftp:// from path separators. thanks... the ; is the same thing as desambiguate - a typo :) The following seems to work for me: # setenv PKG_PATH ftp://ftp3.usa.openbsd.org/pub/OpenBSD/3.9/packages/i386/:/usr/ports/packages/i386/all/ Note that, in most cases, you can use `make update' to update the installation of existing packages. `make update FORCE_UPDATE=Yes' will even force the update with precisely -F update -F updatedepends -F installed I'll make a note to add a quick description of make update to ports(7), since it's currently documented in bsd.port.mk(5), which is a bit too much for the average user Yes... I'm quite sure I would have never found it there ;) Best Rgds, J PS: ports and package management have come an incredibly long way since 2.6. I understand you are largely responsible for these improvements - many thanks for your hard work!
x11/qt4 failure on make package
Hi,
can anyone reproduce the error below, or is it just a pebkac?
...
=== Building package for qt4-4.1.3
Error in package:
/usr/obj/qt4-4.1.3/fake-i386//usr/local/lib/qt4/mkspecs/openbsd-g++/qmake.conf
does not exist
Error in package:
/usr/obj/qt4-4.1.3/fake-i386//usr/local/lib/qt4/mkspecs/openbsd-g++/qplatformdefs.h
does not exist
...
$ ls -l /usr/obj/qt4-4.1.3/fake-i386/usr/local/lib/qt4/mkspecs/openbsd-g++
lrwxr-xr-x 1 root wheel 11 Aug 2 20:46
/usr/obj/qt4-4.1.3/fake-i386/usr/local/lib/qt4/mkspecs/openbsd-g++ -
openbsd-g++
I did a retry just an hour ago with a current system and ports tree
without any change. I also tried to build it on a box with absolutely
no packages installed. And I didn't yet find the reason for the
silly symlink.
FYI:
$ cat /etc/mk.conf
DISTDIR=/var/ports/distfiles
PKGREPOSITORYBASE= /var/ports/packages/${MACHINE_ARCH}
LOCKDIR=/var/ports/locks/${MACHINE_ARCH}
WRKOBJDIR= /usr/obj
PORTSDIR_PATH= ${PORTSDIR}/mystuff:${PORTSDIR}
BULK_COOKIES_DIR= /var/ports/bulk/${MACHINE_ARCH}
UPDATE_COOKIES_DIR= /var/ports/update/${MACHINE_ARCH}
LOCK_CMD= /usr/ports/infrastructure/build/dolock
UNLOCK_CMD= /bin/rm -f
SUDO= /usr/bin/sudo
USE_SYSTRACE= Yes
BULK= Yes
BIN_PACKAGES= Yes
TRUST_PACKAGES= Yes
FORCE_UPDATE= Yes
Ciao,
Kili
--
If you don't know what you are talking about, why are you talking?
-- Theo de Raadt
Re: x11/qt4 failure on make package
Matthias Kilian wrote: can anyone reproduce the error below, or is it just a pebkac? ... === Building package for qt4-4.1.3 Error in package: /usr/obj/qt4-4.1.3/fake-i386//usr/local/lib/qt4/mkspecs/openbsd-g++/qmake.conf does not exist Error in package: /usr/obj/qt4-4.1.3/fake-i386//usr/local/lib/qt4/mkspecs/openbsd-g++/qplatformdefs.h does not exist ... I saw this several months ago, the culprit is: USE_SYSTRACE= Yes Moritz
py-ctypes propolice and AMD64
Hi,
steven@ noticed that py-ctypes was building without ProPolice because
of some legacy paranoia in the build system. This patch re-enables
propolice, enables builds on AMD64 and disabled builds on arm platforms
because libffi (part of py-ctypes) has not been ported.
Ok?
-d
Index: Makefile
===
RCS file: /cvs/ports/devel/py-ctypes/Makefile,v
retrieving revision 1.2
diff -u -p -r1.2 Makefile
--- Makefile22 Jul 2006 00:31:29 - 1.2
+++ Makefile6 Aug 2006 00:37:33 -
@@ -10,6 +10,9 @@ MAINTAINER= Damien Miller [EMAIL PROTECTED]
MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=ctypes/}
HOMEPAGE= http://starship.python.net/crew/theller/ctypes/
+# libffi has not been ported to arm yet
+NOT_FOR_ARCHS= arm
+
# BSD-like
PERMIT_PACKAGE_CDROM= Yes
PERMIT_PACKAGE_FTP=Yes
Index: patches/patch-source_libffi_configure
===
RCS file: patches/patch-source_libffi_configure
diff -N patches/patch-source_libffi_configure
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-source_libffi_configure 6 Aug 2006 00:37:33 -
@@ -0,0 +1,11 @@
+$OpenBSD$
+--- source/libffi/configure.orig Sun Aug 6 00:04:07 2006
source/libffi/configureSun Aug 6 00:04:12 2006
+@@ -3470,6 +3470,7 @@ fi
+
+ TARGETDIR=unknown
+ case $host in
++x86_64-*-openbsd*) TARGET=X86_64; TARGETDIR=x86;;
+ mips*-*-openbsd*) TARGET=MIPS; TARGETDIR=mips;;
+ sparc-*-openbsd*) TARGET=SPARC; TARGETDIR=sparc;;
+ sparc64-*-openbsd*) TARGET=SPARC; TARGETDIR=sparc;;
Index: patches/patch-source_libffi_configure_ac
===
RCS file: patches/patch-source_libffi_configure_ac
diff -N patches/patch-source_libffi_configure_ac
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-source_libffi_configure_ac6 Aug 2006 00:37:33 -
@@ -0,0 +1,11 @@
+$OpenBSD$
+--- source/libffi/configure.ac.origSat Aug 5 20:37:57 2006
source/libffi/configure.ac Sat Aug 5 20:39:30 2006
+@@ -21,6 +21,7 @@ AC_FUNC_MMAP_BLACKLIST
+
+ TARGETDIR=unknown
+ case $host in
++x86_64-*-openbsd*) TARGET=X86_64; TARGETDIR=x86;;
+ mips*-*-openbsd*) TARGET=MIPS; TARGETDIR=mips;;
+ sparc-*-openbsd*) TARGET=SPARC; TARGETDIR=sparc;;
+ sparc64-*-openbsd*) TARGET=SPARC; TARGETDIR=sparc;;
Index: patches/patch-source_libffi_fficonfig_py_in
===
RCS file: patches/patch-source_libffi_fficonfig_py_in
diff -N patches/patch-source_libffi_fficonfig_py_in
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-source_libffi_fficonfig_py_in 6 Aug 2006 00:37:33 -
@@ -0,0 +1,10 @@
+$OpenBSD$
+--- source/libffi/fficonfig.py.in.orig Sat Aug 5 20:35:43 2006
source/libffi/fficonfig.py.in Sat Aug 5 20:36:25 2006
+@@ -43,6 +43,3 @@ ffi_sources += ffi_platforms['@MKTARGET@
+ ffi_sources = [os.path.join('@srcdir@', f) for f in ffi_sources]
+
+ ffi_cflags = '@CFLAGS@'
+-# I think this may no longer be needed:
+-if sys.platform == openbsd3:
+-ffi_cflags += -fno-stack-protector
