Re: ports softlock
* Markus Lude [2010-01-19]: > Packages for sparc64 are somewhat old. Thanks! These are due to hardware problems, we are working on it. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: ports: feature request
* Toni Mueller [2009-12-29]: > I'd very much prefer to be able to set an environment variable, or to > specify a different target which would result in the system not > building all ports and the rest of the world, but fetch as many binary > packages as possible to satisfy dependencies, and only then go to > build the few remaining dependencies that are not available in binary > form. As in FETCH_PACKAGES? See bsd.port.mk(5). Nikolay
Re: Important notice: package version numbering
* Marc Espie [2009-11-12]: > The main consequence is for porters: when you update things, you have > to make sure version numbers go forward. If the version numbering > changes, as happens from time to time, you have to append a v0 to make > sure it still goes forward. That's great! Is "forward" defined anywhere? There are many cases where version numbers change from 1.0alpha -> 1.0rc1 -> 1.0 or something similar. It would be nice to know when to expect having to add "vN" and when there is a bug in your algorithm. ;) cheers, Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: OOo2 build failure
* Mikolaj Kucharski [2009-11-11]: > > editors/openoffice is unlinked from the build. Hence probably not > > tested with latest perl 5.10 and 5.10.1 updates, which probably > > causes this failure. > > Didn't notice that. Makes sense. Thanks. Actually, it doesn't. Why not just remove OO2 completely? Nikolay
Re: Opera - a) Japanese Text b) Flashplugin conflicts with GNOME Dependency
* Robert Gray [2009-10-13]: > 1) First problem - The Port seems to be built with European Language > Settings only; and there are no Asian Language available for the User > to "Choose and Add" to the Browser's language collection. >From my commit message: note: this version of opera has font issue on OpenBSD, there are missing MathML characters and some fonts, like Farsi(?), cannot be displayed at all. I looked at it, but have no clue what to do about it. > 2) Flashplugin - when I tried to do the "make install" listed in the > FAQ #13 Section, > (http://www.openbsd.org/faq/faq13.html#flashplugin) > I ran into some package conflicts involving an rpm pkg (rpm vs > rpm-cpio). This should be fixed in -current (committed on october 4th), just update your ports tree or at least emulators/fedora. cheers, Nikolay
update: nagios 3.2.0
Hi,
attached is an update of Nagios to 3.2.0. Please test and comment.
cheers,
Nikolay
--
"It's all part of my Can't-Do approach to life." Wally
Index: Makefile
===
RCS file: /a/devel/cvs/ports/net/nagios/nagios/Makefile,v
retrieving revision 1.36
diff -p -u -r1.36 Makefile
--- Makefile7 Aug 2009 17:18:34 - 1.36
+++ Makefile10 Sep 2009 08:25:59 -
@@ -3,10 +3,10 @@
COMMENT-main= host and service monitor
COMMENT-web= cgis and webpages for nagios
-V= 3.0.6
+V= 3.2.0
DISTNAME= nagios-${V}
-PKGNAME-main= nagios-${V}p1
-PKGNAME-web= nagios-web-${V}p0
+PKGNAME-main= nagios-${V}
+PKGNAME-web= nagios-web-${V}
CATEGORIES=net
HOMEPAGE= http://www.nagios.org/
@@ -23,12 +23,10 @@ MULTI_PACKAGES= -main -web
MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=nagios/}
-MODULES= devel/gettext
-
BUILD_DEPENDS= :gd-*:graphics/gd
-LIB_DEPENDS= glib-2.0::devel/glib2 \
- ltdl.>=4::devel/libtool,-ltdl
-RUN_DEPENDS= :nagios-plugins-*:net/nagios/plugins
+LIB_DEPENDS= ltdl.>=4::devel/libtool,-ltdl
+RUN_DEPENDS= :nagios-plugins-*:net/nagios/plugins \
+ ::www/php5/core
CONFIGURE_STYLE=gnu
CONFIGURE_ARGS+=--datadir=/var/www/nagios \
@@ -67,11 +65,8 @@ WANTLIB-web=
LIB_DEPENDS-web=
RUN_DEPENDS-web=:nagios-${V}-chroot:net/nagios/nagios,chroot
.else
-WANTLIB-web= ${WANTLIB} fontconfig freetype jpeg z
-LIB_DEPENDS-web=${MODGETTEXT_LIB_DEPENDS} \
- glib-2.0::devel/glib2 \
- gd::graphics/gd \
- png::graphics/png
+WANTLIB-web= ${WANTLIB} fontconfig freetype jpeg png z
+LIB_DEPENDS-web=gd::graphics/gd
RUN_DEPENDS-web=:nagios-${V}-!chroot:net/nagios/nagios
.endif
@@ -81,7 +76,7 @@ NO_REGRESS= Yes
CFLAGS+= -D_DIRENT_HAVE_D_TYPE
-CONFIG_FILES0= cgi.cfg httpd.conf nagios.cfg resource.cfg
+CONFIG_FILES0= cgi.cfg httpd.conf mrtg.cfg nagios.cfg resource.cfg
CONFIG_FILES1= commands.cfg contacts.cfg localhost.cfg printer.cfg switch.cfg \
templates.cfg timeperiods.cfg windows.cfg
EXDIR= ${PREFIX}/share/examples/nagios
Index: distinfo
===
RCS file: /a/devel/cvs/ports/net/nagios/nagios/distinfo,v
retrieving revision 1.21
diff -p -u -r1.21 distinfo
--- distinfo24 Dec 2008 20:03:42 - 1.21
+++ distinfo10 Sep 2009 05:10:21 -
@@ -1,5 +1,5 @@
-MD5 (nagios-3.0.6.tar.gz) = kA4/QWT0sqGEhUIO6u/oEg==
-RMD160 (nagios-3.0.6.tar.gz) = HCsCa03FCV8kGdI5BCt1IA2xuV8=
-SHA1 (nagios-3.0.6.tar.gz) = 1r0gzcItK5Mfmtf5yzP/cdLLfXE=
-SHA256 (nagios-3.0.6.tar.gz) = vt6ywf+/dSXsGayEpmutYKGdKwVEy/BQpTv8NjwJuyI=
-SIZE (nagios-3.0.6.tar.gz) = 2735504
+MD5 (nagios-3.2.0.tar.gz) = NWYWfMYN3qrTTn0uJu1KWA==
+RMD160 (nagios-3.2.0.tar.gz) = IXfl68fKXM7nukBHKOiW0QBjsng=
+SHA1 (nagios-3.2.0.tar.gz) = Bwqjz6lmUZAEjThvRxzU03dID5k=
+SHA256 (nagios-3.2.0.tar.gz) = fL81ugMZ8k+ghZgsA4Q3xFFQA7J4Y8SJfchrGXpQpec=
+SIZE (nagios-3.2.0.tar.gz) = 3170967
cvs diff: Diffing patches
Index: patches/chroot-cgi_Makefile_in
===
RCS file: /a/devel/cvs/ports/net/nagios/nagios/patches/chroot-cgi_Makefile_in,v
retrieving revision 1.2
diff -p -u -r1.2 chroot-cgi_Makefile_in
--- patches/chroot-cgi_Makefile_in 18 Mar 2008 20:25:11 - 1.2
+++ patches/chroot-cgi_Makefile_in 10 Sep 2009 05:24:37 -
@@ -1,12 +1,11 @@
-$OpenBSD: chroot-cgi_Makefile_in,v 1.2 2008/03/18 20:25:11 sturm Exp $
cgi/Makefile.in.orig Mon Jan 24 12:31:08 2005
-+++ cgi/Makefile.inMon Jan 24 12:31:39 2005
-@@ -36,7 +36,7 @@ cfla...@cflags@ @DEFS@ -DNSCGI
+--- cgi/Makefile.in.orig Thu Sep 10 07:17:08 2009
cgi/Makefile.inThu Sep 10 07:17:33 2009
+@@ -36,7 +36,7 @@
#CFLAGS_DEBUG=-ggdb3 -g3
#CFLAGS+=$(CFLAGS_WARN) $(CFLAGS_DEBUG)
-ldfla...@ldflags@
+LDFLAGS=-static @LDFLAGS@
- li...@libs@ @GLIB_LIBS@
+ li...@libs@
CGIS=avail.cgi cmd.cgi config.cgi extinfo.cgi history.cgi notifications.cgi
outages.cgi showlog.cgi status.cgi statuswml.cgi summary.cgi tac.cgi
$(CGIEXTRAS)
Index: patches/patch-base_utils_c
===
RCS file: patches/patch-base_utils_c
diff -N patches/patch-base_utils_c
--- patches/patch-base_utils_c 10 Jan 2009 16:16:21 - 1.6
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,21 +0,0 @@
-$OpenBSD: patch-base_utils_c,v 1.6 2009/01/10 16:16:21 henning Exp $
base/utils.c.orig Sun Nov 30 18:22:58 2008
-+++ base/utils.c Sat Jan 10 17:14:26 2009
-@@ -610,8 +610,16 @@ int my_system(char *cmd,int timeout,int *early_timeout
- /* handle errors */
- if(bytes_read==-1){
- /* we encountered a recoverable error,
so try again */
-- if(errno==EINTR || errn
Re: update: opera 10
* frantisek holop [2009-09-05]: > Nikolay, have you considered an install message that the profile > structure has changed, and while upgrading from 9.* to 10.* is > surprisingly painless, one might consider backing up the profile > directory first? (for example i have no experience with opera's mail, > etc) Well I did that and consider it common sense before doing major upgrades. I use opera to read some rss feeds and haven't had any problem whatsoever with it. So, to answer your question, I think an install message like that suggested would only add to the noise of bulk updates. > one more thing, there is also a gcc3 build. is there a way to compare > it stability-wise to gcc4? Feel free to try it out, I haven't had any unusual experience with the gcc4 build yet but then, I only use it on an UP machine. cheers, Nikolay -- "It's all part of my Can't-Do approach to life." Wally
update: opera 10
Hi,
please find attached an update of the opera port to version 10. Please
test and comment.
cheers,
Nikolay
--
"It's all part of my Can't-Do approach to life." Wally
Index: Makefile
===
RCS file: /a/devel/cvs/ports/www/opera/Makefile,v
retrieving revision 1.59
diff -p -u -r1.59 Makefile
--- Makefile5 Mar 2009 05:38:17 - 1.59
+++ Makefile2 Sep 2009 19:13:19 -
@@ -1,12 +1,13 @@
# $OpenBSD: Makefile,v 1.59 2009/03/05 05:38:17 sturm Exp $
ONLY_FOR_ARCHS=i386
+SHARED_ONLY= Yes
COMMENT= fast and customizable WWW browser
-V= 9.64
-OPERA_BUILD= 2480
-DISTNAME= opera-${V}.gcc295-static-qt3.i386
+V= 10.00
+OPERA_BUILD= 4585
+DISTNAME= opera-${V}.gcc4-shared-qt3.i386
PKGNAME= opera-${V}
CATEGORIES=www
@@ -31,7 +32,7 @@ MASTER_SITES_OPERA= ftp://ftp.opera.com/
ftp://ftp.rge.com/pub/infosystems/opera/ \
http://www.rge.com/pub/infosystems/opera/
-MASTER_SITES= ${MASTER_SITES_OPERA:=linux/${V:S/.//}/final/en/i386/static/}
+MASTER_SITES= ${MASTER_SITES_OPERA:=linux/${V:S/.//}/final/en/i386/shared/}
EXTRACT_SUFX= .tar.bz2
RUN_DEPENDS= ::devel/desktop-file-utils \
@@ -40,7 +41,7 @@ RUN_DEPENDS= ::devel/desktop-file-utils
NO_BUILD= Yes
NO_REGRESS=Yes
-WRKDIST= ${WRKDIR}/opera-$V-${OPERA_BUILD}.gcc295-static-qt3.i386
+WRKDIST= ${WRKDIR}/opera-$V-${OPERA_BUILD}.gcc4-shared-qt3.i386
do-install:
cd ${WRKBUILD} && ./install.sh \
Index: distinfo
===
RCS file: /a/devel/cvs/ports/www/opera/distinfo,v
retrieving revision 1.45
diff -p -u -r1.45 distinfo
--- distinfo5 Mar 2009 05:38:17 - 1.45
+++ distinfo2 Sep 2009 18:46:45 -
@@ -1,5 +1,5 @@
-MD5 (opera-9.64.gcc295-static-qt3.i386.tar.bz2) = t5UtpYI4PkLEW/E6l833tg==
-RMD160 (opera-9.64.gcc295-static-qt3.i386.tar.bz2) =
bkPgoE0X2t9ZtnO8/0z/1deGC6E=
-SHA1 (opera-9.64.gcc295-static-qt3.i386.tar.bz2) = Cq0BFhmfz/Ip9BEgqqy8JkM/ioA=
-SHA256 (opera-9.64.gcc295-static-qt3.i386.tar.bz2) =
PanGOjLkMUFU01KwYagpOQM8NvauifCv2xpyiPpXGp8=
-SIZE (opera-9.64.gcc295-static-qt3.i386.tar.bz2) = 9061350
+MD5 (opera-10.00.gcc4-shared-qt3.i386.tar.bz2) = q/DpD2wGILGx+VuYYLMBiQ==
+RMD160 (opera-10.00.gcc4-shared-qt3.i386.tar.bz2) =
CrRIJX8G12IVALCWb7khvo7Mvgo=
+SHA1 (opera-10.00.gcc4-shared-qt3.i386.tar.bz2) = jF1uGSzraWbzudFVEBVncuXB2rY=
+SHA256 (opera-10.00.gcc4-shared-qt3.i386.tar.bz2) =
imQ2gDpi/Nfrjyx9weKoWSsI1+UkPiSMgCKH/CEjbEo=
+SIZE (opera-10.00.gcc4-shared-qt3.i386.tar.bz2) = 8454441
cvs diff: Diffing files
cvs diff: Diffing patches
Index: patches/patch-install_sh
===
RCS file: /a/devel/cvs/ports/www/opera/patches/patch-install_sh,v
retrieving revision 1.21
diff -p -u -r1.21 patch-install_sh
--- patches/patch-install_sh13 Jun 2008 23:01:36 - 1.21
+++ patches/patch-install_sh2 Sep 2009 18:04:32 -
@@ -1,6 +1,6 @@
$OpenBSD: patch-install_sh,v 1.21 2008/06/13 23:01:36 sturm Exp $
install.sh.origWed Jun 11 23:11:27 2008
-+++ install.sh Sat Jun 14 00:54:10 2008
+--- install.sh.origSun Aug 30 18:25:07 2009
install.sh Wed Sep 2 20:04:25 2009
@@ -870,7 +870,7 @@ generate_wrapper()
wrapper_contain="#!/bin/sh
@@ -20,7 +20,7 @@ $OpenBSD: patch-install_sh,v 1.21 2008/0
# Native Java enviroment
for d in \"\$OPERA_PERSONALDIR\" \"\$HOME/.opera\"
do if test -f \"\$d/javapath.txt\"
-@@ -1060,9 +1057,6 @@ if test \"\${OPERA_JAVA_DIR}\"; then
+@@ -1063,9 +1060,6 @@ if test \"\${OPERA_JAVA_DIR}\"; then
export LD_LIBRARY_PATH OPERA_JAVA_DIR
fi
cvs diff: Diffing pkg
Index: pkg/PLIST
===
RCS file: /a/devel/cvs/ports/www/opera/pkg/PLIST,v
retrieving revision 1.29
diff -p -u -r1.29 PLIST
--- pkg/PLIST 2 Dec 2008 19:54:16 - 1.29
+++ pkg/PLIST 2 Sep 2009 18:47:37 -
@@ -20,6 +20,9 @@ share/examples/opera/opera6rc.fixed
@sample /etc/opera6rc.fixed
share/icons/
share/icons/hicolor/
+share/icons/hicolor/128x128/
+share/icons/hicolor/128x128/apps/
+share/icons/hicolor/128x128/apps/opera.png
share/icons/hicolor/16x16/
share/icons/hicolor/16x16/apps/
share/icons/hicolor/16x16/apps/opera.png
@@ -32,249 +35,191 @@ share/icons/hicolor/32x32/apps/opera.png
share/icons/hicolor/48x48/
share/icons/hicolor/48x48/apps/
share/icons/hicolor/48x48/apps/opera.png
+share/icons/hicolor/scalable/
+share/icons/hicolor/scalable/apps/
+share/icons/hicolor/scalable/apps/opera.svg
share/opera/
+share/opera/defaults/
+share/opera/defaults/bookmarks.adr
+share/opera/defaults/feedreaders.ini
+share/opera/defaults/filehandler.ini
+share/opera/defaults/font.ini
+share/opera/defaults/license.txt
+share/opera/defaults/mailproviders.xml
+share/opera/defaults/pluginpath.ini
+share/opera/defau
Re: Problem doing a bulk package build
* and...@msu.edu [2009-08-30]: > vista /usr/ports/infrastructure/build ./dpb -b -e > ==> creating dependency file > /bin/sh: syntax error: `;' unexpected > /usr/bin/make all-dir-depends: Inappropriate ioctl for device at ./dpb line > 696. Make sure your whole tree is clean and mv mystuff out of the way, just to be sure. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: fedora4 patch
* frantisek holop [2009-07-30]: > i am playing with the idea of going forward and update to fedora 5. > and maybe slowly even higher... > > have you tried in the past and gave up or never even tried? > am i headed for the lion's den? I tried and failed, fedora 4 was the last version to support the old linux threads, afair. Nikolay
Re: fedora4 patch
* frantisek holop [2009-07-30]: > what follows is a patch for fedora4 to include qt and libmng so that > we can use the dynamic opera binaries. Committed, thanks. Next time be sure to use and diff against -current, please. Nikolay
Re: opera10 experimental build with bundled qt4 (and some linux emul questions)
* frantisek holop [2009-07-21]: > i am trying to do the same. if this proves to be usable on the long > run, the question is, how to make this automatised. could these two > other libraries be added to the fedora core makefile? or the Of course, that's the purpose of the fedora port. Nikolay
Re: compat_linux / fedora_base
* Austin Hook [2009-05-18]: > Just curious, which way would you fix it? I wouldn't care about it at all. In my opinion the fedora_base package just exists to make opera and one or two other programs run and that's it. If tools of the fedora_base package don't work, it doesn't matter, we don't need them. Linux compat is dying code anyways... cheers, Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: @sample / config file question
* frantisek holop [2009-03-19]: > some packages seem to have example config files and > ask the user explicitly to copy it to /etc and tailor it. > > some also put this file into /etc during pkg_add. > and i guess some do entirely differently :] As always, it depends. There are some programs out there, providing configuration examples which don't really make any sense as default config. In that case, I would advise the user to configure their software manually. Other programs provide usable defaults in their config files, making it sensible to install them to SYSCONFIDR by default. But that's just my opinion... Nikolay
Re: cannot fork - try again
* patrick keshishian [2009-03-03]: > I installed macppc snapshot from Feb 28 (dmesg attached so gmail > doesn't wrap lines) and I am slowly rebuilding ports and every once in > a while I see make abort with "cannot fork - try again". In OpenBSD we have very conservative settings for number of open files and number of processes. If ports dependencies get too deep or you run one of the bigger desktop systems, you have to increase these limits. Either change them in /etc/login.conf or use ulimit. Nikolay
Re: Get rid of warnings when YP is configured
* Ingo Schwarze [2008-12-27]: > If i understand correctly, local systrace.filter files override the > global systrace.filter in infrastructure/db. Ok, thanks for checking. I just committed your diff. Nikolay
Re: Get rid of warnings when YP is configured
* Ingo Schwarze [2008-12-26]: > I'm not really sure the patch is correct and doesn't break anything > else - though i currently don't see why any port build should be > allowed to bind(2) inet(4) sockets. We have two ports explicitly allowing binding to 0.0.0.0:0, devel/mico and x11/gnustep/base. How do these cope with your proposed change? Does denying bind(2) as suggested and removing their local filter break them? Nikolay
Re: update gnucash to 2.2.8
* Antoine Jacoutot [2008-12-22]: > Hmm... are you certain it needs rarian? Yes, gnucash-docs doesn't build w/o it. > I don't see any build/run depends on it and while you added a rarian goo > in PLIST the port does not install any omf files. gnucash-docs, which also RUN_DEPENDS on rarian, does install omf files. Re gnucash proper, I guess this is the part where my gnome-new-stuff-foo is missing. So I only need to call rarian when omf files are installed, thus I can remove the rarian calls from gnucash/pkg/PLIST? Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: update gnucash to 2.2.8
* Landry Breuil [2008-12-22]: > Had a quick look, what is the problem with libofx ? It's linked against the static sp library, which seems to break with loading shared objects aka the ofx module in gnucash. A quick look suggests exchanging our sp library with opensp, but I have more interesting problems like HBCI. :-) Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: update gnucash to 2.2.8
* Nikolay Sturm [2008-12-22]: > attached you find a new port of gnucash 2.2.8 to replace the existing > port plus a separate port of gnucash's GNOME documentation. I didn't > bother updating the old port as too much changed. Tested on i386 and > sparc64, but someone should verify the gnome/scrollkeeper/rarian bits. And this time with the tarball... Nikolay -- "It's all part of my Can't-Do approach to life." Wally gnucash.tar.gz Description: application/tar-gz
update gnucash to 2.2.8
Hi, attached you find a new port of gnucash 2.2.8 to replace the existing port plus a separate port of gnucash's GNOME documentation. I didn't bother updating the old port as too much changed. Tested on i386 and sparc64, but someone should verify the gnome/scrollkeeper/rarian bits. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Opera stability
Hi, as some people noticed, newer versions of opera tend to freeze quite often on OpenBSD, on MP systems it basically freezes right after startup. By accident I found a way to make opera virtually rock solid on my UP workstation and was able to surf a little on an MP machine. The trick is to use a webproxy like squid. I assume that the freezes happen during DNS resolution of hostnames, which are delegated to the webproxy now, thus stabilizing opera. It would be great if anyone knowledgeable enough could have a look at our emulation code and fix the real problem, though. :) cheers, Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: i386 ports breakage 2008-11-01
* Steven Mestdagh [2008-11-03]: > > net/py-libpcap swig related? > please try this update. Fixes the build on sparc64. Nikolay
Re: Good example of how to setup SQL schema? how to install webapp?
* Jeff R. Allen [2008-10-05]: > I am working on a port where one step is "make install-db", to setup > the schema in the (mysql) database so that the app will be ready to > run with nothing more than "pkg_add $portname". Is there an existing > port that handles this job in a way you'd recommend that I replicate? You have to deal with this in a post-install README. I really doubt you could setup a mysql database and package it in a sensible way. Nikolay
Re: Users and Groups in ports
* Aaron W. Hsu [2008-09-26]: > I am cleaning up a port to be imported into the tree, and I noticed > that even though there are appropriate @newuser and @newgroup lines > in the PLIST, these users are not created by the "make repackage" > command, or any other command for that matter. According to pkg_add(1) > the users will be created when the package is installed, but this > doesn't work so well when the package requires those users when > built, what can I do to fix this? You have to patch the build scripts. No port is permitted to add users. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: UPDATE: TWiki
* Jasper Lievisse Adriaanse [2008-09-23]: > anyone using TWiki who could test this update please? I use it at work, but I don't have a test setup handy. What I can say, though, is that our port sux big time and needs lots of tweaking before it actually works. Regardless, what I don't understand is the username dance throughout PLIST, I really doubt that's sensible. The other problem I have with your update (and the existing port), is the inclusion of perl modules. Why would you not just depend on our ports? Nikolay
Re: obsd43 building opera
* Pau [2008-09-20]: > cd /usr/ports/www/opera > make install clean=depends distclean > > It breaks. I have built it in the past on the same laptop but with > obsd42. I attach in the bottom the make messages. Either your system is totally messed up or your hardware is broken. Nikolay
Re: OpenBSD 4.3 (i386) port for KDEvelop 3.5.3?
* Rick Flores [2008-09-16]: > I was wondering if you guys are currently working on a port on OpenBSD > 4.3 (i386) for KDEvelop 3.5.3? Ports development only happens on -current. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: Update: ruby-gems 1.2.0
* Jeremy Evans [2008-09-09]:
> * /etc/gemrc (and windows equivalent) for global settings
I don't see anything in your diff making sure the port uses
${SYSCONFDIR} instead of hardcoded /etc. An example file would be nice
as well (if upstream provided one, haven't checked that).
Nikolay
Re: NEW: security/aespipe
* Federico G. Schwindt [2008-09-04]: > is this really needed? > you can achieve the same via "openssl enc [-e|-d] -aes128" for example. There are programs which use this tool. Furthermore, usefulness has never been considered when adding ports. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: UPDATE: mozilla-firefox-3.0
* Stefan Sperling [2008-07-19]: > EXA has other advantages over XAA as well. If you use the composite But it has disadvanteges as well, it breaks on my 82855GM by hanging the X server from time to time. Nikolay
Re: Random Opera Freezes, still?
* Aaron W. Hsu [2008-06-16]: > Does anyone have suggestions for improving the stability of Opera? Yes, opera 9.50 does freeze more often and harder than older opera releases did, but I haven't had the time to debug this issue yet. From the state of our linux emulation code, I'd expect this to be some threads issue though. If you want opera to be stable, improve compat linux to the state where we can update the fedora/base port to a more recent version. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: Errors when installing Fedora base in 4.3 -current
* Alexander.Pavlyutin [2008-06-07]:
> ===> Installing fedora_base-4.0p3 from /usr/ports/packages/i386/all/
> Adding fedora_base-4.0p3
> kern.emul.linux: 0 -> 1
> /emul/linux/sbin/ldconfig.bin[1]: syntax error: `(' unexpected
> /usr/local/emul/fedora/usr/bin/gtk-query-immodules-2.0-32: error while
> loading shared libraries: libXrandr.so.2: cannot open shared object file: No
> such file
> or directory
> /usr/local/emul/fedora/usr/bin/pango-querymodules-32: error while loading
> shared libraries: libXft.so.2: cannot open shared object file: No such file
> or direc
> tory
> --- fedora_base-4.0p3 ---
> system(/bin/sh, -c,
> /usr/local/emul/fedora/usr/bin/gtk-query-immodules-2.0-32 >
> /usr/local/emul/fedora/etc/gtk-2.0/gtk.immodules) failed: exit(127)
> system(/bin/sh, -c, /usr/local/emul/fedora/usr/bin/pango-querymodules-32 >
> /usr/local/emul/fedora/etc/pango/i386-redhat-linux-gnu/pango.modules)
> failed: exit
> (127)
This is not reproducable here and your output looks weird, I don't have
that "Adding..." line. Please show your mk.conf and the actual command
line you used. Then try adding the package with pkg_add and show the
output.
Nikolay
Re: Question about "make update"
* Alexander.Pavlyutin [2008-06-03]: > For example: > I install esound(@option manual-installation in +CONTENTS) whitch > autoinstall libaudiofile (no @option manual-installation in +CONTENTS). > If I update libaudiofile from ports with "make update" I get @option > manual-installation in +CONTENTS > How I can update libaudiofile without getting @option manual-installation in > +CONTENTS? The ports tree is not intended for use by endusers, they are supposed to use pkg_add (there are exceptions, I know). So, just run pkg_add -ui and you shouldn't get that 'manual-installation'. If you do, it is either a bug in the pkg tools or the related feature is not finished yet. cheers, Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: pkg_create tweak
* Marc Espie [2008-04-06]: > This adds some supplementary option to pkg_create to expose the > substitution mechanism. The idea being that, sometimes, in ports, we > would like to have the exact same substitutions that are available to > pkg creation. Having read your mail and even the man page bits of your diff, I have no clue at all, what this would accomplish. Your examples don't help either. Could you explain the issue in a little more verbosity, please? cheers, Nikolay
Re: libiconv vs gperf
* Kenneth R Westerback [2008-04-03]: > When I got back it had crashed because the libiconv build could not > find gperf. gperf had been built and installed. However the PATH for > root did not include /usr/local/*, and the invocation of gperf > apparently did not supply the appropriate path. I don't know if this This is a bug in the port, I'll take care of it. Thanks for the report, Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: new: thinkingrock
* Nikolay Sturm [2008-03-18]: > Thinking Rock is a free software application for collecting and > processing your thoughts following the GTD methodology. Updated port attached: - fixed PERMIT_*_CDROM, thanks to Andreas Bihlmaier - use javaPathHelper as suggested by kurt@ Feedback welcome, especially with other jdks/jres than 1.5. > The archive contains a patched java.port.mk which will be committed > soon. Feedback welcome. This is already committed. cheers, Nikolay -- "It's all part of my Can't-Do approach to life." Wally thinkingrock.tar.gz Description: application/tar-gz
Re: new: thinkingrock
* Andreas Bihlmaier [2008-03-18]: > Product Version = ThinkingRock 2.0 Epsilon > Operating System= OpenBSD version 4.2 running on i386 Ups, this confused me. > Java; VM; Vendor; Home = 1.5.0_13-p7; Java HotSpot(TM) Client VM > 1.5.0_13-p7-root_28_jan_2008_09_28; Sun Microsystems Inc.; > /usr/local/jdk-1.5.0/jre I have only tried using thinkingrock with 1.5.0.14 and never had a problem. Maybe updating the jdk helps? > tr.extract.clean [1.0 070616] > [WinSys.PersistenceManager.getTopComponentForID] Problem when deserializing > TopComponent for tcID:'OverviewTopComponent'. Reason: Top component > OverviewTopComponent could not be located or loaded from Components folder. > Cannot find Overview component. It will not be located properly in the window > system. This is a warning only and shouldn't be a problem. I get it as well on a clean system but after that, thinkingrock starts fine. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: new: thinkingrock
* Nikolay Sturm [2008-03-18]: > > To your port: > > - Does CDDL really permit CDROM? (if yes then why doesn't devel/netbeans) > I'll verify this. That license is too hard for me to understand, I'll go with the restrictions netbeans uses. Thanks for the hint. > > - crashes for me on startup: > Works for me, I'll have a look at this as well. You don't run -current and I don't have any 4.2 machine, so I cannot help with this. If anyone sees it fail on 4.3 or -current, let me know. cheers, Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: new: thinkingrock
* Andreas Bihlmaier [2008-03-18]: > I have been working on a port of thinking rock as well, but I also > wanted to split off the included netbeans since we have a port of > netbeans already, but I was not able to. That's trivial but I was warned that might introduce other problems, so I kept it. > To your port: > - Does CDDL really permit CDROM? (if yes then why doesn't devel/netbeans) I'll verify this. > - crashes for me on startup: Works for me, I'll have a look at this as well. Nikolay
new: thinkingrock
Hi, attached is a port of thinkingrock: Thinking Rock is a free software application for collecting and processing your thoughts following the GTD methodology. The archive contains a patched java.port.mk which will be committed soon. Feedback welcome. Nikolay -- "It's all part of my Can't-Do approach to life." Wally thinkingrock.tar.gz Description: application/tar-gz
Re: split www/zope and www/plone into multiple versions
* Marc Winiger [2008-02-14]: > Nikolay, do you see the problem we have? Do you have any objections to > the way we solved it? Makes sense to me, I just wanted to be sure there was a reason to have multiple versions in tree. Nikolay
Re: split www/zope and www/plone into multiple versions
* Marc Winiger [2008-02-08]: > The versions of zope and plone in the -current cvs tree are broken. They > don't work togheter. So we decided to split the ports into multiple > versions, that the user can choose the version he needs. Why would it make sense to have multiple versions of both in-tree? Nikolay
Re: Switch www/webalizer to webalizer xtended
* Marc Balmer [2008-01-30]: > this switches www/webalizer to webalizer xtended, a fork of webalizer > that adds IPv6 support and detailed HTTP 404 statistics. Also set > MAINTAINER to the ports mailing list. What does the old maintainer say about this? Nikolay
Re: LDFLAGS is missing from bsd.port.mk
* Brad [2008-01-13]: > I feel the same way. bsd.port.mk(5) and examples in the tree are more > than good enough. Some people take the template way too seriously, to > the point of even putting variables in the same order as the template > and other silliness. I don't only do that, I actually think it does have a benefit. From my experience it is much easier reviewing a port that follows some pre-defined structure than having none whatsoever. We have KNF in src and IMHO Makefilte.template is a good start at PNF (Ports Normal Form). I use it for virtually every new port I start. Nikolay
Re: Installing Example Files?
* J.C. Roberts [2007-12-10]:
> pre-fake:
Use what everyone else uses, post-install for adding extra files.
> Which is considered more correct (or is preferred) ?
>
> EXDIR= ${PREFIX}/share/examples/${PKGNAME}
> or
> EXDIR= ${PREFIX}/share/examples/${BASE_PKGPATH}
Dito, use ${PREFIX}/share/examples/$portname_without_version.
Looking at other ports is the first step to answer your questions when
porting.
Nikolay
--
"It's all part of my Can't-Do approach to life." Wally
Re: Update sticks on audio/workman
* Peter Hessler [2007-12-09]: > Does pkg_add -u handle ports that can't be distributed as binaries > (i.e. updates to Java)? Last I checked, that wasn't available. Not yet. It is planned to have support for sth called src-URL, which points to your ports tree, where pkg_add will find updates to java and other undistributable packages. Nobody (except for espie@ maybe) knows when this feature will be committed, though. :) Nikolay
Re: Update sticks on audio/workman
* Jason Dixon [2007-12-09]: > Sorry, I should expand on that. What I mean is that it seems odd that > it dives into BUILD_DEPENDS before checking to see if it's installed. > Only after it resolves the state of its dependencies does it bother to > see if they (or itself) is even needed. The ports dir is designed with developer's needs in mind, therefor it might behave a little strange in certain situations. It looks to me as if make update was not supposed to be used recursively as there is a much better mechanism: pkg_add -u. Just use that one and be happy. Nikolay
Re: Firefox 2.0.0.10? For OpenBSD 4.2-CURRENT only??
* Edd Barrett [2007-11-29]: > instead you should urge the project to pay more attention to the > stable branch, if thats how you feel. You cannot urge the project, that's not how we work. You have to find a developer and convince him to take responsibility. I tried it several times and failed, which is easily understandable: -stable maintenance doesn't give anything back to the developer who does it, it's just work and it never ends. > Backporting is quite a simple process. Usually you can try the one Certainly not! Backporting is often enough an extremely complex process, you have to make sure not to break compatibility, sometimes you have to dig through source code to find that one security fix, you are looking for but which isn't marked in any sensible way. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: can't install nagios
* PowerBSD [2007-11-26]: > I know the reason: > > I only use make install clean That's not the reason, the reason is, that you are in the wrong directory. Nikolay
Re: can't install nagios
* PowerBSD [2007-11-26]: > when I build the port nagios,at last with an error: And which line of the output is it that you don't understand? > ===> net/nagios/nagios > ===> net/nagios/nagios,chroot > ===> Installing nagios-2.10-chroot from /usr/ports/packages/i386/all/ > Can't install nagios-2.10-chroot because of conflicts (nagios-2.10) > /usr/sbin/pkg_add: nagios-2.10-chroot:Fatal error > *** Error code 1 > > Stop in /usr/ports/net/nagios/nagios (line 1403 of > /usr/ports/infrastructure/mk/bsd.port.mk). > ===> Exiting net/nagios/nagios,chroot with an error > *** Error code 1 > > Stop in /usr/ports/net/nagios (line 124 of > /usr/ports/infrastructure/mk/bsd.port.subdir.mk). > -- "It's all part of my Can't-Do approach to life." Wally
Re: Ion3 port is obsolete
* Siju George [2007-11-17]: > If OpenBSD changed its license and put in restrictions you would have > no other way but to distribute the old one. > > See it is not the OpenBSD people who tarnishes you. It is your own > license restrictions that are working against you. Could you move this stupid flame war elsewhere, please? Or even better, just stop it. cheers, Nikolay
Re: Python versions, Zope and, Python modules
* Ingo Schwarze [2007-11-13]: > Consistent naming schemes are nice to have. Yes, but we don't have a consistent base to start with... > On the other hand, regarding FLAVORS, aren't they intended for > packages having common basic technology (which is not the case here as > we are using different platforms to start from) and different optional > functionality (which is also not the case here, py-foo will usually do > the same no matter which version of Python is involved). Besides, in > my understanding, choosing one and only one of several available > FLAVORS at install time is central to the whole concept of FLAVORS, > not only technically, but also logically. In my understanding, FLAVORS just differentiate packages built from the same source with different compile time options (i.e. you have to build them twice), just as is the case here: one module source compiled with different versions of python. That's the reason why I want this problem solved with FLAVORS. SUBPACKAGES on the other hand are just used to split files and simplify dependencies (i.e. one build to package them all ;). Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: Python versions, Zope and, Python modules
* Marc Winiger [2007-11-12]: > Python can co-exist in different versions at the same time. Why should > that not be possible for site-packages with our ports tree? Which is exactly the question Espie has to answer, but is currently too busy to do. So why don't we all just wait and maybe poke a little in his general direction. ;) Nikolay
-stable ports maintenance
Hi, as you might have noticed, -stable ports have not been properly updated in the last few months. Due to lack of resources, especially a responsible maintainer, you cannot expect any updates to -stable for the foreseeable future. Although some updates might happen, -stable should be considered unmaintained. It is sad having to make this announcement, but I thought our users should know about the status quo. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: mozilla is backwards (as usual)
* Ian Darwin [2007-10-26]: > I'm with Kurt, not changing the default behaviour for a cross-platform app. me too > So would this do it, or is this document getting too long? README.OpenBSD should talk about OpenBSD specifics, this is just a general configuration issue that does not need to be documented in our files. Nikolay
Re: [Resubmit] swish-e
* Mike Erdely [2007-09-10]: > Simon@ worked with me to add the -perl subpackage. Because the libs > are handled so late in the build process (during do-install), the > -perl subpackage is built in post-install. Since *-install runs as > root, the files built inside w-swish-e-2.4.5/swish-e-2.4.5/perl are > owned by root. That makes 'make clean' puke when running with > SUDO=sudo as a normal user. Don't force a SUBPACKAGE if the port is evil, just make it a FLAVOR. Nikolay
Re: Repost: updates waiting
* [EMAIL PROTECTED] [2007-09-04]: > > cfengine-2.1.22: > > > > http://openbsd.rutgers.edu/cfengine-2.1.22.diff I'll take care of this. Nikolay
sparc64 snapshot packages
Hi, the sparc64 package fuckup is being fixed as I write this mail, expect mirrors to carry real sparc64 packages soon. Thanks for the report! Nikolay -- "It's all part of my Can't-Do approach to life." Wally
sparc64 packages
Hi, I just uploaded a new set of sparc64 packages. Please test and report any problems you see. cheers, Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: Tor 1.2.16 - Security hole
* Peter Thoenen [2007-08-08]: > Tor 1.1.x has BEEN DEPRECIATED from before the time 4.1 STABLE was Tor 1.2 only came out around the release of 4.1 and no update was marked a security update, so there was no reason to update the -stable ports. We have our policies and we do have them for a good reason. > two week per the developers. The developers announced all users > should update immediately yet still not seeing this port updated in > stable when I csup. So what? Where does the OpenBSD project guarantee you security updates or any service whatsoever within any fixed timeframe? -current was updated within a week and 4.1-stable shortly thereafter. 4.0 will have to wait a little longer. > Can you (Rui) update this port finally as it would count as a security > update or you just going to hang out and continue to be a subpar > maintainer. How about you smart ass send diffs instead of being a subpar user? And btw, we have a process for dealing with unresponsive maintainers. It's documented in the archives, but that of course, would mean a little work on your side finding out... > If you don't want to maintain your own port then let me know and I or > somebody else can do it but this is ridiculous. You Since 4.1 the tor port was updated within days of the release of a new version. Get your facts straight. > Sorry to be an ass Rui but with maintenance comes responsibility. You have a responsibilty as well, mr consumer, care to take it on? Nikolay
Re: var/run policy?
* Todd T. Fries [2007-06-19]: > Shouldn't the ports in question either via pkg/MESSAGE or somesuch > instruct the user on proper startup of the software, including > creation of the necessary directory(s) under /var/run so /etc/rc.local > is properly populated? This is definately the right thing to do. Using @unexec on those files and directories doesn't hurt, but for simplicity's sake I wouldn't enforce that. The same goes for creating directories from pkg_add, it's nice as you can just start up the daemon but it's not strictly needed. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
4.0-stable php5 update needs testing
* Nikolay Sturm [2007-06-10]:
> attached is a security update for php5 on 4.1-stable. Please report any
> breakage you see. If I don't get any negative feedback, this will be
> committed next friday.
>
> A diff for 4.0 should follow soon.
Here's the diff.
Nikolay
Index: Makefile.inc
===
RCS file: /cvs/ports/www/php5/Makefile.inc,v
retrieving revision 1.11
diff -p -u -r1.11 Makefile.inc
--- Makefile.inc18 Jul 2006 21:44:40 - 1.11
+++ Makefile.inc10 Jun 2007 18:06:18 -
@@ -79,3 +79,6 @@ PATCH_DIST_STRIP= -p1
.else
SUPDISTFILES= hardening-patch-${V}-${HARDENED_V}.patch.gz:1
.endif
+
+post-extract:
+ @find ${WRKSRC} -name \*.orig -exec rm {} \;
Index: core/Makefile
===
RCS file: /cvs/ports/www/php5/core/Makefile,v
retrieving revision 1.17
diff -p -u -r1.17 Makefile
--- core/Makefile 4 Aug 2006 00:10:42 - 1.17
+++ core/Makefile 10 Jun 2007 18:06:18 -
@@ -5,8 +5,8 @@ SUBPACKAGE?=
COMMENT= "server-side HTML-embedded scripting language"
COMMENT-pear= "base classes for common PHP tasks"
-PKGNAME= php5-core-${V}p1
-FULLPKGNAME-pear= php5-pear-${V}p0
+PKGNAME= php5-core-${V}p2
+FULLPKGNAME-pear= php5-pear-${V}p1
DISTFILES= php-${V}.tar.gz \
pear-${PEAR_V}.tar.gz:0
Index: extensions/Makefile
===
RCS file: /cvs/ports/www/php5/extensions/Makefile,v
retrieving revision 1.23
diff -p -u -r1.23 Makefile
--- extensions/Makefile 4 Aug 2006 00:10:42 - 1.23
+++ extensions/Makefile 10 Jun 2007 18:06:18 -
@@ -1,6 +1,6 @@
# $OpenBSD: Makefile,v 1.23 2006/08/04 00:10:42 espie Exp $
-FULLPKGNAME= php5-extensions-${V}
+FULLPKGNAME= php5-extensions-${V}p0
COMMENT= "informational package about PHP5 extensions"
MULTI_PACKAGES=
@@ -106,11 +106,11 @@ CONFIGURE_ARGS+= --with-gd=shared ${GRAP
. if ${FLAVOR:L:Mno_x11}
CONFIGURE_ARGS+= --without-xpm-dir
-FULLPKGNAME-gd=php5-gd-${V}-no_x11
+FULLPKGNAME-gd=php5-gd-${V}-no_x11p0
. else
USE_X11= yes
CONFIGURE_ARGS+= --with-xpm-dir=${X11BASE}
-FULLPKGNAME-gd=php5-gd-${V}
+FULLPKGNAME-gd=php5-gd-${V}p0
. if !defined(PACKAGING) || ${SUBPACKAGE} == "-gd"
GD_WANTLIB+= X11 Xpm
. endif
@@ -139,7 +139,7 @@ CONFIGURE_ARGS+=--with-imap=shared,${LO
--with-kerberos=/usr --with-imap-ssl
IMAP_DEPENDS= c-client.>=3::mail/c-client
IMAP_WANTLIB= gssapi crypto krb5 ssl
-FULLPKGNAME-imap= php5-imap-${V}
+FULLPKGNAME-imap= php5-imap-${V}p0
.endif
# ldap
@@ -260,7 +260,7 @@ MULTI_PACKAGES+=-soap
COMMENT-soap= "SOAP functions for php5"
CONFIGURE_ARGS+= --enable-soap=shared
SOAP_WANTLIB= xml2 z iconv
-FULLPKGNAME-soap= php5-soap-${V}
+FULLPKGNAME-soap= php5-soap-${V}p0
.endif
# snmp
@@ -318,12 +318,12 @@ COMMENT-xsl= "XSL functions for php5"
CONFIGURE_ARGS+= --with-xsl=shared --enable-dom
XSL_DEPENDS= xslt.>=3,exslt::textproc/libxslt
XSL_WANTLIB= iconv xml2 z
-FULLPKGNAME-xsl= php5-xsl-${V}
+FULLPKGNAME-xsl= php5-xsl-${V}p0
.endif
.for i in ${MULTI_PACKAGES}
. if !defined(FULLPKGNAME${i})
-FULLPKGNAME${i}= php5${i}-${V}
+FULLPKGNAME${i}= php5${i}-${V}p0
. endif
. if ${FLAVOR:L:Mhardened}
FULLPKGNAME${i}:= ${FULLPKGNAME${i}}-hardened
Index: extensions/patches/patch-ext_gd_config_m4
===
RCS file: /cvs/ports/www/php5/extensions/patches/patch-ext_gd_config_m4,v
retrieving revision 1.4
diff -p -u -r1.4 patch-ext_gd_config_m4
--- extensions/patches/patch-ext_gd_config_m4 9 May 2006 14:15:40 -
1.4
+++ extensions/patches/patch-ext_gd_config_m4 10 Jun 2007 18:06:18 -
@@ -1,7 +1,7 @@
$OpenBSD: patch-ext_gd_config_m4,v 1.4 2006/05/09 14:15:40 robert Exp $
ext/gd/config.m4.orig Fri Apr 28 00:56:27 2006
-+++ ext/gd/config.m4 Fri Apr 28 00:57:11 2006
-@@ -87,13 +87,13 @@
+--- ext/gd/config.m4.orig Tue Nov 29 19:25:59 2005
ext/gd/config.m4 Sun Jun 10 19:51:29 2007
+@@ -87,13 +87,13 @@ AC_DEFUN([PHP_GD_PNG],[
AC_MSG_ERROR([PNG support requires ZLIB. Use --with-zlib-dir=])
fi
Index: extensions/patches/patch-ext_gd_gdttf_c
===
RCS file: /cvs/ports/www/php5/extensions/patches/patch-ext_gd_gdttf_c,v
retrieving revision 1.1.1.1
diff -p -u -r1.1.1.1 patch-ext_gd_gdttf_c
--- extensions/patches/patch-ext_gd_gdttf_c 2 Oct 2004 11:32:35 -
1.1.1.1
+++ extensions/patches/patch-ext_gd_gdttf_c
4.1-stable php5 update needs testing
Hi,
attached is a security update for php5 on 4.1-stable. Please report any
breakage you see. If I don't get any negative feedback, this will be
committed next friday.
A diff for 4.0 should follow soon.
Nikolay
--
"It's all part of my Can't-Do approach to life." Wally
Index: Makefile.inc
===
RCS file: /cvs/ports/www/php5/Makefile.inc,v
retrieving revision 1.13
diff -p -u -r1.13 Makefile.inc
--- Makefile.inc24 Oct 2006 09:30:25 - 1.13
+++ Makefile.inc10 Jun 2007 16:44:37 -
@@ -70,3 +70,6 @@ PATCH_LIST= ${.CURDIR}/../patches/patch
patch-*
PATCHFILES=suhosin-patch-${V}-${SUHOSIN_V}.patch.gz:1
PATCH_DIST_STRIP= -p1
+
+post-extract:
+ @find ${WRKSRC} -name \*.orig -exec rm {} \;
Index: core/Makefile
===
RCS file: /cvs/ports/www/php5/core/Makefile,v
retrieving revision 1.19
diff -p -u -r1.19 Makefile
--- core/Makefile 25 Nov 2006 11:09:04 - 1.19
+++ core/Makefile 10 Jun 2007 16:44:37 -
@@ -4,8 +4,8 @@ MULTI_PACKAGES= -main -pear
COMMENT-main= "server-side HTML-embedded scripting language"
COMMENT-pear= "base classes for common PHP tasks"
-PKGNAME-main= php5-core-${V}p0
-FULLPKGNAME-pear= php5-pear-${V}p0
+PKGNAME-main= php5-core-${V}p1
+FULLPKGNAME-pear= php5-pear-${V}p1
DISTFILES= php-${V}.tar.gz \
pear-${PEAR_V}.tar.gz:0
@@ -50,7 +50,6 @@ PREFIX-pear= ${CHROOT_DIR}
RUN_DEPENDS-pear= :php5-core-${V}:www/php5/core
WANTLIB-pear=
LIB_DEPENDS-pear=
-
pre-fake:
${INSTALL_DATA_DIR} ${PREFIX}/${APACHE_MODULE_SUBDIR}
Index: extensions/Makefile
===
RCS file: /cvs/ports/www/php5/extensions/Makefile,v
retrieving revision 1.32
diff -p -u -r1.32 Makefile
--- extensions/Makefile 9 Feb 2007 13:48:20 - 1.32
+++ extensions/Makefile 10 Jun 2007 16:44:37 -
@@ -1,6 +1,6 @@
# $OpenBSD: Makefile,v 1.32 2007/02/09 13:48:20 bernd Exp $
-FULLPKGNAME-main= php5-extensions-${V}p0
+FULLPKGNAME-main= php5-extensions-${V}p1
COMMENT-main= "informational package about PHP5 extensions"
MULTI_PACKAGES=-main
@@ -48,7 +48,7 @@ COMMENT-curl= "curl URL library extensi
CONFIGURE_ARGS+= --with-curl=shared,${LOCALBASE}
LIB_DEPENDS-curl= curl.>=2::net/curl
WANTLIB-curl= ${WANTLIB} crypto ssl z
-FULLPKGNAME-curl= php5-curl-${V}p2
+FULLPKGNAME-curl= php5-curl-${V}p3
.endif
# dba
@@ -101,11 +101,11 @@ CONFIGURE_ARGS+= --with-gd=shared ${GRAP
. if ${FLAVOR:L:Mno_x11}
CONFIGURE_ARGS+= --without-xpm-dir --without-freetype-dir
-FULLPKGNAME-gd=php5-gd-${V}-no_x11p2
+FULLPKGNAME-gd=php5-gd-${V}-no_x11p3
. else
USE_X11= yes
CONFIGURE_ARGS+= --with-xpm-dir=${X11BASE}
-FULLPKGNAME-gd=php5-gd-${V}p1
+FULLPKGNAME-gd=php5-gd-${V}p2
WANTLIB-gd+= X11 Xpm freetype
. endif
.endif
@@ -290,7 +290,7 @@ MULTI_PACKAGES+=-sybase_ct
COMMENT-sybase_ct= "sybase database access extensions for php5"
CONFIGURE_ARGS+=--with-sybase-ct=shared,${LOCALBASE}
LIB_DEPENDS-sybase_ct= ct::databases/freetds
-FULLPKGNAME-sybase_ct= php5-sybase_ct-${V}p2
+FULLPKGNAME-sybase_ct= php5-sybase_ct-${V}p3
.endif
# xml-rpc
@@ -318,7 +318,7 @@ WANTLIB-xsl=${WANTLIB} iconv xml2 z
.endif
.for i in ${MULTI_PACKAGES}
-FULLPKGNAME${i} ?= php5${i}-${V}p1
+FULLPKGNAME${i} ?= php5${i}-${V}p2
RUN_DEPENDS${i}= :php5-core-${V}:www/php5/core
MODULE_NAME${i}= ${i:S/-//g}
.endfor
Index: extensions/patches/patch-ext_gd_config_m4
===
RCS file: /cvs/ports/www/php5/extensions/patches/patch-ext_gd_config_m4,v
retrieving revision 1.4
diff -p -u -r1.4 patch-ext_gd_config_m4
--- extensions/patches/patch-ext_gd_config_m4 9 May 2006 14:15:40 -
1.4
+++ extensions/patches/patch-ext_gd_config_m4 10 Jun 2007 16:44:37 -
@@ -1,7 +1,7 @@
$OpenBSD: patch-ext_gd_config_m4,v 1.4 2006/05/09 14:15:40 robert Exp $
ext/gd/config.m4.orig Fri Apr 28 00:56:27 2006
-+++ ext/gd/config.m4 Fri Apr 28 00:57:11 2006
-@@ -87,13 +87,13 @@
+--- ext/gd/config.m4.orig Tue Nov 29 19:25:59 2005
ext/gd/config.m4 Sun Jun 10 18:10:52 2007
+@@ -87,13 +87,13 @@ AC_DEFUN([PHP_GD_PNG],[
AC_MSG_ERROR([PNG support requires ZLIB. Use --with-zlib-dir=])
fi
Index: extensions/patches/patch-ext_gd_gdttf_c
===
RCS file: /cvs/ports/www/php5/extensions/patches/patch-ext_gd_gdttf_c,v
retrieving revision 1.1.1.1
diff -p -u -r1.1.1.1 patch-ext_gd_gdttf_c
--- extensions/patches/patch-ext_gd_gdttf_c 2 Oct 2004 11:32:35 -
1.1.1.1
+++ ex
Re: Checksum mismatch (4.1-current)
* Stephan A. Rickauer [2007-05-16]: > I seem to have messed up my ports-tree/machine, since all attempts to > build any port from -current fails with this (example for subversion): > > cksum: -: no properly formatted checksum lines found I assume your system and ports tree are out of sync. Nikolay
Re: Systrace - 'systrace: intercept_get_string: ioctl: Invalid
* Otto Moerbeek [2007-05-14]: > > Has the patch been committed to CVS? > No, I just committed it. Nikolay
Re: NEW: textproc/gsed
* Paul de Weerd [2007-05-15]: > The problem is in the 'we should have everything + the kitchen sink in > the portstree'-attitude. Ports/packages should add functionality. In > the past, software has been removed from the portstree because the > functionality provided by them had been added to the base OS. There's > no need to add ports that have no additional value other than "hey, > that's one more package we've ported to OpenBSD". This is your opinion, the ports tree works differently. Nikolay
Re: gettext and expat
* Todd T. Fries [2007-05-10]: > There is an expat in src/. What are the plans for that? Would solve > a lot of `must install xbase in order to get expat' complaints. The > counter is that one already installs xbase to get libs. My understanding is, that it won't be activated until it's audited, which has not happened for about 1.5 years now. Nikolay
Re: gettext and expat
* Ketelaars, Björn [2007-05-10]: > I really dislike the idea of installing binaries/libraries, which are > not going to be used for a certain task/application. It's the same > with the compiler; if you do not use it, do not install it. Do you use tcpdump on your webserver or sudo or ospfd or bgpd or ...? You install all kinds of crap on your webserver that you will never use. > I am wondering what the reason is for abandoning expat from ports - > Xenocara maybe? Yes, xenocara provides its own libexpat and we ran into trouble with both, xenocara's libexpat and the expat package being installed. Nikolay
Re: gettext and expat
* Damien Miller [2007-05-10]: > I'd be more happy with this argument if installing the necessary libraries > didn't also install setuid X11 binaries. It doesn't have to be this way, I suggested an additional x-package just containing libraries and support files (maybe even only freetype and expat) before. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: Systrace - 'systrace: intercept_get_string: ioctl: Invalid
* Otto Moerbeek [2007-05-09]: > I see systrace things during make: > > systrace: deny user: otto, prog: /usr/local/bin/fontforge, pid: > 32458(0)[25445], > policy: /usr/bin/env, filters: 193, syscall: native-fswrite(136), > filename: /lilypond-2.8.8_writes_to_HOME/.PfaEdits legitimate (i.e. systrace is behaving correctly) > [the build continues] > > and during make fake: > > /usr/ports/print/lilypond/w-lilypond-2.8.8/fake-i386/usr/local/share/lilypond/2.8.8/fonts/type1: > failed to write cachesystrace: deny user: root, prog: > /usr/X11R6/bin/fc-cache, pid: 2897(0)[24786], policy: /usr/bin/env, > filters: 193, syscall: native-fswrite(10), filename: > /var/cache/fontconfig/9d5904fd1710b46775398a2ce65f03cc-x86.cache-2 > gmake[1]: *** [install-fc-cache] Error 5 > gmake[1]: Leaving directory > `/usr/ports/print/lilypond/w-lilypond-2.8.8/lilypond-2.8.8/mf' > gmake: *** [install] Error 2 > *** Error code 2 and this as well If this is all there is, then the port is just broken. But what I got from the discussion so far, is that it fails randomly with the intercept_get_string error. I started a lilypond build myself a couple of minutes ago. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: Systrace - 'systrace: intercept_get_string: ioctl: Invalid argument' during make fake
* Edd Barrett [2007-05-08]: > >I am still porting texlive. My problem now is that systrace is > >inconsistently chucking up this error: > > > >systrace: intercept_get_string: ioctl: Invalid argument Which is, to the best of my knowledge, harmless. > >Which kills the build. It does not occur every time. So you say your build works sometimes and sometimes not? > >The whole transcript of a failing "make fake" (which is where this > >dies) is here: Your logfile has this to say: fmtutil: Error! Not all formats have been built successfully. Visit the log files in directory /usr/ports/print/texlive/base/w-texlive_base-2007/fake-sparc64/var/web2c for details. So what is up with those logs? Nikolay
EOL OpenBSD 3.9 ports
With the release of OpenBSD 4.1, there will be no more updates to the OpenBSD 3.9 stable ports tree. We will start updating the OpenBSD 4.1 stable ports tree today. cheers, Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: Fix installation of ImageMagic, lib_depends+=expat..
* Mikolaj Kucharski [2007-04-27]: > ImageMagic fails to install even when expat port is installed. This is a well known issue and discussed here before (it was with gd, afair). With the bump of libexpat in xenocara, ImageMagic and the other ports that had problems with expat should install again. We will soon start getting rid of the expat port, so everyone install their xbase. :) Nikolay
Re: Embedding MODPY_VERSION in PKGNAME
* Damien Miller [2007-04-26]: > At present it is not possible to install Python modules for more than > one version of Python concurrently. A simple way around this would be > to embed MODPY_VERSION (or some modified form thereof) into PKGNAME. > For example "py-flowd-0.9-py2.4" or something like "py24-flowd-0.9". This will result in tons of conflicts, complicating issues for porters of python modules. You should provide some very strong reason why this extension would be useful and worth the hassle. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
new: grub
Hi, attached is a first shot at a grub port. It works for me with the chainloader, personally I don't care for native kernel loading. grub over serial works as well, which is all I need. :) grub is i386-only. Feedback welcome, Nikolay -- "It's all part of my Can't-Do approach to life." Wally grub.tar.gz Description: application/tar-gz
Re: update: audio/libgpod: 0.99.8p1 -> 0.99.8p2
* Chris Kuethe [2007-03-15]: > This makes gtkpod compile with the new libgpod Works for me, ok sturm. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: update: audio/libgpod: 0.4.0 -> 0.4.2
* Chris Kuethe [2007-03-15]: > Works on i386 talking to my 5th gen ipod gtkpod doesn't compile with this lib: file.c: In function `copy_new_info': file.c:854: error: structure has no member named `unk208' file.c:854: error: structure has no member named `unk208' file.c: In function `get_track_info_from_file': file.c:1073: error: structure has no member named `unk208' file.c:1082: error: structure has no member named `unk208' file.c:1090: error: structure has no member named `unk208' file.c:1102: error: structure has no member named `unk208' file.c:1113: error: structure has no member named `unk208' Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: obsd4.0 - nikto-1.3.5
* Zahid Bukhari [2007-02-18]: > The checksums seem to be off in distinfo. fixed in -current and -stable, thanks Nikolay
Re: update: www/privoxy: 3.0.3 -> 3.0.6
* Chris Kuethe [2007-02-09]: > Seems to be working well on i386. ? patches/patch-cgiedit_c ? patches/patch-encode_c Index: Makefile ... What about those two patches? Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: circular requirements: cyrys-sasl
* Chris Paul [2007-01-23]: > $ sudo pkg_add /var/pkgs/openldap-server-2.3.24-bdb.tgz > Can't install openldap-client-2.3.24: can't resolve cyrus-sasl-2.1.21p2-ldap > Can't install cyrus-sasl-2.1.21p2-ldap: can't resolve openldap-client-2.3.24 > Can't install openldap-client-2.3.24: can't resolve cyrus-sasl-2.1.21p2-ldap > Can't install cyrus-sasl-2.1.21p2-ldap: can't resolve openldap-client-2.3.24 > Can't install /var/pkgs/openldap-server-2.3.24-bdb.tgz: can't resolve > cyrus-sasl-2.1.21p2-ldap > > Please advice. The obvious answer is to talk to the port's maintainers. Nikolay
Re: patching: strcpy versus strlcpy
* J.C. Roberts [2007-01-17]: > In a case where you know the external project can/will not accept > patches replacing strcpy with strlcpy due to maintaining support of > other systems and compilers (DOS v4.0, OS2, ...), is it best to take > the #ifdef approach so you can submit them, or is it better do just do > the outright replacement? We don't encourage the replacement in a port anymore. It's way too much work to maintain and not as trivial as one might think. Upstream can easily package strl* with their software. So push them to fix it in their distribution. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: gnash
* Deanna Phillips [2007-01-16]:
> Index: Makefile
> ===
> RCS file: /cvs/ports/www/gnash/Makefile,v
> retrieving revision 1.6
> diff -u -p -r1.6 Makefile
> --- Makefile 8 Dec 2006 15:40:26 - 1.6
> +++ Makefile 16 Jan 2007 02:23:59 -
> @@ -1,9 +1,9 @@
> # $OpenBSD: Makefile,v 1.6 2006/12/08 15:40:26 deanna Exp $
> SHARED_ONLY= Yes
>
> -COMMENT= "flash player with firefox browser plugin"
> +COMMENT= "fake flash player that really just sucks"
> DISTNAME=gnash-0.7.2
> -PKGNAME= ${DISTNAME}p0
> +PKGNAME= ${DISTNAME}p1
> EXTRACT_SUFX=.tar.bz2
>
> CATEGORIES= www
>
> ok?
I suggest a change in DESCR, describing the issue in one or two
sentences, i.e. that gnash is currently unusable to surf most
flash-based websites.
Nikolay
--
"It's all part of my Can't-Do approach to life." Wally
Re: databases/postgresql, update to 8.2.1.
* Marc Balmer [2007-01-15]:
> SHARED_LIBS= ecpg5.2 \
> - ecpg_compat 2.2 \
> - pgtypes 2.2 \
> - pq 4.2
> + ecpg_compat 3.0 \
> + pgtypes 3.0 \
> + pq 5.0
Why do you leave ecpg at 5.2 when even the postgres guys bumped the
lib's minor?
> NAME= ecpg
> -SO_MAJOR_VERSION= 5
> --SO_MINOR_VERSION= 1
> +-SO_MINOR_VERSION= 2
> +SO_MAJOR_VERSION= ${LIBecpg_MAJOR}
> +SO_MINOR_VERSION= ${LIBecpg_MINOR}
Nikolay
Re: do-install or patch Makefile?
* Johan Fredin [2007-01-09]: > What's the preferred way of installing files from a port? Use the > do-install target, or patching a broken (at least for OpenBSD and it's > DESTDIR) install target in the programs own Makefile? I'm seeing both > examples in the tree, and have used the first method myself for a few > small ports. It depends. In general less patches is better, but then it doesn't make sense to rewrite complex install targets. Just choose the more simple solution. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: print/acroread doesn't fetch
* Jack J. Woehr [2007-01-04]: > With current up-to-date, print/acroread doesn't fetch: Actually it does, if you are lucky. Regardless, update to -current and it should work thanks to [EMAIL PROTECTED] Nikolay
Re: vlc security fix.
* Marc Espie [2007-01-04]: > Doesn't hurt to fix it... but we do not have any kind of working vcd > support. I had a look, I've never been able to get the right ioctl to > work (diffs with netbsd, etc). AFAIR mplayer supports playing vcds. Nikolay
Re: x11/openmotif building package problem
* Mikolaj Kucharski [2006-12-14]: > Can anyone reproduce that? not me Nikolay
Re: x11/openmotif building package problem
* Mikolaj Kucharski [2006-12-14]: > Can anyone reproduce that? If you have problems building a port and expect anyone to help you, it is mandatory to provide a build log. Nikolay
Re: how to get new port versions when following 4.0-patch
* Marc Balmer [2006-12-11]: > I can provide patches for the supported -stable OpenBSD versions if > the other ports committers are fine with it (especially nikolay, who > would have to committ it.) We have a simple process, as outlined to developers. If people just followed it, updates would be committed. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: pfstat-2.2.tgz in OpenBSD 4.0
* Ted Buchan [2006-12-07]: > Is it me or is pfstat-2.2.tgz compiled against an old libexpat not > included in OpenBSD4.0? It's probably linked against /usr/X11R6/lib/libexpat.so.5.0. No idea, whether this is correct or not. Nikolay
Re: NEW: www/mozilla-seamonkey 1.0.6
* James Wright [2006-12-02]: > latest -stable mozilla seamonkey (the next generation suite) version > 1.0.6 (equivalent to the 1.5.0.8 releases of firefox/thunderbird), What is the point of this port? Is there anything genuine in seamonkey that is not in firefox and friends? Nikolay
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: [EMAIL PROTECTED] 2006/11/25 01:22:27 Modified files: graphics/agg : Makefile Log message: fix automake build dependency no PKGNAME bump needed
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: [EMAIL PROTECTED] 2006/11/24 10:35:32 Modified files: devel/automake/1.8: Makefile devel/automake/1.9: Makefile Log message: bump PKGNAME after dependency changes from gnu.port.mk
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: [EMAIL PROTECTED] 2006/11/24 10:28:20 Modified files: x11/gtkglarea : Makefile Log message: fix LIB_DEPENDS
Re: Strange problems with shared objects in 4.0
* Alexander Wirt [2006-11-24]: > I had some compile problems with orbit and detected that several libs > were shipped without dependency informations in the ELF Headers. We don't do inter-library dependencies on OpenBSD. Nikolay
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: [EMAIL PROTECTED] 2006/11/23 22:02:05 Modified files: databases : Makefile Log message: p5-DBIx-Class-Validator -> p5-DBIx-Class-Validation
Re: make update failed (4.0/i386)
* Antti Harri [2006-11-16]: > is this supposed to work: I ran make update against installed package > and it complained that it's not installed. However pkg_add -r updated > the package without any problems. make update is quite dumb when it comes to figuring out whether or not a package is installed. FORCE_UPDATE=Yes make install should work, although this will, at least on a -current box, reinstall many packages all the time. Nikolay
Re: tcpstat
* Mark l [2006-11-11]: > Hi, is there any reason the tcpstat port has not been moved from 1.4 > to 1.5? No. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
Re: What's the difference beetwen systrace.filter and systrace.policy?
* Mikolaj Kucharski [2006-11-04]: > What is the difference beetwen systrace.filter and systrace.policy > files in ports? Some ports have policy, some have filter, and some > have both. It's look like policy file has policies separate fo each > executable file, and filter for all files in general. Am I right? A policy is a complete set of filters for one binary, a filter describes how to deal with a single syscall. Therefor systrace.policy is used when when binaries have to be run that don't match the standard policies at all, like linux binaries. systrace.filter is used, when you just need additional rules for single syscalls. Nikolay -- "It's all part of my Can't-Do approach to life." Wally
