CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2010/08/25 00:34:51 Modified files: databases/mysql-zrm: Makefile devel/fossil : Makefile mail/enigmail : Makefile.inc mail/enigmail/seamonkey: Makefile mail/enigmail/thunderbird: Makefile mail/mailcrypt : Makefile mail/p5-Mail-SpamAssassin: Makefile security/gpa : Makefile security/gpgme : Makefile security/p5-GPG: Makefile security/p5-GnuPG: Makefile security/p5-GnuPG-Interface: Makefile security/p5-Module-Signature: Makefile security/p5-PGP-Sign: Makefile security/py-gnupg: Makefile www/rt : Makefile Log message: Force dependency on gnupg1 since gnupg2 has the same PKGNAME, it can create some depency mess. The long term solution is to update+merge gnupg2 into gnupg1 but for now at least we have a clean dependency chain. looks ok to jasper@ and pea@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2010/08/25 00:55:45 Modified files: sysutils/gamin : Makefile Log message: When the no_python FLAVOR is selected, make sure the server subpackage does not depend on python. spotted by naddy@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2010/08/25 01:40:13 Modified files: x11/gtk+2 : Makefile x11/gtk+2/pkg : MESSAGE-main Log message: Reword MESSAGE-main. Discussed with sthen@
KIT PARA QUE NO LE FALTE NADA A SUS CLIENTES
KIT COMPLETO PARA QUE NO LE FALTE NADA INCLUYE LOS SIGUIENTES 6 ARTICULOS PERSONALIZADOS CON SU LOGOTIPO A TODO COLOR MAS 400 ETIQUETAS AUTOADHESIVAS TAMBIEN IMPRESAS SIN CARGO . Cofias para baqo cocidas envasadas en bolsitas de POLIPROPILENO (REFORZADAS) 100 unid. Kit paquelitos tissue (3 paquelitos) 100 unid. Peines unisex Blancos impresos full color envasados en flow pack 100 unid. Kit dental (1 cepillo + sachet dental) 100 unid. Vanity Pack (2 discos desmaquillantes,2 hisopos y lima para uqas) 100 unid. Costureritos muy completos envasados por unidad impresos c/etiqueta full color 100 unid. TOTAL DEL COMBO IVA INCLUIDO $ 684.80 Amenitie Solutions Con 2 paginas Web (triple www punto ) jaboncitos.com.ar (Triple www punto) amenities01.com.ar Si no desea estar informado de nuestro servicio , responder con asubto remover ., Gracias .
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2010/08/25 04:13:33 Modified files: emulators/stella: Makefile distinfo emulators/stella/pkg: PLIST Added files: emulators/stella/patches: patch-configure Log message: - update stella to 3.2 ok benoit@ (MAINTAINER)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2010/08/25 04:16:48 Modified files: infrastructure/templates: network.conf.template Log message: - drop some sites which no longer mirror cpan - add some new sites to replace the above - switch some sites over to using http - notably, the first site on the list is now http, which is a big win on networks with broken ftp (note that peek-ftp skips http mirrors, so this doesn't cause any problem there). okay ajacoutot@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2010/08/25 05:17:42 Modified files: devel/p5-Config-Simple: Makefile Log message: - add mikolaj (submitter) as maintainer
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2010/08/25 07:43:58 Log message: Import gnomint-1.2.1. gnoMint is a tool for an easy creation and management of Certification Authorities. It allows a fancy visualization of all the pieces that conform a CA: x509 certificates, CSRs, CRLs... Currently, it allows the creation of CAs, CSRs and Certificates, and export both public and private parts of them into PEM formatted files. It manages the revocation of the created certificates, as well as the creation of CRLs. ... ok jasper@ Status: Vendor Tag: ajacoutot Release Tags: ajacoutot_20100825 N ports/security/gnomint/distinfo N ports/security/gnomint/Makefile N ports/security/gnomint/patches/patch-configure N ports/security/gnomint/pkg/PLIST N ports/security/gnomint/pkg/DESCR No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2010/08/25 07:45:25 Modified files: security : Makefile Log message: +gnomint
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2010/08/25 08:29:56 Modified files: devel/cmake: Makefile Log message: arm needs -O0
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2010/08/25 10:37:34 Added files: emulators/spim/patches: patch-CPU_run_c Log message: - fix build on mips64, no change for other arches
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2010/08/25 11:26:54 ports/textproc/gnuvd/patches Update of /cvs/ports/textproc/gnuvd/patches In directory cvs.openbsd.org:/tmp/cvs-serv9524/patches Log Message: Directory /cvs/ports/textproc/gnuvd/patches added to the repository
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2010/08/25 11:37:41 Added files: textproc/gnuvd/patches: patch-src_Makefile_in Log message: Remove compiler flag that's not supported by all three gcc versions, -Werror. Fixes build on at least sgi. ok steven@ (MAINTAINER)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2010/08/25 12:00:56 Modified files: graphics/xbmbrowser: Makefile Added files: graphics/xbmbrowser/patches: patch-Imakefile Removed files: graphics/xbmbrowser/patches: patch-a Log message: - fix build on hppa - rename patch - use SUBST_CMD - set license
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2010/08/25 12:04:37 Modified files: games/xboing/patches: patch-Imakefile Log message: - fix build on hppa
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2010/08/25 14:27:16 Modified files: devel/boost: Makefile databases/db/v4: Makefile x11/dbus-glib : Makefile devel/desktop-file-utils: Makefile audio/flac : Makefile security/gnutls: Makefile Log message: add missing wantlibs, new-style lib_depends/revision, bump
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2010/08/25 14:45:52 Modified files: databases/iodbc: Makefile audio/ladspa : Makefile www/libcroco : Makefile devel/libnotify: Makefile devel/libglade2: Makefile multimedia/mjpegtools: Makefile x11/gtk-engines2: Makefile graphics/py-cairo: Makefile Log message: add missing wantlibs, new-style lib_depends/revision, bump
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs
Module name:ports
Changes by: st...@cvs.openbsd.org 2010/08/25 17:06:00
Modified files:
databases/db/v4: Makefile
Log message:
When bumping -main here in my last commit, I inadvertently replaced
PKGNAME-main=${DISTNAME}p0 with REVISION=1. As it is a multi-package port,
the PKGNAME-main=${DISTNAME} must be kept, REVISION must be an extra line.
Breakage noted by jasper@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2010/08/25 17:25:36 Modified files: multimedia/schroedinger: Makefile distinfo multimedia/schroedinger/patches: patch-configure_ac patch-schroedinger_pc_in multimedia/schroedinger/pkg: PLIST Removed files: multimedia/schroedinger/patches: patch-schroedinger_schroroughmotion_c Log message: update schroedinger to 1.0.9; from Brad. this changes a lib dependency from liboil to liborc; dependent ports will be fixed up shortly.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2010/08/25 17:27:21 Modified files: audio/akode: Makefile audio/xmms2: Makefile graphics/amide : Makefile graphics/ffmpeg: Makefile graphics/ffmpegthumbnailer: Makefile graphics/gegl : Makefile graphics/swfdec: Makefile multimedia/dvdstyler: Makefile multimedia/ffmpeg2theora: Makefile multimedia/gstreamer-0.10/plugins-bad: Makefile multimedia/gstreamer-0.10/plugins-ffmpeg: Makefile multimedia/libquicktime: Makefile multimedia/transcode: Makefile multimedia/xine-lib: Makefile telephony/iaxclient: Makefile www/swfdec-plugin: Makefile x11/mplayer: Makefile x11/vlc: Makefile x11/wxSVG : Makefile Log message: - chase the library change in multimedia/schroedinger; oil-orc - add various other missing WANTLIB (and LIB_DEPENDS in some cases) - while there move PKGNAME=..pX to REVISION, and move some ports to new-style LIB_DEPENDS original diff from Brad, extensive checking by me (clean build of everything related to liboil).
Re: CVS: cvs.openbsd.org: ports
On 2010/08/25 17:27, Stuart Henderson wrote: original diff from Brad, extensive checking by me checking + fixing up, to be more correct. (clean build of everything related to liboil). i.e. 300-odd packages and a check-lib-depends sweep of all of them.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ke...@cvs.openbsd.org 2010/08/25 19:10:16 Log message: Initial import of libdisorder 0.0.2 libdisorder is a small, simple C library for use by programmers in other programs. ok jasper@ Status: Vendor Tag: kevlo Release Tags: kevlo_2010-Aug-26 N ports/devel/libdisorder/Makefile N ports/devel/libdisorder/distinfo N ports/devel/libdisorder/files/Makefile N ports/devel/libdisorder/pkg/DESCR N ports/devel/libdisorder/pkg/PLIST N ports/devel/libdisorder/pkg/PFRAG.shared N ports/devel/libdisorder/pkg/PFRAG.no_mips64 No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ke...@cvs.openbsd.org 2010/08/25 19:11:28 Modified files: devel : Makefile Log message: add libdisorder
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jake...@cvs.openbsd.org 2010/08/25 19:44:26 Modified files: multimedia/get_iplayer: Makefile Log message: * set $HOME during do-build, so the manual gets built when USE_SYSTRACE is set * switch to REVISION, bump ok sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2010/08/25 23:16:12 Modified files: devel/avr : Makefile devel/avr/gcc : Makefile devel/avr/gdb : Makefile Log message: - new style wantlib/lib_depends/revisions - order SUBDIRs ok sthen@
KIT PARA QUE NO LE FALTE NADA A SUS CLIENTES
KIT COMPLETO PARA QUE NO LE FALTE NADA INCLUYE LOS SIGUIENTES 6 ARTICULOS PERSONALIZADOS CON SU LOGOTIPO A TODO COLOR MAS 400 ETIQUETAS AUTOADHESIVAS TAMBIEN IMPRESAS SIN CARGO . Cofias para baño cocidas envasadas en bolsitas de POLIPROPILENO (REFORZADAS) 100 unid. Kit pañuelitos tissue (3 pañuelitos) 100 unid. Peines unisex Blancos impresos full color envasados en flow pack 100 unid. Kit dental (1 cepillo + sachet dental) 100 unid. Vanity Pack (2 discos desmaquillantes,2 hisopos y lima para uñas) 100 unid. Costureritos muy completos envasados por unidad impresos c/etiqueta full color 100 unid. TOTAL DEL COMBO IVA INCLUIDO $ 684.80 Amenitie Solutions Con 2 paginas Web (triple www punto ) jaboncitos.com.ar (Triple www punto) amenities01.com.ar Si no desea estar informado de nuestro servicio , responder con asubto remover ., Gracias .
Re: NEW/UPDATE: games/ioquake3
On Tue, Aug 24, 2010 at 10:30:46PM +, Jacob Meuser wrote: On Tue, Aug 24, 2010 at 10:27:37PM +0100, Stuart Henderson wrote: On 2010/08/24 20:29, Jacob Meuser wrote: This seems to be a bug intruduced with xorg 7.5 with this commit: http://cgit.freedesktop.org/xorg/xserver/commit/?id=91f73b79b7ae64e5b846d1efeb470bb61a913720 as reported here: http://bugzilla.libsdl.org/show_bug.cgi?id=971 quick fix (but wrong) here: http://www.mail-archive.com/x...@lists.freedesktop.org/msg05439.html At least that's my understanding. I use the quick and wrong fix and r_gamma works again. Checking the bug report there is also (comment 10) a patch for SDL to support xrandr gamma ramps. (Note X extensions are by no means promised, relying on them for functionality means this happens if the one you depend on becomes deprecated). Does that also work for you? -0- -- It is not true that life is one damn thing after another -- it's one damn thing over and over. -- Edna St. Vincent Millay That patch works for me. So no need for the quick and wrong fix. I split the comment-10 patch into something that ports(7) can eat. Alf I don't follow. please explain why *this port* needs SDL to have patches. or does it affect other ports too? there are many ports that use SDL. Grepping over a (couple-of-months-old) unpacked ports src, the other ports using SDL that are in our tree don't use r_gamma. But from reading the other comments in the bug, if anything else was to start using it, it looks like they would need this patch too. what I don't get is that not everyone is seeing this ... I tried this port on a fresh install of -current. I personally can not explain it because I have no insight into what is going on. However, it worked before (and has always worked since drm(4) is there as well as on Linux boxen with xorg radeon drivers and it broke on Linux too. With this patch (or that other hack) it works again ( - looks like in Windoze)) Of course that does not mean that it should go in/does not break other things. If you do not see it on your machine, you may have set r_ignorehwgamma set to 1. I tried xgamma and although it lets me adjust rgb brightness and general brightness, it does not come close to what q3 is supposed to look like. Most obvious on armor items and brightskins (OSP mod and others). Alf -- jake...@sdf.lonestar.org SDF Public Access UNIX System - http://sdf.lonestar.org
network.conf.template; CPAN updates
- drop some sites which no longer mirror cpan - add some new sites to replace the above - switch some sites over to using http - notably, the first site on the list is now http, which is a big win on networks with broken ftp (note that peek-ftp skips http mirrors, so this doesn't cause any problem there). ok? at least the following could do with similar treatment: gnu, sunsite, xcontrib, ctan, mysql - any volunteers? Index: network.conf.template === RCS file: /cvs/ports/infrastructure/templates/network.conf.template,v retrieving revision 1.144 diff -u -p -r1.144 network.conf.template --- network.conf.template 30 Jul 2010 21:24:39 - 1.144 +++ network.conf.template 25 Aug 2010 09:25:56 - @@ -137,25 +137,25 @@ MASTER_SITE_GNOME+= \ ftp://ftp.nara.wide.ad.jp/pub/X11/GNOME/ \ ftp://ftp.gnome.org/pub/GNOME/ +# http://www.cs.uu.nl/stats/mirmon/cpan.html MASTER_SITE_PERL_CPAN+=\ + http://mirror.sov.uk.goscomb.net/CPAN/modules/by-module/ \ ftp://ftp.funet.fi/pub/languages/perl/CPAN/modules/by-module/ \ - ftp://ftp.mirror.anlx.net/CPAN/modules/by-module/ \ + http://cpan.mirror.anlx.net/CPAN/modules/by-module/ \ ftp://ftp.cpan.org/pub/CPAN/modules/by-module/ \ + http://ftp.kddilabs.jp/lang/perl/CPAN/modules/by-module/ \ ftp://ftp.dnsbalance.ring.gr.jp/pub/lang/perl/CPAN/modules/by-module/ \ http://www.dnsbalance.ring.gr.jp/archives/lang/perl/CPAN/modules/by-module/ \ - ftp://ftp.kddilabs.jp/lang/perl/CPAN/modules/by-module/ \ - ftp://ftp.sunet.se/pub/lang/perl/CPAN/modules/by-module/ \ - ftp://mirror.hiwaay.net/CPAN/modules/by-module/ \ - ftp://csociety-ftp.ecn.purdue.edu/pub/CPAN/modules/by-module/ \ + http://mir2.ovh.net/ftp.cpan.org/modules/by-module/ \ + http://mirror.hiwaay.net/CPAN/modules/by-module/ \ + http://mirrors1.kernel.org/cpan/modules/by-module/ \ ftp://ftp.auckland.ac.nz/pub/perl/CPAN/modules/by-module/ \ - ftp://ftp.isu.net.sa/pub/CPAN/modules/by-module/ \ - ftp://mirrors.ucr.ac.cr/CPAN/modules/by-module/ \ - ftp://mirror.datapipe.net/pub/CPAN/modules/by-module/ \ - ftp://cpan.mirrors.tds.net/pub/CPAN/modules/by-module/ \ + http://mirrors.ucr.ac.cr/CPAN/modules/by-module/ \ + http://mirror.datapipe.net/CPAN/modules/by-module/ \ + http://cpan.mirrors.tds.net/modules/by-module/ \ ftp://ftp.epix.net/pub/languages/perl/modules/by-module/ \ ftp://cpan.pop-mg.com.br/pub/CPAN/modules/by-module/ \ - ftp://ftp.demon.co.uk/pub/CPAN/modules/by-module/ \ - ftp://ftp.chg.ru/pub/lang/perl/CPAN/modules/by-module/ \ + http://cpan.rinet.ru/modules/by-module/ \ http://www.cpan.dk/modules/by-module/ \ ftp://mirrors.dotsrc.org/cpan/modules/by-module/
Re: NEW/UPDATE: games/ioquake3
On 2010 Aug 24 (Tue) at 15:16:54 +0200 (+0200), Alf Schlichting wrote: :That patch works for me. :So no need for the quick and wrong fix. :I split the comment-10 patch into something that ports(7) can eat. : :Alf This patch prevents 'marathoninfinity' from crashing on me in amd64 mode, but just turns the screen black and won't recover. I can vt switch out just once. Userland still runs, and I have to reboot to get my screen back. I should have some time to look at this in more detail later this week. -- Parallel lines never meet, unless you bend one or both of them.
Re: NEW/UPDATE: games/ioquake3
so basically X.org fucked up big time with this? I don't think SDL is an insignificant user of X. oh but that was an extention. yes, we provide these extensions for use, but then they might just go away. I don't see how calling it an extension makes it any less of an API/ABI change. or shall we blame SDL for taking forever to come out with a new release? it's been promised to be ready soon for like 2 years now. either way, this is incredibly annoying. On Wed, Aug 25, 2010 at 09:35:26AM +0200, Alf Schlichting wrote: On Tue, Aug 24, 2010 at 10:30:46PM +, Jacob Meuser wrote: On Tue, Aug 24, 2010 at 10:27:37PM +0100, Stuart Henderson wrote: On 2010/08/24 20:29, Jacob Meuser wrote: This seems to be a bug intruduced with xorg 7.5 with this commit: http://cgit.freedesktop.org/xorg/xserver/commit/?id=91f73b79b7ae64e5b846d1efeb470bb61a913720 as reported here: http://bugzilla.libsdl.org/show_bug.cgi?id=971 quick fix (but wrong) here: http://www.mail-archive.com/x...@lists.freedesktop.org/msg05439.html At least that's my understanding. I use the quick and wrong fix and r_gamma works again. Checking the bug report there is also (comment 10) a patch for SDL to support xrandr gamma ramps. (Note X extensions are by no means promised, relying on them for functionality means this happens if the one you depend on becomes deprecated). Does that also work for you? -0- -- It is not true that life is one damn thing after another -- it's one damn thing over and over. -- Edna St. Vincent Millay That patch works for me. So no need for the quick and wrong fix. I split the comment-10 patch into something that ports(7) can eat. Alf I don't follow. please explain why *this port* needs SDL to have patches. or does it affect other ports too? there are many ports that use SDL. Grepping over a (couple-of-months-old) unpacked ports src, the other ports using SDL that are in our tree don't use r_gamma. But from reading the other comments in the bug, if anything else was to start using it, it looks like they would need this patch too. what I don't get is that not everyone is seeing this ... I tried this port on a fresh install of -current. I personally can not explain it because I have no insight into what is going on. However, it worked before (and has always worked since drm(4) is there as well as on Linux boxen with xorg radeon drivers and it broke on Linux too. With this patch (or that other hack) it works again ( - looks like in Windoze)) Of course that does not mean that it should go in/does not break other things. If you do not see it on your machine, you may have set r_ignorehwgamma set to 1. I tried xgamma and although it lets me adjust rgb brightness and general brightness, it does not come close to what q3 is supposed to look like. Most obvious on armor items and brightskins (OSP mod and others). Alf -- jake...@sdf.lonestar.org SDF Public Access UNIX System - http://sdf.lonestar.org -- jake...@sdf.lonestar.org SDF Public Access UNIX System - http://sdf.lonestar.org
Re: NEW/UPDATE: games/ioquake3
On 2010/08/25 11:32, Jacob Meuser wrote: so basically X.org fucked up big time with this? I don't think SDL is an insignificant user of X. oh but that was an extention. yes, we provide these extensions for use, but then they might just go away. they don't keep the core particularly stable, I don't see what hope extensions have.. :(
OpenBSD Port: rxvt race causes winsize of terminal to be incorrect
Hello. I emailed Jason Ish about this bug, however he says that
he does not have time to work on this port.
Thanks,
-Nate
- Forwarded message from Nathan Houghton nate...@gmail.com -
Date: Sat, 21 Aug 2010 15:27:42 -0400
From: Nathan Houghton nate...@gmail.com
To: Jason Ish i...@openbsd.org
Subject: OpenBSD Port: rxvt race causes winsize of terminal to be incorrect
Hi Jason,
Here is a bug report for the openbsd rxvt port.. I initially
submitted it using sendbug, but was instructed to send it on to
you.
Number: 6450
Category: ports
Synopsis: rxvt race causes winsize of terminal to be
incorrect
Confidential: yes
Severity: serious
Priority: medium
Responsible:bugs
State: closed
Quarter:
Keywords:
Date-Required:
Class: sw-bug
Submitter-Id: unknown
Arrival-Date: Sat Aug 21 17:20:01 GMT 2010
Closed-Date:Sat Aug 21 11:42:46 MDT 2010
Last-Modified: Sat Aug 21 11:42:46 MDT 2010
Originator:
Release:
Organization:
Environment:
System : OpenBSD 4.5
Details : OpenBSD 4.5-stable (GENERIC) #2: Tue Aug 17
23:04:08 EDT 2010
r...@iron.brainwerk.org:/usr/src/sys/arch/i386/compile/GENERIC
Architecture: OpenBSD.i386
Machine : i386
Description:
A race condition exists in rxvt during startup that can
cause the terminal size to be set (with ioctl TIOCSWINSZ)
to the size determined by the rxvt window size, followed
by setting the terminal size to 80x24, resulting in a
large
rxvt window that has only an 80x24 size terminal usable
inside.
How-To-Repeat:
Use a window manager such as scrotwm that will
immediately
resize the terminal window when it is launched. Simply
launch rxvt. Inside rxvt run something like vi, or use
TIOCGWINSZ to get the terminal size. It will be set to
80x24. After a window resize the terminal size will be
corrected.
Fix:
Don't call rxvt_tt_winsize in both processes during
initialization.
--- src/init.c.orig Sat Aug 21 00:30:38 2010
+++ src/init.c Sat Aug 21 00:30:56 2010
@@ -1316,6 +1316,9 @@
/* need to trap SIGURG for SVR4 (Unixware) rlogin */
/* signal (SIGURG, SIG_DFL); */
+/* set window size */
+rxvt_tt_winsize(STDIN_FILENO, r-TermWin.ncol,
r-TermWin.nrow);
+
#ifndef __QNX__
/* spin off the command interpreter */
switch (r-h-cmd_pid = fork()) {
@@ -1398,9 +1401,6 @@
}
#endif /* SRIOCSREDIR */
}
-
-/* set window size */
-rxvt_tt_winsize(STDIN_FILENO, r-TermWin.ncol,
r-TermWin.nrow);
/* reset signals and spin off the command interpreter */
signal(SIGINT, SIG_DFL);
-Nate
- End forwarded message -
Re: OpenBSD Port: rxvt race causes winsize of terminal to be incorrect
I think you need to go talk to the rxvt folks.
this says you're using obsd 4.5. what's the status on more recent
releases?
On Wed, Aug 25, 2010 at 02:43:18PM -0400, Nathan Houghton wrote:
Hello. I emailed Jason Ish about this bug, however he says that
he does not have time to work on this port.
Thanks,
-Nate
- Forwarded message from Nathan Houghton nate...@gmail.com -
Date: Sat, 21 Aug 2010 15:27:42 -0400
From: Nathan Houghton nate...@gmail.com
To: Jason Ish i...@openbsd.org
Subject: OpenBSD Port: rxvt race causes winsize of terminal to be incorrect
Hi Jason,
Here is a bug report for the openbsd rxvt port.. I initially
submitted it using sendbug, but was instructed to send it on to
you.
Number: 6450
Category: ports
Synopsis: rxvt race causes winsize of terminal to be
incorrect
Confidential: yes
Severity: serious
Priority: medium
Responsible:bugs
State: closed
Quarter:
Keywords:
Date-Required:
Class: sw-bug
Submitter-Id: unknown
Arrival-Date: Sat Aug 21 17:20:01 GMT 2010
Closed-Date:Sat Aug 21 11:42:46 MDT 2010
Last-Modified: Sat Aug 21 11:42:46 MDT 2010
Originator:
Release:
Organization:
Environment:
System : OpenBSD 4.5
Details : OpenBSD 4.5-stable (GENERIC) #2: Tue Aug 17
23:04:08 EDT 2010
r...@iron.brainwerk.org:/usr/src/sys/arch/i386/compile/GENERIC
Architecture: OpenBSD.i386
Machine : i386
Description:
A race condition exists in rxvt during startup that can
cause the terminal size to be set (with ioctl TIOCSWINSZ)
to the size determined by the rxvt window size, followed
by setting the terminal size to 80x24, resulting in a
large
rxvt window that has only an 80x24 size terminal usable
inside.
How-To-Repeat:
Use a window manager such as scrotwm that will
immediately
resize the terminal window when it is launched. Simply
launch rxvt. Inside rxvt run something like vi, or use
TIOCGWINSZ to get the terminal size. It will be set to
80x24. After a window resize the terminal size will be
corrected.
Fix:
Don't call rxvt_tt_winsize in both processes during
initialization.
--- src/init.c.orig Sat Aug 21 00:30:38 2010
+++ src/init.cSat Aug 21 00:30:56 2010
@@ -1316,6 +1316,9 @@
/* need to trap SIGURG for SVR4 (Unixware) rlogin */
/* signal (SIGURG, SIG_DFL); */
+/* set window size */
+rxvt_tt_winsize(STDIN_FILENO, r-TermWin.ncol,
r-TermWin.nrow);
+
#ifndef __QNX__
/* spin off the command interpreter */
switch (r-h-cmd_pid = fork()) {
@@ -1398,9 +1401,6 @@
}
#endif /* SRIOCSREDIR */
}
-
-/* set window size */
-rxvt_tt_winsize(STDIN_FILENO, r-TermWin.ncol,
r-TermWin.nrow);
/* reset signals and spin off the command interpreter */
signal(SIGINT, SIG_DFL);
-Nate
- End forwarded message -
--
jake...@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org
rm nawk (was: Re: CVS: cvs.openbsd.org: src)
Theo de Raadt dera...@cvs.openbsd.org wrote: CVSROOT: /cvs Module name: src Changes by: dera...@cvs.openbsd.org 2010/08/25 13:21:04 Modified files: usr.bin/awk: Makefile awk.1 Log message: get rid of the nawk hard link ok millert guenther People following -current should remove the nawk files from their system... and pay attention to possible ports fallout. -- Christian naddy Weisgerber na...@mips.inka.de
Re: OpenBSD Port: rxvt race causes winsize of terminal to be incorrect
As far as I can tell from the OpenBSD web cvs, this port has not
changed in a long while (years). I can try contacting the rxvt
folks, but the project seems to be pretty dead lately (as far as I
can see on the sourceforge page).
-Nate
On Wed, Aug 25, 2010 at 09:47:48PM +, Jacob Meuser wrote:
I think you need to go talk to the rxvt folks.
this says you're using obsd 4.5. what's the status on more recent
releases?
On Wed, Aug 25, 2010 at 02:43:18PM -0400, Nathan Houghton wrote:
Hello. I emailed Jason Ish about this bug, however he says that
he does not have time to work on this port.
Thanks,
-Nate
- Forwarded message from Nathan Houghton nate...@gmail.com -
Date: Sat, 21 Aug 2010 15:27:42 -0400
From: Nathan Houghton nate...@gmail.com
To: Jason Ish i...@openbsd.org
Subject: OpenBSD Port: rxvt race causes winsize of terminal to be incorrect
Hi Jason,
Here is a bug report for the openbsd rxvt port.. I initially
submitted it using sendbug, but was instructed to send it on to
you.
Number: 6450
Category: ports
Synopsis: rxvt race causes winsize of terminal to be
incorrect
Confidential: yes
Severity: serious
Priority: medium
Responsible:bugs
State: closed
Quarter:
Keywords:
Date-Required:
Class: sw-bug
Submitter-Id: unknown
Arrival-Date: Sat Aug 21 17:20:01 GMT 2010
Closed-Date:Sat Aug 21 11:42:46 MDT 2010
Last-Modified: Sat Aug 21 11:42:46 MDT 2010
Originator:
Release:
Organization:
Environment:
System : OpenBSD 4.5
Details : OpenBSD 4.5-stable (GENERIC) #2: Tue Aug 17
23:04:08 EDT 2010
r...@iron.brainwerk.org:/usr/src/sys/arch/i386/compile/GENERIC
Architecture: OpenBSD.i386
Machine : i386
Description:
A race condition exists in rxvt during startup that can
cause the terminal size to be set (with ioctl TIOCSWINSZ)
to the size determined by the rxvt window size, followed
by setting the terminal size to 80x24, resulting in a
large
rxvt window that has only an 80x24 size terminal usable
inside.
How-To-Repeat:
Use a window manager such as scrotwm that will
immediately
resize the terminal window when it is launched. Simply
launch rxvt. Inside rxvt run something like vi, or use
TIOCGWINSZ to get the terminal size. It will be set to
80x24. After a window resize the terminal size will be
corrected.
Fix:
Don't call rxvt_tt_winsize in both processes during
initialization.
--- src/init.c.orig Sat Aug 21 00:30:38 2010
+++ src/init.c Sat Aug 21 00:30:56 2010
@@ -1316,6 +1316,9 @@
/* need to trap SIGURG for SVR4 (Unixware) rlogin */
/* signal (SIGURG, SIG_DFL); */
+/* set window size */
+rxvt_tt_winsize(STDIN_FILENO, r-TermWin.ncol,
r-TermWin.nrow);
+
#ifndef __QNX__
/* spin off the command interpreter */
switch (r-h-cmd_pid = fork()) {
@@ -1398,9 +1401,6 @@
}
#endif /* SRIOCSREDIR */
}
-
-/* set window size */
-rxvt_tt_winsize(STDIN_FILENO, r-TermWin.ncol,
r-TermWin.nrow);
/* reset signals and spin off the command interpreter */
signal(SIGINT, SIG_DFL);
-Nate
- End forwarded message -
--
jake...@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org
Re: OpenBSD Port: rxvt race causes winsize of terminal to be incorrect
On 2010/08/25 18:28, Nathan Houghton wrote:
As far as I can tell from the OpenBSD web cvs, this port has not
changed in a long while (years). I can try contacting the rxvt
folks, but the project seems to be pretty dead lately (as far as I
can see on the sourceforge page).
I agree, no substantial changes since 2004. I think this should go
in. Any ok's?
Index: Makefile
===
RCS file: /cvs/ports/x11/rxvt/Makefile,v
retrieving revision 1.41
diff -u -p -r1.41 Makefile
--- Makefile30 Jul 2010 06:59:00 - 1.41
+++ Makefile25 Aug 2010 22:58:25 -
@@ -6,9 +6,9 @@ COMMENT-clock= analog clock for X11 w/ap
VER= 2.7.10
DISTNAME= rxvt-${VER}
PKGNAME-main= rxvt-${VER}
-REVISION-main= 4
+REVISION-main= 5
PKGNAME-clock= rclock-${VER}
-REVISION-clock=0
+REVISION-clock=1
CATEGORIES=x11
MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=rxvt/} \
ftp://ftp.fu-berlin.de/unix/X11/terms/rxvt/ \
@@ -19,7 +19,7 @@ HOMEPAGE= http://www.rxvt.org/
MAINTAINER=Jason Ish i...@openbsd.org
-# GPL
+# GPLv2+
PERMIT_PACKAGE_CDROM= Yes
PERMIT_PACKAGE_FTP=Yes
PERMIT_DISTFILES_CDROM=Yes
Index: patches/patch-src_init_c
===
RCS file: patches/patch-src_init_c
diff -N patches/patch-src_init_c
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-src_init_c25 Aug 2010 22:58:25 -
@@ -0,0 +1,26 @@
+$OpenBSD$
+
+avoid a race commonly hit by tiling WMs
+
+--- src/init.c.origWed Aug 25 23:52:24 2010
src/init.c Wed Aug 25 23:53:06 2010
+@@ -1316,6 +1316,9 @@ rxvt_run_command(rxvt_t *r, const char *const
*argv)
+ /* need to trap SIGURG for SVR4 (Unixware) rlogin */
+ /* signal (SIGURG, SIG_DFL); */
+
++/* set window size */
++rxvt_tt_winsize(STDIN_FILENO, r-TermWin.ncol, r-TermWin.nrow);
++
+ #ifndef __QNX__
+ /* spin off the command interpreter */
+ switch (r-h-cmd_pid = fork()) {
+@@ -1398,9 +1401,6 @@ rxvt_run_child(rxvt_t *r, const char *const
*argv)
+ }
+ #endif/* SRIOCSREDIR */
+ }
+-
+-/* set window size */
+-rxvt_tt_winsize(STDIN_FILENO, r-TermWin.ncol, r-TermWin.nrow);
+
+ /* reset signals and spin off the command interpreter */
+ signal(SIGINT, SIG_DFL);
Index: pkg/PLIST-clock
===
RCS file: /cvs/ports/x11/rxvt/pkg/PLIST-clock,v
retrieving revision 1.2
diff -u -p -r1.2 PLIST-clock
--- pkg/PLIST-clock 15 Sep 2004 19:26:31 - 1.2
+++ pkg/PLIST-clock 25 Aug 2010 22:58:25 -
@@ -1,3 +1,3 @@
@comment $OpenBSD: PLIST-clock,v 1.2 2004/09/15 19:26:31 espie Exp $
-bin/rclock
+...@bin bin/rclock
@man man/man1/rclock.1
Index: pkg/PLIST-main
===
RCS file: /cvs/ports/x11/rxvt/pkg/PLIST-main,v
retrieving revision 1.1
diff -u -p -r1.1 PLIST-main
--- pkg/PLIST-main 21 Nov 2006 21:04:56 - 1.1
+++ pkg/PLIST-main 25 Aug 2010 22:58:25 -
@@ -1,6 +1,6 @@
@comment $OpenBSD: PLIST-main,v 1.1 2006/11/21 21:04:56 espie Exp $
@pkgpath x11/rxvt
@mode 4755
-bin/rxvt
+...@bin bin/rxvt
@mode
@man man/man1/rxvt.1
Re: rm nawk (was: Re: CVS: cvs.openbsd.org: src)
On 2010/08/25 21:54, Christian Weisgerber wrote: Theo de Raadt dera...@cvs.openbsd.org wrote: CVSROOT:/cvs Module name:src Changes by: dera...@cvs.openbsd.org 2010/08/25 13:21:04 Modified files: usr.bin/awk: Makefile awk.1 Log message: get rid of the nawk hard link ok millert guenther People following -current should remove the nawk files from their system... and pay attention to possible ports fallout. s/possible//. nawk is in our autoconf cache.
Re: OpenBSD Port: rxvt race causes winsize of terminal to be incorrect
Apparently this bug was reported in 2004.
http://sourceforge.net/tracker/?func=detailaid=981781group_id=221atid=100221
I went ahead and posted a patch there for the next version (if
there is one).
-Nate
On Thu, Aug 26, 2010 at 12:01:25AM +0100, Stuart Henderson wrote:
On 2010/08/25 18:28, Nathan Houghton wrote:
As far as I can tell from the OpenBSD web cvs, this port has not
changed in a long while (years). I can try contacting the rxvt
folks, but the project seems to be pretty dead lately (as far as I
can see on the sourceforge page).
I agree, no substantial changes since 2004. I think this should go
in. Any ok's?
Index: Makefile
===
RCS file: /cvs/ports/x11/rxvt/Makefile,v
retrieving revision 1.41
diff -u -p -r1.41 Makefile
--- Makefile 30 Jul 2010 06:59:00 - 1.41
+++ Makefile 25 Aug 2010 22:58:25 -
@@ -6,9 +6,9 @@ COMMENT-clock=analog clock for X11 w/ap
VER= 2.7.10
DISTNAME=rxvt-${VER}
PKGNAME-main=rxvt-${VER}
-REVISION-main= 4
+REVISION-main= 5
PKGNAME-clock= rclock-${VER}
-REVISION-clock= 0
+REVISION-clock= 1
CATEGORIES= x11
MASTER_SITES=${MASTER_SITE_SOURCEFORGE:=rxvt/} \
ftp://ftp.fu-berlin.de/unix/X11/terms/rxvt/ \
@@ -19,7 +19,7 @@ HOMEPAGE= http://www.rxvt.org/
MAINTAINER= Jason Ish i...@openbsd.org
-# GPL
+# GPLv2+
PERMIT_PACKAGE_CDROM=Yes
PERMIT_PACKAGE_FTP= Yes
PERMIT_DISTFILES_CDROM= Yes
Index: patches/patch-src_init_c
===
RCS file: patches/patch-src_init_c
diff -N patches/patch-src_init_c
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-src_init_c 25 Aug 2010 22:58:25 -
@@ -0,0 +1,26 @@
+$OpenBSD$
+
+avoid a race commonly hit by tiling WMs
+
+--- src/init.c.orig Wed Aug 25 23:52:24 2010
src/init.c Wed Aug 25 23:53:06 2010
+@@ -1316,6 +1316,9 @@ rxvt_run_command(rxvt_t *r, const char *const
*argv)
+ /* need to trap SIGURG for SVR4 (Unixware) rlogin */
+ /* signal (SIGURG, SIG_DFL); */
+
++/* set window size */
++rxvt_tt_winsize(STDIN_FILENO, r-TermWin.ncol, r-TermWin.nrow);
++
+ #ifndef __QNX__
+ /* spin off the command interpreter */
+ switch (r-h-cmd_pid = fork()) {
+@@ -1398,9 +1401,6 @@ rxvt_run_child(rxvt_t *r, const char *const
*argv)
+ }
+ #endif /* SRIOCSREDIR */
+ }
+-
+-/* set window size */
+-rxvt_tt_winsize(STDIN_FILENO, r-TermWin.ncol, r-TermWin.nrow);
+
+ /* reset signals and spin off the command interpreter */
+ signal(SIGINT, SIG_DFL);
Index: pkg/PLIST-clock
===
RCS file: /cvs/ports/x11/rxvt/pkg/PLIST-clock,v
retrieving revision 1.2
diff -u -p -r1.2 PLIST-clock
--- pkg/PLIST-clock 15 Sep 2004 19:26:31 - 1.2
+++ pkg/PLIST-clock 25 Aug 2010 22:58:25 -
@@ -1,3 +1,3 @@
@comment $OpenBSD: PLIST-clock,v 1.2 2004/09/15 19:26:31 espie Exp $
-bin/rclock
+...@bin bin/rclock
@man man/man1/rclock.1
Index: pkg/PLIST-main
===
RCS file: /cvs/ports/x11/rxvt/pkg/PLIST-main,v
retrieving revision 1.1
diff -u -p -r1.1 PLIST-main
--- pkg/PLIST-main21 Nov 2006 21:04:56 - 1.1
+++ pkg/PLIST-main25 Aug 2010 22:58:25 -
@@ -1,6 +1,6 @@
@comment $OpenBSD: PLIST-main,v 1.1 2006/11/21 21:04:56 espie Exp $
@pkgpath x11/rxvt
@mode 4755
-bin/rxvt
+...@bin bin/rxvt
@mode
@man man/man1/rxvt.1
audio/timidity cleanup
* quit building audio(4) backend
* fix compilation of xaw flavor
* REVISION and bump
* fix and new style LIB_DEPENDS/WANTLIB
ok?
--
jake...@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org
Index: Makefile
===
RCS file: /cvs/ports/audio/timidity/Makefile,v
retrieving revision 1.36
diff -N -u -p Makefile
--- Makefile23 Aug 2010 11:17:17 - 1.36
+++ Makefile26 Aug 2010 02:11:33 -
@@ -4,7 +4,8 @@ COMMENT=MIDI to WAV renderer and player
V= 2.13.2
DISTNAME= TiMidity++-${V}
-PKGNAME= timidity-${V}p2
+PKGNAME= timidity-${V}
+REVISION= 3
CATEGORIES=audio
MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=timidity/}
@@ -25,7 +26,7 @@ WANTLIB= c m ncurses sndio
CONFIGURE_STYLE=autoconf automake gnu
AUTOCONF_VERSION = 2.59
AUTOMAKE_VERSION = 1.9
-CONFIGURE_ARGS= --enable-audio=sndio,sun \
+CONFIGURE_ARGS= --enable-audio=sndio \
--with-default-output=sndio \
--enable-vt100 \
--enable-ncurses \
@@ -39,18 +40,20 @@ FLAVOR?=
CONFIGURE_ARGS+=--enable-gtk
MODULES= devel/gettext
USE_X11= Yes
-LIB_DEPENDS+= gdk-x11-2.0,gdk_pixbuf-2.0,gtk-x11-2.0::x11/gtk+2
+LIB_DEPENDS+= ::x11/gtk+2
WANTLIB+= X11 Xau Xcomposite Xcursor Xdamage Xdmcp Xext Xfixes \
Xi Xinerama Xrandr Xrender atk-1.0 cairo expat \
- fontconfig freetype glib-2.0 glitz gio-2.0 gmodule-2.0 \
- gobject-2.0 pango-1.0 pangocairo-1.0 pangoft2-1.0 \
- pixman-1 png pthread-stubs xcb z
+ fontconfig freetype gdk-x11-2.0 gdk_pixbuf-2.0 \
+ glib-2.0 glitz gio-2.0 gmodule-2.0 gobject-2.0 \
+ gtk-x11-2.0 gthread-2.0 pango-1.0 pangocairo-1.0 \
+ pangoft2-1.0 pixman-1 png pthread pthread-stubs xcb \
+ xcb-render xcb-render-util z
.endif
.if ${FLAVOR:L:Mxaw}
CONFIGURE_ARGS+=--enable-xaw
USE_X11= Yes
-WANTLIB+= ICE SM X11 Xaw Xext Xmu Xt pthread-stubs xcb
+WANTLIB+= ICE SM X11 Xaw Xext Xmu Xt
.endif
.if ! ${FLAVOR:L:Mgtk2} ! ${FLAVOR:L:Mxaw}
Index: patches/patch-TiMidity++-2_13_2_interface_xaw_i_c
===
RCS file: patches/patch-TiMidity++-2_13_2_interface_xaw_i_c
diff -N -u -p patches/patch-TiMidity++-2_13_2_interface_xaw_i_c
--- /dev/null 25 Aug 2010 20:11:33 -
+++ patches/patch-TiMidity++-2_13_2_interface_xaw_i_c 26 Aug 2010 02:11:33
-
@@ -0,0 +1,30 @@
+$OpenBSD$
+--- TiMidity++-2.13.2/interface/xaw_i.c.orig Wed Aug 25 19:07:52 2010
TiMidity++-2.13.2/interface/xaw_i.cWed Aug 25 19:08:36 2010
+@@ -261,7 +261,7 @@ int amplitude = DEFAULT_AMPLIFICATION;
+ String bitmapdir = XAW_BITMAP_DIR;
+ Boolean arrangetitle,savelist;
+ static char **current_flist = NULL;
+-static int voices = 0, last_voice = 0, voices_num_width;
++static int nvoices = 0, last_voice = 0, voices_num_width;
+ static int maxentry_on_a_menu = 0,submenu_n = 0;
+ #define OPTIONS_WINDOW 1
+ #define FLIST_WINDOW 2
+@@ -1329,7 +1329,7 @@ static void handle_input(XtPointer data,int *source,Xt
+ c= *(local_buf+1);
+ n= atoi(local_buf+2);
+ if(c == 'L')
+- voices = n;
++ nvoices = n;
+ else
+ last_voice = n;
+ if(IsTracePlaying()) drawVoices();
+@@ -2070,7 +2070,7 @@ static void drawVoices(void) {
+ XSetForeground(disp, gct, tracecolor);
+ XFillRectangle(disp,XtWindow(trace),gct,voices_num_width +4,
+
MAX_XAW_MIDI_CHANNELS*BAR_SPACE+TRACEV_OFS+1,VOICENUM_WIDTH,TRACE_FOOT);
+- sprintf(local_buf, %3d/%d, last_voice, voices);
++ sprintf(local_buf, %3d/%d, last_voice, nvoices);
+ XSetForeground(disp, gct, capcolor);
+ XDrawString(disp, XtWindow(trace),gct,voices_num_width+6,
+
MAX_XAW_MIDI_CHANNELS*BAR_SPACE+TRACEV_OFS+16,local_buf,strlen(local_buf));
