Re: (n to go) progress bar
On Thu, 17 Dec 2009, Marc Espie wrote: On Thu, Dec 17, 2009 at 12:26:42PM -0600, Marco Peereboom wrote: On Thu, Dec 17, 2009 at 07:00:08PM +0100, Marc Espie wrote: On Thu, Dec 17, 2009 at 06:43:02PM +0100, frantisek holop wrote: on the internet, you are never alone ;] -f Yeah, but Paul is our resident gypsy, so we routinely ignore him (and doubly so if we're rude frogs) I am on Paul's side here. So the evil capitalists are also against rude frogging one's eyes too. You're probably one of those guys who either: - uses <10 packages - has a killer-fast connection to the nearest mirror. Everybody else out there is very happy to know that they still have 190 packages to go through when doing an update that last for 3 hours. Think about it. If you don't like it, use -x, all the progress messages should go away. Hi, I got 306 packages on my desktop and a crappy 8Mbps adsl, so I like the new "n to go". I can see some kind of estimate of the time left. However I kind of agree with the other side too, it could be more accurate if that's technically possible and don't make things slower than they are now. Then I'm going to hijack this thread a little: have others seen ftp hangings recently with the new pkg_add? It always hangs where it finds new candidates, never when downloading or actually installing them. When I do pkill ftp it continues, usually without an error. I'm thinking it's either a problem with the adsl modem I recently got, or pkg_add. i...@kameli:~$ sudo pkg_add -ui Not updating .libs*, remember to clean them Looking for packages|No need to update a2ps-4.14p1-a4 (300 to go) -- Antti Harri
Re: (n to go) progress bar
hmm, on Thu, Dec 17, 2009 at 04:36:16PM -0600, Marco Peereboom said that > Speaking of the reason why pkg_add -ui is annoying. It takes hours to > update all my packages and when I get to my laptop in the morning it > sits there waiting for me to answer y. Repeat like 10 times. i am sure there is a keyword for this with -F, but i never understood the documentation very much there, and my repeated pleas for either examples or rephrasing went unheard. > > 190 packages to go through when doing an update that last for 3 hours. > > Think about it. > > I like to know it but it is inaccurate. it's not just inaccurate, or an estimate, but as i said before, as all the packages have different sizes, different dependencies, it is a useless metric timewise. (190 to go) could mean 3h or 30m. i am _not_ saying it should go away, i am merely suggesting it could find a new home on a higher verbose level. > > If you don't like it, use -x, all the progress messages should go away. i am fine with the basic progress bar. i could also use > /dev/null to make everything go away. i'd just like a very simple, very spartan ftp-like progress bar. is there an option for that? -f -- don't eat yellow snow.
sysutils/fsstress
I'd like to commit this. I promise I'll take care of it and feed it. Peter had a peek on icb. fsstress.tgz Description: GNU Zip compressed data
rwhoisd
I'm going to start work on porting rwhoisd, is anyone else working on this? -- Thanks Andrew Klettke Optic Fusion NOC 253-830-2943
Re: sndio for games/xblast
On Wednesday 16 December 2009 23:50:45 Jacob Meuser wrote:
> also, port-lib-depends-check says pthread-stubs and xcb shouldn't be
> WANTLIB.
I have seen this before and don't really understand how it thinks they're
not necessary. libX11 needs pthread-stubs and xcb.
> --
> jake...@sdf.lonestar.org
> SDF Public Access UNIX System - http://sdf.lonestar.org
>
> Index: Makefile
> ===
> RCS file: /cvs/ports/games/xblast/Makefile,v
> retrieving revision 1.23
> diff -N -u -p Makefile
> --- Makefile 10 Aug 2009 06:31:43 - 1.23
> +++ Makefile 17 Dec 2009 04:44:57 -
> @@ -3,7 +3,7 @@
> COMMENT= graphical multi-player real-time strategy game for X11
>
> DISTNAME=xblast-2.10.4
> -PKGNAME= ${DISTNAME}p0
> +PKGNAME= ${DISTNAME}p1
> DIST_SUBDIR= xblast
> DISTFILES= ${DISTNAME}${EXTRACT_SUFX} \
> images-2005-01-06${EXTRACT_SUFX} \
> @@ -21,7 +21,7 @@ PERMIT_PACKAGE_FTP= Yes
> PERMIT_DISTFILES_CDROM= Yes
> PERMIT_DISTFILES_FTP=Yes
>
> -WANTLIB= ICE X11 c m ossaudio pthread-stubs xcb
> +WANTLIB= ICE X11 c m sndio
>
> MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=xblast/}
>
> @@ -31,7 +31,7 @@ MODULES=devel/gettext
>
> CONFIGURE_STYLE=autoconf
> AUTOCONF_VERSION=2.59
> -CONFIGURE_ENV= LDFLAGS="-lossaudio \
> +CONFIGURE_ENV= LDFLAGS="-lsndio \
> -L${LOCALBASE}/lib -liconv -lintl"
> CONFIGURE_ARGS= --enable-sound \
> --with-otherdatadir=${PREFIX}/share/xblast \
> Index: patches/patch-xbsndsrv_c
> ===
> RCS file: /cvs/ports/games/xblast/patches/patch-xbsndsrv_c,v
> retrieving revision 1.3
> diff -N -u -p patches/patch-xbsndsrv_c
> --- patches/patch-xbsndsrv_c 20 Dec 2007 22:03:05 - 1.3
> +++ patches/patch-xbsndsrv_c 17 Dec 2009 04:44:57 -
> @@ -1,24 +1,116 @@
> $OpenBSD: patch-xbsndsrv_c,v 1.3 2007/12/20 22:03:05 ajacoutot Exp $
> xbsndsrv.c.orig Mon Jun 12 13:06:36 2006
> -+++ xbsndsrv.c Sun Dec 16 15:37:02 2007
> -@@ -75,6 +75,9 @@
> +--- xbsndsrv.c.orig Mon Jun 12 04:06:36 2006
> xbsndsrv.c Wed Dec 16 19:48:41 2009
> +@@ -75,6 +75,10 @@
> #ifdef HAVE_LINUX_SOUNDCARD_H
> #include
> #endif
> +#ifdef __OpenBSD__
> -+#include
> ++#define USE_SNDIO
> ++#include
> +#endif
>
> #endif
>
> -@@ -263,6 +266,10 @@ static struct _sound_name
> +@@ -263,6 +267,10 @@ static struct _sound_name
> #define SAMPLE_CHANNELS 2
> #define SAMPLE_SIZE 16
> #define SUN_AUDIO_REF 0
> -+#elif __OpenBSD__
> -+#define SOUND_DEVICE "/dev/sound"
> ++#elif defined(USE_SNDIO)
> ++#define SOUND_DEVICE "default"
> +#define SAMPLE_CHANNELS 1
> +#define SAMPLE_SIZE 8
> #else
> #define SOUND_DEVICE "/dev/dsp"
> #define SAMPLE_CHANNELS 1
> +@@ -283,6 +291,10 @@ static int sample_rate = SAMPLE_RATE;
> + static int sample_channels = SAMPLE_CHANNELS;
> + static int sample_size = SAMPLE_SIZE;
> +
> ++#ifdef USE_SNDIO
> ++struct sio_hdl *hdl;
> ++#endif
> ++
> + /*
> + * outcomment the following line to suppress server statistics
> + */
> +@@ -371,8 +383,20 @@ init_dsp (dsp)
> + fprintf (stderr, " (wanted %d channels, %d bits, %dHz)\n",
> + sample_channels, sample_size, sample_rate);
> +
> +-#else
> ++#elif defined(USE_SNDIO)
> ++struct sio_par par;
> +
> ++sio_initpar(&par);
> ++par.rate = sample_rate;
> ++par.bits = sample_size;
> ++par.sig = par.bits == 8 ? 0 : 1;
> ++par.pchan = sample_channels ? 2 : 1;
> ++par.appbufsz = 1024;
> ++
> ++if (!sio_setpar(hdl, &par) || !sio_getpar(hdl, &par) || !sio_start(hdl))
> ++fprintf (stderr, "XBlast sound server: could not configure
> sndio\n");
> ++
> ++#else
> + if (ioctl (dsp, SNDCTL_DSP_SETFRAGMENT, &fragsize) < 0) {
> + fprintf (stderr, "XBlast sound server: could not set fragment
> size %8x
> on sound device\n", + fragsize);
> +@@ -418,7 +442,7 @@ resync (dsp)
> + #ifdef __sun__
> + ioctl (dsp, I_FLUSH, NULL);
> + fprintf (stderr, "\nsync\n");
> +-#else
> ++#elif !defined(USE_SNDIO)
> + /* resync sound device to correct any channel flipping */
> + write (dsp, playbuff, SUBSIZE);
> + write (dsp, playbuff, SUBSIZE);
> +@@ -601,7 +625,11 @@ main (argc, argv)
> + /*
> + * open and prepare sound device
> + */
> ++#ifdef USE_SNDIO
> ++if ((hdl = sio_open (NULL, SIO_PLAY, 0)) == NULL) {
> ++#else
> + if ((dsp = open (SOUND_DEVICE, O_WRONLY)) < 0) {
> ++#endif
> + fprintf (stderr, "XBlast sound server: could not open sound
> device
> %s\n", SOUND_DEVICE); + ack_val = SND_ACK_ERROR;
> + write (1, &ack_val, sizeof (ack_val));
> +@@ -780,7 +808,11 @@ main (argc, argv)
> + }
> +
> + /* play buffer */
> ++#ifdef USE_SNDIO
> ++
Re: UPDATE: Warzone 2100 2.3
On Sun, Dec 13, 2009 at 10:41:29PM -0500, Brad wrote:
> Here is an update to Warzone 2100 2.3 beta 4. I'm putting this
> out for a bit more testing. Please test and provide any feedback.
Here is an updated diff after the sndio addition to OpenAL.
BTW, I also forgot to mention that Warzone seems to expose
bug(s) in DRI with at least Radeon hw. Can anyone try this
with an Intel GPU?
Index: Makefile
===
RCS file: /cvs/ports/games/warzone2100/Makefile,v
retrieving revision 1.3
diff -u -p -r1.3 Makefile
--- Makefile10 Aug 2009 06:31:43 - 1.3
+++ Makefile17 Dec 2009 00:38:12 -
@@ -1,48 +1,48 @@
# $OpenBSD: Makefile,v 1.3 2009/08/10 06:31:43 kili Exp $
-COMMENT= 3d realtime strategy on a future earth
+COMMENT= 3D realtime strategy on a future earth
-DISTNAME= warzone2100-2.0.7
-PKGNAME= ${DISTNAME}p0
+DISTNAME= warzone2100-2.3_beta4
CATEGORIES=games
-EXTRACT_SUFX= .tar.bz2
+MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=warzone2100/}
HOMEPAGE= http://wz2100.net/
MAINTAINER=Simon Bertrang
-# GPL
+# GPLv2+
PERMIT_PACKAGE_CDROM= Yes
PERMIT_PACKAGE_FTP=Yes
PERMIT_DISTFILES_CDROM=Yes
PERMIT_DISTFILES_FTP= Yes
-WANTLIB= GL GLU SDL X11 Xext c m ogg ossaudio pthread pthread-stubs \
- usbhid xcb z
+WANTLIB= GL GLU X11 c m pthread sndio stdc++ z
-MASTER_SITES= http://download.gna.org/warzone/releases/2.0/
-
-LIB_DEPENDS= SDL_net.>=1::devel/sdl-net \
+MODULES= devel/gettext
+LIB_DEPENDS= SDL.>=8::devel/sdl \
physfs.>=1::devel/physfs \
openal.>=1::audio/openal \
+ ogg.>=6::audio/libogg \
vorbis.>=5,vorbisfile.>=4::audio/libvorbis \
- mad.>=2::audio/libmad \
- jpeg.>=62::graphics/jpeg \
- png.>=5::graphics/png
+ theora::multimedia/libtheora \
+ png.>=5::graphics/png \
+ popt::devel/popt \
+ GLC::graphics/quesoglc
BUILD_DEPENDS= ::devel/bison \
::archivers/unzip \
::archivers/zip
-RUN_DEPENDS= :desktop-file-utils-*:devel/desktop-file-utils
+RUN_DEPENDS= :desktop-file-utils-*:devel/desktop-file-utils \
+ :dejavu-fonts-*:x11/dejavu-fonts
USE_GMAKE= Yes
USE_X11= Yes
-
-CONFIGURE_STYLE=gnu
-CONFIGURE_ARGS+=${CONFIGURE_SHARED}
-CONFIGURE_ENV+=CPPFLAGS='-I${LOCALBASE}/include
-I${LOCALBASE}/include/libpng \
- -I${X11BASE}/include' \
- LDFLAGS='-L${LOCALBASE}/lib -L${X11BASE}/lib' \
- LIBS='-pthread -lossaudio' \
- YACC='bison -y'
+CONFIGURE_STYLE= autoconf
+AUTOCONF_VERSION= 2.61
+CONFIGURE_ARGS+= --with-distributor="OpenBSD"
+CONFIGURE_ENV+=CPPFLAGS="-I${X11BASE}/include
-I${LOCALBASE}/include/libpng \
+ -I${LOCALBASE}/include" \
+ LDFLAGS="-L${X11BASE}/lib -L${LOCALBASE}/lib" \
+ YACC="bison -y" \
+ LEX="gflex"
.include
Index: distinfo
===
RCS file: /cvs/ports/games/warzone2100/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo22 Jul 2007 14:57:52 - 1.1.1.1
+++ distinfo14 Dec 2009 00:35:04 -
@@ -1,5 +1,5 @@
-MD5 (warzone2100-2.0.7.tar.bz2) = mDzMZ8/V3zGFxoNsYupyLw==
-RMD160 (warzone2100-2.0.7.tar.bz2) = qNSDZmq2YhtLP9jdl/xOlanZxWs=
-SHA1 (warzone2100-2.0.7.tar.bz2) = +4W+ejAWvSwAbKxkVTjaOn1wJdE=
-SHA256 (warzone2100-2.0.7.tar.bz2) =
5q4rNYTG8YlofL/HejQ5foIQy86rg3ERVcNllOmw+so=
-SIZE (warzone2100-2.0.7.tar.bz2) = 19622856
+MD5 (warzone2100-2.3_beta4.tar.gz) = sjvIORfQsnwpApIxICtZyA==
+RMD160 (warzone2100-2.3_beta4.tar.gz) = VbwLxUXM+dclXuIAPRnVHSpsC2M=
+SHA1 (warzone2100-2.3_beta4.tar.gz) = yJ4Gm2trIaWDSbpThKKdO5MtS1E=
+SHA256 (warzone2100-2.3_beta4.tar.gz) =
YUGiaIIuvLS5IFnPguN3dNhBTqw6xLFT9lRAD0inlnU=
+SIZE (warzone2100-2.3_beta4.tar.gz) = 50661004
Index: patches/patch-Makefile_in
===
RCS file: patches/patch-Makefile_in
diff -N patches/patch-Makefile_in
--- patches/patch-Makefile_in 22 Jul 2007 14:57:52 - 1.1.1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,12 +0,0 @@
-$OpenBSD: patch-Makefile_in,v 1.1.1.1 2007/07/22 14:57:52 simon Exp $
Makefile.in.orig Sun Jul 22 00:02:49 2007
-+++ Makefile.inSun Jul 22 00:03:05 2007
-@@ -199,7 +199,7 @@ target_alias = @target_alias@
- top_builddir = @top_builddir@
- top_srcdir = @top_srcdir@
- SUBDIRS = win32 lib src data icons nsis
--dist_doc_DATA = AUTHORS ChangeLog README COPYING COPYING.README
-+dist_doc_DATA = README
- ACLOCAL_AMFLAGS = -I m4
- EXTRA_DIST = \
- autogen.sh \
Index: patches/patch-configure
===
RCS file: patches/patch-configure
diff -N patches
ports that use audio but not with sndio
in case anyone was curious of the current state. ports that have AUDIO_SETINFO in the sources and don't have ossaudio in WANTLIB (probably are using audio(4)): audio/gqmpeg (mixer only) audio/libdca (example player) audio/libmikmod audio/libsamplerate (example player) audio/libsndfile (example player) audio/libworkman (has sun audio but no audio support built) audio/rplay (has sun audio but uses generic audio; sndio backend written) audio/soundtracker audio/speex (example player) audio/tracker audio/workman (has sun audio but no audio support built) comms/gmfsk devel/jdk/1.7 done: patches sent to MAINTAINER devel/mico devel/plib done: patches sent to MAINTAINER games/cgoban games/crossfire-client games/eboard games/xblast done: patches sent to ports@ multimedia/transcode x11/qvwm done: patches sent to ports@ x11/rdesktop x11/xmascot ports that have ossaudio in WANTLIB: audio/ac3dec (example player) audio/aumix (mixer) audio/bonk audio/cmu-sphinx3 audio/cmu-sphinxbase audio/festival/core audio/freealut done (uses audio/openal) audio/hydrogen (backend written, port needs update to latest release) audio/liba52 (example player) audio/mixer.app (mixer) audio/mp3blaster audio/mppenc (example player) audio/nosefart audio/nspmod audio/rsynth audio/sidplay audio/tosha audio/waveplay (will be removed from ports?) audio/wmix (mixer) audio/wmmixer (mixer) audio/xmix (mixer) audio/xmmix (mixer) comms/gpsk31 comms/tlf devel/jdk/1.5 (unportable API; needs a device filehandle) devel/jdk/1.6 done: patches sent to MAINTAINER editors/emacs21 editors/emacs22 emulators/qemu emulators/wine games/gtkpool games/quake2 (have patches to use sdl by default, need to revisit them) games/xboing games/xbreaky games/xscavenger graphics/ffmpeg graphics/flash graphics/fxtv graphics/synaesthesia math/freemat multimedia/avidemux multimedia/libquicktime multimedia/recordmydesktop/base sysutils/cdrtools sysutils/conky sysutils/gkrellm/plugins/volume sysutils/tpb telephony/pjsua (should probably just use portaudio-svn port) x11/kde/multimedia3 (only mixer parts left, iirc) x11/kde4/multimedia x11/kmplayer (port marked as BROKEN) x11/wxWidgets x11/xawtv about the ports where I put "(example player)". these are audio encoding/ decoding libraries that have code to play to an audio device. imo, kinda dumb. what's the point? to test decoder? then output to a WAV, or at least use libao or the like. these have a really low priority for me. ports marked "mixer" only use the mixer interface, and are also low priority. use mixerctl. if you are working on porting any of these, please let me know. testing of patches sent to ports@ greatly appreciated also. -- jake...@sdf.lonestar.org SDF Public Access UNIX System - http://sdf.lonestar.org
sndio for x11/qvwm
kill the esound FLAVOR too.
--
jake...@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org
Index: Makefile
===
RCS file: /cvs/ports/x11/qvwm/Makefile,v
retrieving revision 1.13
diff -N -u -p Makefile
--- Makefile7 Mar 2009 15:37:38 - 1.13
+++ Makefile17 Dec 2009 22:48:18 -
@@ -3,7 +3,7 @@
COMMENT= Windows 95/98/NT like window manager
DISTNAME= qvwm-1.1.8
-PKGNAME= ${DISTNAME}p1
+PKGNAME= ${DISTNAME}p2
CATEGORIES= x11
MASTER_SITES= ftp://ftp.qvwm.org/pub/qvwm/
@@ -14,9 +14,9 @@ PERMIT_PACKAGE_CDROM= Yes
PERMIT_PACKAGE_FTP=Yes
PERMIT_DISTFILES_CDROM=Yes
PERMIT_DISTFILES_FTP= Yes
-WANTLIB= ICE SM X11 Xext Xss c m stdc++
+WANTLIB= ICE SM X11 Xext Xss c m sndio stdc++
-FLAVORS= esound imlib
+FLAVORS= imlib
FLAVOR?=
USE_X11= Yes
@@ -24,17 +24,8 @@ USE_X11= Yes
SEPARATE_BUILD=flavored
CONFIGURE_STYLE= gnu
-# the build-scheme will get confused otherwise
-post-patch:
- rm ${WRKDIST}/src/yaccsrc.{cc,h}
-
-.if ${FLAVOR:L:Mesound}
-LIB_DEPENDS+= esd.>=2::audio/esound
-CONFIGURE_ARGS+= --with-esd-prefix=${LOCALBASE}
-WANTLIB+= audiofile sndio
-.else
CONFIGURE_ARGS+= --without-esd
-.endif
+CONFIGURE_ENV= CPPFLAGS="-DUSE_SNDIO" LIBS="-lsndio"
.if ${FLAVOR:L:Mimlib}
LIB_DEPENDS+= Imlib.>=19::graphics/imlib
@@ -44,5 +35,9 @@ WANTLIB+= tiff ungif z jpeg m png
CONFIGURE_ARGS+= --without-imlib
WANTLIB+= Xpm
.endif
+
+# the build-scheme will get confused otherwise
+post-patch:
+ rm ${WRKDIST}/src/yaccsrc.{cc,h}
.include
Index: patches/patch-src_Makefile_in
===
RCS file: /cvs/ports/x11/qvwm/patches/patch-src_Makefile_in,v
retrieving revision 1.1
diff -N -u -p patches/patch-src_Makefile_in
--- patches/patch-src_Makefile_in 3 Jul 2000 00:39:09 - 1.1
+++ patches/patch-src_Makefile_in 17 Dec 2009 22:48:18 -
@@ -1,7 +1,7 @@
$OpenBSD: patch-src_Makefile_in,v 1.1 2000/07/03 00:39:09 espie Exp $
src/Makefile.in.orig Mon Jul 3 02:22:01 2000
-+++ src/Makefile.inMon Jul 3 02:22:11 2000
-@@ -86,7 +86,7 @@ qvwm_SOURCES = button.cc button.h callba
+--- src/Makefile.in.orig Wed May 31 00:39:47 2000
src/Makefile.inWed Dec 16 20:53:13 2009
+@@ -86,7 +86,7 @@ qvwm_SOURCES = button.cc button.h callback.h colormap.
# VPATH is not enough for lexsrc.ll and yaccsrc.yy
Index: patches/patch-src_audio_cc
===
RCS file: /cvs/ports/x11/qvwm/patches/patch-src_audio_cc,v
retrieving revision 1.1
diff -N -u -p patches/patch-src_audio_cc
--- patches/patch-src_audio_cc 15 Jan 2004 18:25:36 - 1.1
+++ patches/patch-src_audio_cc 17 Dec 2009 22:48:18 -
@@ -1,30 +1,117 @@
$OpenBSD: patch-src_audio_cc,v 1.1 2004/01/15 18:25:36 espie Exp $
src/audio.cc.orig 2004-01-15 19:20:05.0 +0100
-+++ src/audio.cc 2004-01-15 19:20:15.0 +0100
-@@ -232,7 +232,7 @@ int Audio::setFormat(int bits, int encod
+--- src/audio.cc.orig Wed May 31 00:17:59 2000
src/audio.cc Thu Dec 17 01:55:40 2009
+@@ -32,6 +32,8 @@
+ // sound playback type definition
+ #if defined(USE_ESD)
+ #define ESD
++#elif defined(USE_SNDIO)
++#define SNDIO
+ #else
+
+ #if defined(__linux__) && defined(USE_ALSA)
+@@ -51,6 +53,8 @@
+ // include header file
+ #ifdef ESD
+ #include
++#elif defined(SNDIO)
++#include
+ #else
+
+ #ifdef __linux__
+@@ -165,6 +169,11 @@ int Audio::openDevice()
+ return 1;
+ #elif defined(ESD)
+ return 1;
++#elif defined(SNDIO)
++ if ((hdl = sio_open(NULL, SIO_PLAY, 0)) == NULL)
++return -1;
++ started = 0;
++ return 1;
+ #else
+ m_audiofd = open(m_audiodev, O_WRONLY);
+ if (m_audiofd < 0)
+@@ -184,6 +193,8 @@ int Audio::closeDevice()
+ if (!(m_audiofd < 0))
+ if (esd_close(m_audiofd) < 0)
+ return -1;
++#elif defined(SNDIO)
++ sio_close(hdl);
+ #else
+ if (close(m_audiofd) < 0)
+ return -1;
+@@ -222,6 +233,15 @@ int Audio::setFormat(int bits, int encoding)
+ return -1;
#endif
++#ifdef SNDIO
++ if (encoding != EN_LINEAR && encoding != EN_NONE)
++return -1;
++ sio_initpar(&par);
++ par.bits = bits;
++ par.sig = bits == 8 ? 0 : 1;
++ par.le = SIO_LE_NATIVE;
++#endif
++
+ #ifdef OSS
+ if (encoding != EN_LINEAR && encoding != EN_NONE)
+ return -1;
+@@ -232,7 +252,7 @@ int Audio::setFormat(int bits, int encoding)
+ #endif
+
#ifdef SUN
- struct audio_info_t ainfo;
+ audio_info_t ainfo;
AUDIO_INITINFO(&ainfo);
-@@ -271,7 +271,7 @@ int Audio::setChannels(int channels)
+@@ -270,8 +290,12 @@ int Audio::setChannels(int channels)
+ return -1;
#endif
++#ifdef SNDIO
++ par.pchan = channels;
++#endif
++
#ifdef SUN
- struct audio_info_t ainfo;
+ audio_info_t ainfo;
AUDIO_INITINFO(&ainfo);
ai
Re: (n to go) progress bar
On Thu, Dec 17, 2009 at 10:29:39PM +0100, Marc Espie wrote: > On Thu, Dec 17, 2009 at 12:26:42PM -0600, Marco Peereboom wrote: > > On Thu, Dec 17, 2009 at 07:00:08PM +0100, Marc Espie wrote: > > > On Thu, Dec 17, 2009 at 06:43:02PM +0100, frantisek holop wrote: > > > > on the internet, you are never alone ;] > > > > > > > > -f > > > Yeah, but Paul is our resident gypsy, so we routinely ignore him (and > > > doubly > > > so if we're rude frogs) > > > > I am on Paul's side here. So the evil capitalists are also against rude > > frogging one's eyes too. > > You're probably one of those guys who either: > - uses <10 packages $ pkg_info | wc -l 255 > - has a killer-fast connection to the nearest mirror. Rural cable, so hardly. Speaking of the reason why pkg_add -ui is annoying. It takes hours to update all my packages and when I get to my laptop in the morning it sits there waiting for me to answer y. Repeat like 10 times. > Everybody else out there is very happy to know that they still have > 190 packages to go through when doing an update that last for 3 hours. > Think about it. I like to know it but it is inaccurate. > If you don't like it, use -x, all the progress messages should go away. >
Re: (n to go) progress bar
On Thu, Dec 17, 2009 at 12:26:42PM -0600, Marco Peereboom wrote: > On Thu, Dec 17, 2009 at 07:00:08PM +0100, Marc Espie wrote: > > On Thu, Dec 17, 2009 at 06:43:02PM +0100, frantisek holop wrote: > > > on the internet, you are never alone ;] > > > > > > -f > > Yeah, but Paul is our resident gypsy, so we routinely ignore him (and doubly > > so if we're rude frogs) > > I am on Paul's side here. So the evil capitalists are also against rude > frogging one's eyes too. You're probably one of those guys who either: - uses <10 packages - has a killer-fast connection to the nearest mirror. Everybody else out there is very happy to know that they still have 190 packages to go through when doing an update that last for 3 hours. Think about it. If you don't like it, use -x, all the progress messages should go away.
Re: UPDATE: net/snort-2.8.5.1
On Wed, Dec 16, 2009 at 08:11:27PM +0100, Markus Lude wrote:
> On Wed, Dec 16, 2009 at 03:02:59PM +, Stuart Henderson wrote:
> > On 2009-10-25, Markus Lude wrote:
> > > You now need to add the correct library version number to
> > > libsf_engine.so in the dynamicengine config line in your snort.conf.
> > > This applies only to those which use an older already adjusted
> > > snort.conf. The one in the diff below already has it. Should I add a
> > > MESSAGE file for this?
> >
> > This is a bit of a problem, it means users will need to hand-edit their
> > config file every time they upgrade. Were you able to work out what changed
> > between releases that has made this necessary? dlopen(3) should be quite
> > capable of locating the library from the basename without specifying the
> > particular version...
> >
> > Alternatively, does it even make sense to give this a versioned filename?
> > Plugins for e.g. php, apache and others typically just use unversioned
> > names.
>
> Thanks for the feedback. I'll have a look at it.
Please test the new attached diff. Only changes are in snort.conf and
the Makefile.
Any hints on how to remove the versions from the libs? Maybe we could
try that later on...
Regards,
Markus
Index: Makefile
===
RCS file: /cvs/ports/net/snort/Makefile,v
retrieving revision 1.54
diff -u -p -r1.54 Makefile
--- Makefile16 Aug 2009 13:31:27 - 1.54
+++ Makefile17 Dec 2009 08:37:02 -
@@ -4,7 +4,7 @@ SHARED_ONLY = Yes
COMMENT = highly flexible sniffer/NIDS
-DISTNAME = snort-2.8.4.1
+DISTNAME = snort-2.8.5.1
CATEGORIES = net security
MASTER_SITES = http://dl.snort.org/snort-current/
@@ -19,22 +19,21 @@ PERMIT_DISTFILES_CDROM = Yes
PERMIT_DISTFILES_FTP = Yes
WANTLIB = c m pcap
-SHARED_LIBS = sf_engine 3.0 \
- sf_dce2_preproc 0.0 \
- sf_dcerpc_preproc 2.0 \
- sf_dns_preproc 3.0 \
- sf_ftptelnet_preproc 3.0 \
- sf_smtp_preproc 3.0 \
- sf_ssh_preproc 2.0 \
- sf_ssl_preproc 1.0 \
- _sfdynamic_example_rule 0.0 \
- _sfdynamic_preprocessor_example 0.0
+SHARED_LIBS = sf_engine 4.0 \
+ sf_dce2_preproc 1.0 \
+ sf_dcerpc_preproc 3.0 \
+ sf_dns_preproc 4.0 \
+ sf_ftptelnet_preproc 4.0 \
+ sf_smtp_preproc 4.0 \
+ sf_ssh_preproc 3.0 \
+ sf_ssl_preproc 2.0
USE_LIBTOOL = Yes
SEPARATE_BUILD = concurrent
CONFIGURE_STYLE = simple
CONFIGURE_ARGS += ${CONFIGURE_SHARED} \
+ --enable-ipv6 \
--enable-dynamicplugin
MAKE_FLAGS = mandir=${TRUEPREFIX}/man
Index: distinfo
===
RCS file: /cvs/ports/net/snort/distinfo,v
retrieving revision 1.18
diff -u -p -r1.18 distinfo
--- distinfo16 Aug 2009 13:31:27 - 1.18
+++ distinfo17 Dec 2009 08:37:02 -
@@ -1,5 +1,5 @@
-MD5 (snort-2.8.4.1.tar.gz) = Y/TnaulqLRM/THt0G61UWA==
-RMD160 (snort-2.8.4.1.tar.gz) = jzijfJ8Wm6uVKUbMq6JMktJnuIA=
-SHA1 (snort-2.8.4.1.tar.gz) = P/q7U2bAiJUv76sZqNwZzoTWJkU=
-SHA256 (snort-2.8.4.1.tar.gz) = 7n95DrPfTmFWp8hl8boihVOUyePBP9xX1gqGRyZ/wgk=
-SIZE (snort-2.8.4.1.tar.gz) = 4567713
+MD5 (snort-2.8.5.1.tar.gz) = savzqfo0hnIMmite/5IEFw==
+RMD160 (snort-2.8.5.1.tar.gz) = +/q0Xx19gVUWBDWS6rjPHMbsk9A=
+SHA1 (snort-2.8.5.1.tar.gz) = uXEFLN1LNSegYDhUlTED/prYpFs=
+SHA256 (snort-2.8.5.1.tar.gz) = reGw9K50/WI8Yz0otvFCkYd1GzWzaj+KDBl9IQS15a4=
+SIZE (snort-2.8.5.1.tar.gz) = 4715078
Index: patches/patch-etc_snort_conf
===
RCS file: /cvs/ports/net/snort/patches/patch-etc_snort_conf,v
retrieving revision 1.4
diff -u -p -r1.4 patch-etc_snort_conf
--- patches/patch-etc_snort_conf16 Aug 2009 13:31:27 - 1.4
+++ patches/patch-etc_snort_conf17 Dec 2009 08:37:02 -
@@ -1,19 +1,17 @@
$OpenBSD: patch-etc_snort_conf,v 1.4 2009/08/16 13:31:27 rui Exp $
etc/snort.conf.origWed Mar 11 14:22:03 2009
-+++ etc/snort.conf Wed Apr 8 12:47:17 2009
-@@ -78,7 +78,10 @@ var SNMP_SERVERS $HOME_NET
- # like this:
- #
- # portvar HTTP_PORTS 8081
--#
-+
+--- etc/snort.conf.origMon Oct 19 23:09:14 2009
etc/snort.conf Thu Dec 17 08:25:52 2009
+@@ -99,6 +99,9 @@ portvar HTTP_PORTS 80
+ # including the rules file twice is obsolete. See README.variables for more
+ # details.
+
+# Ports you run ssh servers on
+portvar SSH_PORTS 22
+
- # Ports you run web servers on
- portvar HTTP_PORTS 80
+ # Ports you want to look for SHELLCODE on.
+ portvar S
Re: (n to go) progress bar
On Thu, Dec 17, 2009 at 07:00:08PM +0100, Marc Espie wrote: > On Thu, Dec 17, 2009 at 06:43:02PM +0100, frantisek holop wrote: > > on the internet, you are never alone ;] > > > > -f > Yeah, but Paul is our resident gypsy, so we routinely ignore him (and doubly > so if we're rude frogs) I am on Paul's side here. So the evil capitalists are also against rude frogging one's eyes too.
Re: (n to go) progress bar
On Thu, 17 Dec 2009, Marc Espie wrote: > so if we're rude frogs) That's a pleonasm. -- Antoine
Re: (n to go) progress bar
On Thu, Dec 17, 2009 at 06:43:02PM +0100, frantisek holop wrote: > on the internet, you are never alone ;] > > -f Yeah, but Paul is our resident gypsy, so we routinely ignore him (and doubly so if we're rude frogs)
Re: (n to go) progress bar
hmm, on Thu, Dec 17, 2009 at 09:26:53AM +0100, Marc Espie said that > Turning it off should go with the progress meter, not the verbose mode (.e.g, > -x flag) obviously, verbosity is in the eye of the beholder. in the old times, it was in the eye of the modem :] but as -v has multiple levels, i think sans -v should be quite spartan. i alos think that the (n to go) estimates will have a constant appearance on the mailing list because of the newcomers to openbsd. > > when i am doing an update (full or partial), it kind of doesn't > > matter how many packages are there "to go", esp if it's an estimate > > at best.. i think all the chat should go to -v (verbose). > > You're very much alone in there, you know.. on the internet, you are never alone ;] -f -- latin is a real angina gluteus maximus.
UPDATE: games/sudoku-solver
-lm is needed after openmotif-2.3.2p0 update
regen plist
Index: ports/games/sudoku-solver/Makefile
===
RCS file: /pub/anoncvs/cvs/OpenBSD/ports/games/sudoku-solver/Makefile,v
retrieving revision 1.5
diff -u -p -w -b -u -r1.5 Makefile
--- ports/games/sudoku-solver/Makefile 15 Sep 2009 17:37:21 - 1.5
+++ ports/games/sudoku-solver/Makefile 17 Dec 2009 13:57:57 -
@@ -3,7 +3,7 @@
COMMENT= sudoku puzzle solver with cli, gui, and web ui
DISTNAME= sudoku-solver-1.0.1
-PKGNAME= ${DISTNAME}p0
+PKGNAME= ${DISTNAME}p1
CATEGORIES=games www x11
@@ -14,13 +14,14 @@ PERMIT_PACKAGE_CDROM= Yes
PERMIT_PACKAGE_FTP=Yes
PERMIT_DISTFILES_CDROM=Yes
PERMIT_DISTFILES_FTP= Yes
-WANTLIB= c Xt crypto iconv pthread z
+WANTLIB= c Xt crypto iconv pthread m z
MASTER_SITES= ${HOMEPAGE}
EXTRACT_SUFX= .tgz
USE_X11= Yes
USE_MOTIF= OpenMotif
+MAKE_FLAGS=LDFLAGS="-lm"
MODULES= devel/gettext
Index: ports/games/sudoku-solver/pkg/PLIST
===
RCS file: /pub/anoncvs/cvs/OpenBSD/ports/games/sudoku-solver/pkg/PLIST,v
retrieving revision 1.2
diff -u -p -w -b -u -r1.2 PLIST
--- ports/games/sudoku-solver/pkg/PLIST 22 Jul 2007 16:43:30 - 1.2
+++ ports/games/sudoku-solver/pkg/PLIST 17 Dec 2009 13:57:57 -
@@ -1,7 +1,7 @@
@comment $OpenBSD: PLIST,v 1.2 2007/07/22 16:43:30 mbalmer Exp $
-bin/sudoku-solver
-bin/xsudoku
-libexec/sudoku-handler
+...@bin bin/sudoku-solver
+...@bin bin/xsudoku
+...@bin libexec/sudoku-handler
@man man/cat1/xsudoku.0
@man man/cat6/sudoku-solver.0
share/sudoku/
ΔΥΟ ΣΕΜΙΝΑΡΙΑ ΠΩΛΗΣΕΩΝ ΜΕ ΕΛΕΥΘΕΡΗ ΕΙΣΟΔΟ!
Αγαπητές φίλες και φίλοι, http://www.moneyshowonline.com/system/assets/000/000/111/original_AD_AGGELAKHS_Money_09_EXPRESS_ATHENS_09.pdf Μια τελευταία υπενθύμιση για το Σάββατο 19 Δεκεμβρίου , ώρες 18:00 – 20:00 και Κυριακή 20 Δεκεμβρίου, ώρες 14:00 – 16:00, στο ξενοδοχείο HILTON για να παρακολουθήσετε σε δυο σεμινάρια ιδιαίτερης επικαιρότητας και αποτελεσματικότητας με ελεύθερη είσοδο. Με αυτό τον τρόπο θέλω να εκφράσω για άλλη μια φορά την ευγνωμοσύνη μου για την εμπιστοσύνη, τη συμμετοχή και το ενδιαφέρον που δείχνετε σε όλες μας τις εκπαιδευτικές δραστηριότητες. Η συμμετοχή σας θα με τιμήσει ιδιαίτερα, Άκης Αγγελάκης LLM, SAC Dip., Authorized Trainer & Counselor Εκδότης, Συγγραφέας, Σύμβουλος Επιχειρήσεων Αν δεν επιθυμείτε να λαμβάνετε την παραπάνω αλληλογραφία, παρακαλούμε πατήστε εδώ
Re: (n to go) progress bar
On Thu, Dec 17, 2009 at 09:26:53AM +0100, Marc Espie wrote: > On Wed, Dec 16, 2009 at 05:12:35PM +0100, frantisek holop wrote: > > hmm, on Mon, Dec 14, 2009 at 10:04:15AM +0100, Marc Espie said that > > > Please note that (n to go) is an estimate. In some cases, it can be off. > > > That said, I introduced an off-by-1 the other way round in pkg_add... > > > > i have noticed in pkg_add when doing a full upgrade > > the "n to go" to skip about crazily, sometimes from 96 to 89, etc. > > > > not to be an ass, but if it is an estimate, it should say so... > > i find "n to go" to be pretty empirical when it comes to software. > > I won't change it back, because I like it better this way, and I'm not > alone. It's much more useful than the previous progressmeter. I'm with holop here. I see no use for this on my machines. Might be of very much use for others though and its your work so you know best. > > Turning it off should go with the progress meter, not the verbose mode (.e.g, > -x flag) > > > when i am doing an update (full or partial), it kind of doesn't > > matter how many packages are there "to go", esp if it's an estimate > > at best.. i think all the chat should go to -v (verbose). > > You're very much alone in there, you know.. I do not fancy it much either. I think leaving it to -v would be best. But this is all imho. The point being he is not alone :-)
Re: NEW: security/pdfcrack
On 2009/12/17 12:06, Benoit Lecocq wrote: > Hi, > > A new port tested on amd64/sparc64. > > $cat pkg/DESCR > PDFCrack is a GNU/Linux (other POSIX-compatible systems should work too) > tool for recovering passwords and content from PDF-files. It is small, > command line driven without external dependencies. updated tgz; honour CC and INSTALL_STRIP, and don't require gmake. pdfcrack.tgz Description: application/tar-gz
NEW: security/pdfcrack
Hi, A new port tested on amd64/sparc64. $cat pkg/DESCR PDFCrack is a GNU/Linux (other POSIX-compatible systems should work too) tool for recovering passwords and content from PDF-files. It is small, command line driven without external dependencies. Comments ? OK ? Cheers, benoit pdfcrack.tar.gz Description: application/gzip
Re: (n to go) progress bar
On Thu, 17 Dec 2009 09:26:53 +0100, Marc Espie wrote: >You're very much alone in there, you know.. > >The fun thing is that, actually, the package tools are now chatting ways less >and giving more useful information than before. > +1! *** NOTE *** Please DO NOT CC me. I subscribed to the list. Mail to the sender address that does not originate at the list server is tarpitted. The reply-to: address is provided for those who feel compelled to reply off list. Thankyou. Rod/ --- This life is not the real thing. It is not even in Beta. If it was, then OpenBSD would already have a man page for it.
Re: [patch] graphics/xv - patch for right-side-up BMP data
a better patch (I think). On Sat, Dec 12, 2009 at 7:43 PM, patrick keshishian wrote: > Slight adjustment to patch. > > On Sat, Dec 12, 2009 at 7:24 PM, patrick keshishian > wrote: >> Hello, >> >> This patch lets xv load uncompressed BMP files with negative height >> value in the header. >> >> $ file rightsideup.bmp >> rightsideup.bmp: PC bitmap data, Windows 3.x format, 320 x -239 x 32 >> >> Google pointed me to wikipedia[0] on this subject. Evidently, if the >> height value is negative for uncompressed BMP files, it is an >> indication that the bitmap data is stored top to bottom[1]. >> >> Pre-patch, xv refuses to open these images saying that dimensions are >> too large. This patch addresses this issue (the patch is >> post-jumbo-patch applied). >> >> I have already forwarded this patch to Greg Roelofs who maintains the >> xv-Jumbo patch. >> >> I tested xv after my patch to make sure it didn't break anything using >> the "Test BMP images"[2] referenced from the wikipedia page[0]. All >> work except the bottom three which xv didn't like prior to the patch. >> >> Unfortunately, the "rightsideup.bmp" image, and friends, I ran into >> are copyrighted content which I do not have permission to >> redistribute. Also, ATM I don't know what tool was used to generate >> them. >> >> Comments? OK? >> >> Tested on macppc late November snapshot. >> >> Cheers, >> --patrick >> >> >> [0] http://en.wikipedia.org/wiki/BMP_file_format >> [1] From [0]: "Uncompressed Windows bitmaps can also be stored from >> the top row to the bottom, if the image height value is negative." >> [2] http://wvnvaxa.wvnet.edu/vmswww/bmp.html >> > xv-patch Description: Binary data
Re: (n to go) progress bar
On Wed, Dec 16, 2009 at 05:12:35PM +0100, frantisek holop wrote: > hmm, on Mon, Dec 14, 2009 at 10:04:15AM +0100, Marc Espie said that > > Please note that (n to go) is an estimate. In some cases, it can be off. > > That said, I introduced an off-by-1 the other way round in pkg_add... > > i have noticed in pkg_add when doing a full upgrade > the "n to go" to skip about crazily, sometimes from 96 to 89, etc. > > not to be an ass, but if it is an estimate, it should say so... > i find "n to go" to be pretty empirical when it comes to software. I won't change it back, because I like it better this way, and I'm not alone. It's much more useful than the previous progressmeter. Turning it off should go with the progress meter, not the verbose mode (.e.g, -x flag) > when i am doing an update (full or partial), it kind of doesn't > matter how many packages are there "to go", esp if it's an estimate > at best.. i think all the chat should go to -v (verbose). You're very much alone in there, you know.. The fun thing is that, actually, the package tools are now chatting ways less and giving more useful information than before.
