libao issueshttps://gitlab.xiph.org/xiph/libao/-/issues2010-01-30T13:47:13Zhttps://gitlab.xiph.org/xiph/libao/-/issues/654LibAO should support format change without closing/opening devices.2010-01-30T13:47:13ZtruLibAO should support format change without closing/opening devices.To achive gapless playback between two files with diffrent samplerate you should really have the option to change the audio format without closing and opening the device. This is supported by most audio backends and should be a new api c...To achive gapless playback between two files with diffrent samplerate you should really have the option to change the audio format without closing and opening the device. This is supported by most audio backends and should be a new api call to libao.Monty MontgomeryMonty Montgomeryhttps://gitlab.xiph.org/xiph/libao/-/issues/651gaim crashes the sound system2007-06-17T08:54:16Zmstevens15gaim crashes the sound systemIn linux, I have to change sound daemons to get the
event sounds to even play. It won't coexist happily
with XMMS. Sometimes when I switch sound daemons, it
segfaults.
SEGFAULT:
Program received signal SIGSEGV, Segmentation fault.
[Swi...In linux, I have to change sound daemons to get the
event sounds to even play. It won't coexist happily
with XMMS. Sometimes when I switch sound daemons, it
segfaults.
SEGFAULT:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 16384 (LWP 2922)]
0x415fd33c in Arts::StartupManager::startup () from
/usr/lib/libmcop.so.1
(gdb) bt
#0 0x415fd33c in Arts::StartupManager::startup () from
/usr/lib/libmcop.so.1
#1 0x415e2ff4 in Arts::Dispatcher::Dispatcher () from
/usr/lib/libmcop.so.1
#2 0x41191c95 in arts_backend_init () from
/usr/lib/libartscbackend.so.0
#3 0x40bad756 in arts_init () from /usr/lib/libartsc.so.0
#4 0x40baaaa9 in ao_plugin_test () from
/usr/lib/ao/plugins-2/libarts.so
#5 0x4003054f in ?? () from /usr/lib/libao.so.2
#6 0x08152238 in ?? ()
#7 0x00000000 in ?? ()
#8 0xffffffff in ?? ()
#9 0x400334b4 in ?? () from /usr/lib/libao.so.2
#10 0x08152232 in ?? ()
#11 0x0810d9c1 in top_left_corner ()
#12 0xbfff8c78 in ?? ()
#13 0x400310a2 in ao_default_driver_id () from
/usr/lib/libao.so.2
#14 0x400310a2 in ao_default_driver_id () from
/usr/lib/libao.so.2
#15 0x080f2ed9 in gaim_gtk_sound_get_ui_ops ()
#16 0x0808521e in gaim_prefs_destroy ()
#17 0x405963b6 in g_cclosure_marshal_VOID__VOID ()
from /usr/lib/libgobject-2.0.so.0
#18 0x405846b6 in g_closure_invoke () from
/usr/lib/libgobject-2.0.so.0
#19 0x40595ec8 in g_signal_emit_by_name () from
/usr/lib/libgobject-2.0.so.0
#20 0x40594f4c in g_signal_emit_valist () from
/usr/lib/libgobject-2.0.so.0
#21 0x405951e6 in g_signal_emit () from
/usr/lib/libgobject-2.0.so.0
---Type <return> to continue, or q <return> to quit---
#22 0x402ba1f5 in gtk_menu_item_activate () from
/usr/lib/libgtk-x11-2.0.so.0
#23 0x402cfc57 in gtk_option_menu_get_history ()
from /usr/lib/libgtk-x11-2.0.so.0
#24 0x402ae6de in _gtk_marshal_BOOLEAN__BOXED ()
from /usr/lib/libgtk-x11-2.0.so.0
#25 0x40584949 in g_cclosure_new_swap () from
/usr/lib/libgobject-2.0.so.0
#26 0x405846b6 in g_closure_invoke () from
/usr/lib/libgobject-2.0.so.0
#27 0x40595925 in g_signal_emit_by_name () from
/usr/lib/libgobject-2.0.so.0
#28 0x40594d3a in g_signal_emit_valist () from
/usr/lib/libgobject-2.0.so.0
#29 0x405951e6 in g_signal_emit () from
/usr/lib/libgobject-2.0.so.0
#30 0x4039d8a7 in gtk_widget_send_expose () from
/usr/lib/libgtk-x11-2.0.so.0
#31 0x402ad1f2 in gtk_propagate_event () from
/usr/lib/libgtk-x11-2.0.so.0
#32 0x402ac046 in gtk_main_do_event () from
/usr/lib/libgtk-x11-2.0.so.0
#33 0x404a9815 in _gdk_events_queue () from
/usr/lib/libgdk-x11-2.0.so.0
#34 0x405dc6b2 in g_main_depth () from
/usr/lib/libglib-2.0.so.0
#35 0x405dd738 in g_main_context_dispatch () from
/usr/lib/libglib-2.0.so.0
#36 0x405dda70 in g_main_context_dispatch () from
/usr/lib/libglib-2.0.so.0
#37 0x405de013 in g_main_loop_run () from
/usr/lib/libglib-2.0.so.0
#38 0x402ab8f3 in gtk_main () from
/usr/lib/libgtk-x11-2.0.so.0
#39 0x080f8463 in main ()Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/648[PATCH] audio_out.c(243) : error C2065: 'PATH_MAX' : undeclared identifier2010-01-28T11:51:48ZGitlab Bot[PATCH] audio_out.c(243) : error C2065: 'PATH_MAX' : undeclared identifierC:\libao-0.8.6\src\audio_out.c(243) : error C2065: 'PATH_MAX' : undeclared identifier
C:\libao-0.8.6\src\audio_out.c(243) : error C2057: expected constant expression
C:\libao-0.8.6\src\audio_out.c(243) : error C2466: cannot allocate an a...C:\libao-0.8.6\src\audio_out.c(243) : error C2065: 'PATH_MAX' : undeclared identifier
C:\libao-0.8.6\src\audio_out.c(243) : error C2057: expected constant expression
C:\libao-0.8.6\src\audio_out.c(243) : error C2466: cannot allocate an array of constant size 0
C:\libao-0.8.6\src\audio_out.c(243) : error C2133: 'fullpath' : unknown size
config.c
With MSVS6, LIMITS.H only defines PATH_MAX if _POSIX_ is defined,
but libao-0.8.6\win32\include\dirent.h doesn't like _POSIX_,
so to fix this modify audio_out.c and change
#include <limits.h>
to
#define _POSIX_
#include <limits.h>
#undef _POSIX_
that's how I get it compiled Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/647all *.dsw and *.dsp files need win32 line endings (cr+lf)2010-01-30T13:45:54ZGitlab Botall *.dsw and *.dsp files need win32 line endings (cr+lf)microsoft visual studio requires all *.dsw and *.dsp to have win32 line endings,
otherwise you even attempt to compile libao on win32
Also, the sourcecode tarball doesn't even include the win32 directory
microsoft visual studio requires all *.dsw and *.dsp to have win32 line endings,
otherwise you even attempt to compile libao on win32
Also, the sourcecode tarball doesn't even include the win32 directory
Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/644libao crashes with arts2010-01-30T13:39:47ZGitlab Botlibao crashes with artsI maintain gaim for gentoo linux. I see one bug every other week about crashes when sending/receiving IMs. Gaim devs have confirmed this to be crashing when gaim attempts to play notification sounds through libao, which crashes when tr...I maintain gaim for gentoo linux. I see one bug every other week about crashes when sending/receiving IMs. Gaim devs have confirmed this to be crashing when gaim attempts to play notification sounds through libao, which crashes when trying to use arts.
The latest bug report I have is here: http://bugs.gentoo.org/show_bug.cgi?id=87348
Gaim devs said they just close their bugs from other distros so I don't have much other evidence to show you. I'm sure one of them may comment here now that we've found your tracker.Monty MontgomeryMonty Montgomeryhttps://gitlab.xiph.org/xiph/libao/-/issues/612ALSA plugin fixes2005-07-16T19:16:21ZClemensALSA plugin fixesThe attached patch makes various enhancements to libao's ALSA plugin:
* it isn't necessary to use a separate snd_pcm_access_mask_t variable to set the access mask;
* add a sample rate check (similar to the OSS plugin);
* return the co...The attached patch makes various enhancements to libao's ALSA plugin:
* it isn't necessary to use a separate snd_pcm_access_mask_t variable to set the access mask;
* add a sample rate check (similar to the OSS plugin);
* return the correct error code when calling snd_pcm_prepare() after an underrun;
* set the start_threshold to a positive value – using zero can result in an immediate underrun when the first write is very small;
* various comments were wrong or misleading.Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/611[PATCH] Current ao_sun.c does not support network environments2010-01-29T11:13:12Zosmail[PATCH] Current ao_sun.c does not support network environmentsSun has established environment variables to be used in determining the appropriate audio device. These environment variables were not being checked by src/plugins/sun/ao_sun.c, which simply hardcoded /dev/audio as the device.
I'm atta...Sun has established environment variables to be used in determining the appropriate audio device. These environment variables were not being checked by src/plugins/sun/ao_sun.c, which simply hardcoded /dev/audio as the device.
I'm attaching a context diff (as soon as I figure out how, or I'll mail it on request if I can't) which changes ao_sun.c so that it checks the environment. This updated ao_sun.c will:
1) use the UTAUDIODEV environment variable (used by the SunRay thin client), if set
2) try to use the AUDIODEV environment variable, if set
3) fall back to the hard-coded /dev/audio device if neither of the above two environment variables are set.
Also, I'm compiling libao-0.8.6, which isn't in the version list.Monty MontgomeryMonty Montgomeryhttps://gitlab.xiph.org/xiph/libao/-/issues/604not playing anything in musicpd2010-01-29T10:58:32Zmtenot playing anything in musicpdI've installed musicpd on freebsd 5.3 and libao came with it. But whenever i tried to play a file with musicpd i got an error "libao - OSS cannot get buffer size for device" in /var/log/mpd.error.
I'm using OSS from 4front-tech.com and ...I've installed musicpd on freebsd 5.3 and libao came with it. But whenever i tried to play a file with musicpd i got an error "libao - OSS cannot get buffer size for device" in /var/log/mpd.error.
I'm using OSS from 4front-tech.com and i have two sound cards (1 is integrated on nforce2 mobo and the other is terratec ewx24/96) and I've experienced the same problem on both. If you need more info please let me know.Monty MontgomeryMonty Montgomeryhttps://gitlab.xiph.org/xiph/libao/-/issues/490libao2: libao driver fails to start on ppc2010-01-29T10:59:49Zccheneylibao2: libao driver fails to start on ppc```
Package: libao2
Version: 0.8.4-1
Severity: grave
Tags: sid
Justification: renders package unusable
Latest version of libao2 on ppc makes it impossible to use any player
based on it (like mpg321 or ogg123) because it always gi...```
Package: libao2
Version: 0.8.4-1
Severity: grave
Tags: sid
Justification: renders package unusable
Latest version of libao2 on ppc makes it impossible to use any player
based on it (like mpg321 or ogg123) because it always gives the error
message :
Can't find a suitable libao driver. (Is device in use?)
Downgrading to 0.8.3-1 fixes the problem. I couldn't rebuild 0.8.4-1
because it has a build dependency on libartsc0-dev which is unavailable.
-- System Information:
Debian Release: testing/unstable
Architecture: powerpc
Kernel: Linux motoko 2.4.23-pre5-ben0 #1 mer oct 15 21:42:53 CEST 2003 ppc
Locale: LANG=fr_FR@euro, LC_CTYPE=fr_FR@euro
Versions of packages libao2 depends on:
ii libc6 2.3.2-8 GNU C Library: Shared libraries
an
-- no debconf information
```Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/487MPD/mpg321 doesn't work with this mp32006-06-12T11:39:50ZstarzMPD/mpg321 doesn't work with this mp3```
OK, there's no 'Version 0.8.4' for libao in bugzilla so I couldn't put it as
0.8.4, so I just set it as CVS.
There is a problem in mpg321 & MPD (http://musicpd.org) with playing this file:
http://thelinuxshow.com/archives/2003/tls-1...```
OK, there's no 'Version 0.8.4' for libao in bugzilla so I couldn't put it as
0.8.4, so I just set it as CVS.
There is a problem in mpg321 & MPD (http://musicpd.org) with playing this file:
http://thelinuxshow.com/archives/2003/tls-11-25-2003.mp3
1) ALSA09 + 0.8.4 doesn't work
2) OSS + 0.8.4 does work
3) ALSA09 + 0.8.3 does work
4) OSS + 0.8.3 does work
5) ALSA09 + 0.8.3 + ao_alsa09.c from 0.8.4 doesn't work.
6) ALSA09 + 0.8.4 + ao_alsa09.c from 0.8.3 doesn't work
After all of this, I really don't know what the problem could be.
```Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/466libao 0.8.4 doesn't compile with --disable-esd2006-06-12T11:22:15Zberolibao 0.8.4 doesn't compile with --disable-esd```
[0.8.4 is missing in the Version list on enter_bug.cgi; this is about final
0.8.4 despite the cvs mark]
Trying to compile with esd disabled results in
configure: error: conditional "HAVE_ESD" was never defined.
Usually this m...```
[0.8.4 is missing in the Version list on enter_bug.cgi; this is about final
0.8.4 despite the cvs mark]
Trying to compile with esd disabled results in
configure: error: conditional "HAVE_ESD" was never defined.
Usually this means the macro was only invoked conditionally.
```Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/437libao: WAV device doesn't print error when device is full2010-01-29T10:33:20Zccheneylibao: WAV device doesn't print error when device is full```
This is copied from Debian BTS bug #130816:
I used ogg123 to convert an OGG file to a WAV. However, my hard drive
was full. ogg123 didn't tell me this, but happily continued just as
if everything was working fine. In the end,...```
This is copied from Debian BTS bug #130816:
I used ogg123 to convert an OGG file to a WAV. However, my hard drive
was full. ogg123 didn't tell me this, but happily continued just as
if everything was working fine. In the end, I winded up with an empty
file. It would be nice if a "No space left on device" error would be
printed.
1097:tbm@fisch: ..iseducation_of_lauryn_hill] ogg123 -d wav -f ~/audio-test-2/
lh-005.wav 005-doo_wop_\(that_thing\).ogg
Device: WAV file output
Author: Aaron Holtzman <aholtzma@ess.engr.uvic.ca>
Comments: Sends output to a .wav file
Playing: 005-doo_wop_(that_thing).ogg
Title: Doo Wop (That Thing)
Artist: Lauryn Hill
Album: The Miseducation Of Lauryn Hill
Track number: 5
Done.
1098:tbm@fisch: ..iseducation_of_lauryn_hill] ls -l ~/audio-test-2/lh-005.wav
-rw-r--r-- 1 tbm tbm 0 Jan 25 20:23 /home/tbm/
audio-test-2/lh-005.wav
1099:tbm@fisch: ..iseducation_of_lauryn_hill] df ~/audio-test-2
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/hda7 1912360 1815216 0 100% /home
```Monty MontgomeryMonty Montgomeryhttps://gitlab.xiph.org/xiph/libao/-/issues/418cvs_nighly libao ao_aixs.c build problem2006-06-12T10:44:12Ztomppacvs_nighly libao ao_aixs.c build problem```
libao Makefile tries to build src/ao_aixs.c on my Solaris box. This looks
something which should happen only with AIX
make
Making all in src
Making all in plugins
Making all in oss
Making all in esd
Making all in arts
Making all i...```
libao Makefile tries to build src/ao_aixs.c on my Solaris box. This looks
something which should happen only with AIX
make
Making all in src
Making all in plugins
Making all in oss
Making all in esd
Making all in arts
Making all in alsa
Making all in alsa09
Making all in sun
Making all in irix
Making all in macosx
Making all in nas
source='ao_aixs.c' object='ao_aixs.lo' libtool=yes \
depfile='.deps/ao_aixs.Plo' tmpdepfile='.deps/ao_aixs.TPlo' \
depmode=none /bin/bash ../depcomp \
/bin/bash ../libtool --mode=compile cc -DPACKAGE_NAME=\"\"
-DPACKAGE_TARNAME=\"\" -DPACKAGE_VERSION=\"\" -DPACKAGE_STRING=\"\"
-DPACKAGE_BUGREPORT=\"\" -DPACKAGE=\"libao\" -DVERSION=\"0.8.3\"
-DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1
-DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1
-DHAVE_UNISTD_H=1 -DHAVE_DLFCN_H=1 -DHAVE_LIBPTHREAD=1 -DSIZEOF_SHORT=2
-DSIZEOF_INT=4 -DSIZEOF_LONG=4 -DHAVE_SYS_AUDIOIO_H=1 -DHAVE_SYS_AUDIO_H=1 -I.
-I. -I../include/ao -I../include
-DAO_PLUGIN_PATH=\"/usr/local/lib/ao/plugins-2\" -xO4 -fast -w -fsimple
-native -xcg92 -g -c -o ao_aixs.lo `test -f 'ao_aixs.c' || echo './'`ao_aixs.c
cc -DPACKAGE_NAME=\"\" -DPACKAGE_TARNAME=\"\" -DPACKAGE_VERSION=\"\"
-DPACKAGE_STRING=\"\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE=\"libao\"
-DVERSION=\"0.8.3\" -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1
-DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1
-DHAVE_INTTYPES_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DLFCN_H=1 -DHAVE_LIBPTHREAD=1
-DSIZEOF_SHORT=2 -DSIZEOF_INT=4 -DSIZEOF_LONG=4 -DHAVE_SYS_AUDIOIO_H=1
-DHAVE_SYS_AUDIO_H=1 -I. -I. -I../include/ao -I../include
-DAO_PLUGIN_PATH=\"/usr/local/lib/ao/plugins-2\" -xO4 -fast -w -fsimple -native
-xcg92 -g -c ao_aixs.c -KPIC -DPIC -o ao_aixs.lo
"ao_aixs.c", line 130: undefined symbol: audio_init
"ao_aixs.c", line 130: syntax error before or at: init
"ao_aixs.c", line 131: undefined symbol: audio_control
"ao_aixs.c", line 132: undefined symbol: audio_change
"ao_aixs.c", line 137: undefined symbol: init
"ao_aixs.c", line 137: undefined struct/union member: srate
"ao_aixs.c", line 138: undefined struct/union member: bits_per_sample
"ao_aixs.c", line 139: non-unique member requires struct/union object: channels
"ao_aixs.c", line 139: cannot access member of non-struct/union object
"ao_aixs.c", line 140: undefined struct/union member: mode
"ao_aixs.c", line 140: undefined symbol: AUDIO_PCM
"ao_aixs.c", line 141: undefined struct/union member: flags
"ao_aixs.c", line 141: undefined symbol: AUDIO_BIG_ENDIAN
"ao_aixs.c", line 141: undefined symbol: AUDIO_TWOS_COMPLEMENT
"ao_aixs.c", line 142: undefined struct/union member: operation
"ao_aixs.c", line 144: undefined symbol: AUDIO_INIT
"ao_aixs.c", line 149: undefined symbol: change
"ao_aixs.c", line 149: cannot access member of non-struct/union object
"ao_aixs.c", line 150: undefined struct/union member: volume
"ao_aixs.c", line 151: undefined struct/union member: monitor
"ao_aixs.c", line 151: undefined symbol: AUDIO_IGNORE
"ao_aixs.c", line 152: undefined struct/union member: input
"ao_aixs.c", line 153: undefined struct/union member: output
"ao_aixs.c", line 153: undefined symbol: AUDIO_OUTPUT_1
"ao_aixs.c", line 155: undefined symbol: control
"ao_aixs.c", line 155: undefined struct/union member: ioctl_request
"ao_aixs.c", line 155: undefined symbol: AUDIO_CHANGE
"ao_aixs.c", line 156: undefined struct/union member: position
"ao_aixs.c", line 157: undefined struct/union member: request_info
"ao_aixs.c", line 159: undefined symbol: AUDIO_CONTROL
"ao_aixs.c", line 164: undefined symbol: AUDIO_START
cc: acomp failed for ao_aixs.c
*** Error code 1
make: Fatal error: Command failed for target `ao_aixs.lo'
Current working directory /v/t/src/ogg/build/ao/src
*** Error code 1
make: Fatal error: Command failed for target `all-recursive'
Current working directory /v/t/src/ogg/build/ao/src
*** Error code 1
make: Fatal error: Command failed for target `all-recursive'
```Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/310config files not closed2003-08-07T18:45:05Zfaceprintconfig files not closed```
read_config_file does not close the file that it opens
``````
read_config_file does not close the file that it opens
```Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/260if dlopen fails, the next plugin will also fail.2003-08-31T18:10:02Zd95hjortif dlopen fails, the next plugin will also fail.```
The call to dlopen fails to give a handle then the next time _get_plugin is
called it will fail.
This is because dlerror() is never called after the failing dlopen().
Insted the error messag is returned in the test for the first symb...```
The call to dlopen fails to give a handle then the next time _get_plugin is
called it will fail.
This is because dlerror() is never called after the failing dlopen().
Insted the error messag is returned in the test for the first symbol in the next
plugin.
dt->functions->test = dlsym(dt->handle, "ao_plugin_test");
if (dlerror()) { free(dt->functions); free(dt); return NULL; }
```Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/258device is not compared with NULL in ao_play()2003-08-26T23:16:58Zchristianbieredevice is not compared with NULL in ao_play()```
This bug report is for libao-0.8.3. The "Version" selector of your Bugzilla is
out of date.
ogg123 crashes on Solaris if /dev/audio doesn't exist. The crash happends in
ao_play() line 631 of src/audio_output.c:
if (device->s...```
This bug report is for libao-0.8.3. The "Version" selector of your Bugzilla is
out of date.
ogg123 crashes on Solaris if /dev/audio doesn't exist. The crash happends in
ao_play() line 631 of src/audio_output.c:
if (device->swap_buffer != NULL) {
IMO there's a missing "if (device == NULL) return 0;" although ogg123 should not
call ao_play() with device == NULL.I would also insert NULL checks in ao_close()
just because someone could accidently call it with device == NULL.
BTW, in doc/ao_play.html "ao_play" is wrongly called "ao_open".
```Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/244libao doesn't understand MacOS X dynamic libraries2007-06-17T08:52:27Zscrewtapelibao doesn't understand MacOS X dynamic libraries```
I'm using Mac OS X 10.1.5, and fink from cvs. The standard fink package of libao
doesn't allow any output besides .wav, .raw and null, despite the fact that the
libao tarball includes an ESD driver (and I have ESD installed, madplay ...```
I'm using Mac OS X 10.1.5, and fink from cvs. The standard fink package of libao
doesn't allow any output besides .wav, .raw and null, despite the fact that the
libao tarball includes an ESD driver (and I have ESD installed, madplay and xmms
use it happily) and an OS X CoreAudio driver.
So I downloaded libao-0.8.3 from the Vorbis site and compiled it - exactly the
same situation. I began to poke around in the code.
It seems libao has some issues with Mac OS X's implementation of dynamic
libraries. I have the fink dlcompat library installed, which mostly does the
right thing, but there's two places that stuff up:
* In src/audio_out.c at line 43, "SHARED_LIB_EXT" is defined as ".so". On Mac OS
X, shared libraries end in ".dylib". Oddly, the libao plugins are built and
installed as files ending in ".dylib", but libao ignores them because they have
the wrong extension.
* At about line 112 and following of the same file, libao has dlopened the file
in question and begins poking around for proper symbols to be defined. By
default all of these fail - libao looks for functions like "ao_plugin_test",
while the library exports "_ao_plugin_test".
After fixing both of the above in a highly hackish manner, recompiling and
installing libao, I now have the ESD plugin working properly. and ao_example.c
behaves as appropriate. I won't attach a patch, though, because I'm sure that my
changes would break every other platform out there.
```https://gitlab.xiph.org/xiph/libao/-/issues/240dlopen() fails upon loading plugins2006-06-12T11:25:17ZMichael Ritzertdlopen() fails upon loading plugins```
1) in contrast to my entry above this report concerns libao 0.8.3, not
0.8.2. bugzilla did not offer me to select this version. PLease fix the
corresponding form.
bug report:
after building ogg123 on my solaris7/sparc box i n...```
1) in contrast to my entry above this report concerns libao 0.8.3, not
0.8.2. bugzilla did not offer me to select this version. PLease fix the
corresponding form.
bug report:
after building ogg123 on my solaris7/sparc box i noticed that ogg123 showed
up only with the default drivers. i built the example program in libao's
doc subdirectory and noticed in a gdb session that the calls to dlopen()
failed. A research with google showed that the same problem has been
reported earlier for OpenBSD and for NetBSD. So i enforced the flag to the
dlopen() call to
#define DLOPEN_FLAG (RTLD_LAZY)
and the example program worked.
some thoughts to fix this:
the dlopen() problem occurs on OpenBSD, NetBSD, and possibly FreeBSD, and
it occurs on Solaris7 (and hence probably on most other svr4 systems)
On solaris boxes and the other commercial unices You have to deal with at
least two compilers.
So i suggest to enter a check into the configure script which tests for the
need of RTLD_LAZY or at least the system type and sets a preprocessor
symbol of, say, -DNEEDS_RTLD_LAZY according to the result of the test (or
the system type)
ao_private.h will be changed to check for NEEDS_RTLD_LAZY instead of OpenBSD
Thanks for writing this whole stuff!
Michael Ritzert
```Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/238FILENAME_MAX only 14 characters on HP-UX2006-06-12T11:25:58ZAlexander RossFILENAME_MAX only 14 characters on HP-UX```
HP-UX fails because in audio_out.c the code uses FILENAME_PATH. I think
either PATH_MAX from limits.h or use pathconf.
Also, SHARED_LIB_EXTENSION isn't set automatically. I'm not sure what
can be done without explicit
#ifdef _HPU...```
HP-UX fails because in audio_out.c the code uses FILENAME_PATH. I think
either PATH_MAX from limits.h or use pathconf.
Also, SHARED_LIB_EXTENSION isn't set automatically. I'm not sure what
can be done without explicit
#ifdef _HPUX_SOURCE
#endif
code.
Also, dlopen doesn't seem to work but hacking shl_open does. I don't really
know if it is worth the effort.
Cheers,
Ross
```Stan SeibertStan Seiberthttps://gitlab.xiph.org/xiph/libao/-/issues/219libao rpm doesn't build/work on distributions w/o alsa (i.e. redhat)2006-06-12T11:40:57Znoalibao rpm doesn't build/work on distributions w/o alsa (i.e. redhat)```
The file libao.spec included in the libao distribution contains a
build-dependency on alsa-lib-devel, a package that is unavailable in several
major distributions. This prevents the libao package from being built and used
on for exam...```
The file libao.spec included in the libao distribution contains a
build-dependency on alsa-lib-devel, a package that is unavailable in several
major distributions. This prevents the libao package from being built and used
on for example redhat.
The src and binary rpms on www.vorbis.com also needs to be updated once this is
fixed.
```Stan SeibertStan Seibert