Currently, if we start decoding while the pause flag is set, we open the
audio device and leave it opened, blocking other apps from using it. The
obvious thing to do is to not open the audio device if the pause flag is
set, but the open call also sets the audio format. Therefore I'm leaving
the open call in, and just closing it immediately afterwards if the pause
flag is set.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6745 09075e82-0dd4-0310-85a5-a0d7c8717e4f
The shout plugin will now feign playback until the connect timeout is hit,
preventing connection attempts from blocking playback on local outputs.
Note that this patch is very different from remiss' original one.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6738 09075e82-0dd4-0310-85a5-a0d7c8717e4f
Previously, the warning log was only flushed if creating the db or logging
to stdout. This meant that under normal circumstances (no db creation,
logging to files) the warning log was never flushed. This caused a bug
when a warning was printed for each call to the status command where the
warning buffer would grow endlessly, eventually using more and more CPU to
reallocate it.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6660 09075e82-0dd4-0310-85a5-a0d7c8717e4f
Turns out the fix was as simple as specifying the OPEN_TAGS flag when
opening the file. Thanks again to Kodest for figuring this one out.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6657 09075e82-0dd4-0310-85a5-a0d7c8717e4f
This ReplayGain code is currently disabled because WavpackGetTagItem can't
seem to find replaygain_* fields in APEv2 tags (which is how wvgain stores
ReplayGain values). Additionally, because APEv2 tags are stored at the end
of the file, this code is only implemented for regular files and not HTTP
streams. Using HTTP seeking it *may* be possible to implement it for both.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6656 09075e82-0dd4-0310-85a5-a0d7c8717e4f
Only wavpack implements both fileDecodeFunc and streamDecodeFunc, and it's
fileDecodeFunc provides more functionality. So try using that first.
This commit also fixes a bug where the plugin test loop wouldn't break once
a suitable plugin was found if it used fileDecodeFunc.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6655 09075e82-0dd4-0310-85a5-a0d7c8717e4f
if the clock ticks right after we get the start time and the timeout is
only one second, we'll still wait a full second instead of returning
immediately.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6557 09075e82-0dd4-0310-85a5-a0d7c8717e4f
pretending to play while we wait for the connection to timeout. This
removes the need for timers, and thus removes the now unnecessary
timer_get_runtime_* function(s) from the timer code.
The changes made compared to the pre-patch shout plugin are:
* Block while connecting, timing out after 2 seconds.
* Close the device, and not just the connection, if play returns -1.
* Remove sd->last_err (it's always assigned before use).
* Some minor cleanups.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6555 09075e82-0dd4-0310-85a5-a0d7c8717e4f
outputs, which is actually desired behaviour. This way if the shout server
takes a while to respond, the shout output can block until connected
without messing up other audio outputs.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6554 09075e82-0dd4-0310-85a5-a0d7c8717e4f
leave it in that state. Likewise, if an audio output is in state
DEVICE_ON, and reopening the device due to a format change fails, change it
to state DEVICE_ENABLE. This will prevent flushAudioBuffer from even
attempting to play audio on a closed device (even though it would fail
anyway).
git-svn-id: https://svn.musicpd.org/mpd/trunk@6529 09075e82-0dd4-0310-85a5-a0d7c8717e4f
top depending on !quit, which doesn't set it anywhere before the if (quit)
block is reached, and the inner one which doesn't set quit at all. Since
it's a local variable and can't be modified externally, it'll never be hit.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6524 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Wait ten seconds before declearing the shout server unreachable
* Fix a state where it would never attempt to connect if it had previously failed
It isn't perfect yet, but I'd like some testing on it from other setups
git-svn-id: https://svn.musicpd.org/mpd/trunk@6523 09075e82-0dd4-0310-85a5-a0d7c8717e4f
state (when users press stop, previous snd_pcm_drop(), then
snd_pcm_drain() was called. this would lockup dmix)
git-svn-id: https://svn.musicpd.org/mpd/trunk@6517 09075e82-0dd4-0310-85a5-a0d7c8717e4f
completely stopped. Instead, send them SIGSTOP to pause the process until
they're needed again. Then send them SIGCONT instead of re-spawning them.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6485 09075e82-0dd4-0310-85a5-a0d7c8717e4f
some versions of shoutcast send "content-type" in all lowercase, and I
don't trust other servers to get the case right for the rest of the headers
we look for.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6482 09075e82-0dd4-0310-85a5-a0d7c8717e4f
have any effect until the aac and mp4 input plugins actually support a
stream decoding API.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6481 09075e82-0dd4-0310-85a5-a0d7c8717e4f
been redirected. This prevents zeroconf from blocking daemonization, and
makes sure any errors get sent to the logs and not stdout.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6477 09075e82-0dd4-0310-85a5-a0d7c8717e4f
header. While this is odd for an HTTP header, it's actually quite common
for streaming clients to send it without a space. Some clients do send
with a space as well, but without one has always worked fine and may in
fact be more compatible.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6472 09075e82-0dd4-0310-85a5-a0d7c8717e4f
silence a warning about an unused variable without using stupid checks for
HAVE_AVAHI || HAVE_BONJOUR.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6471 09075e82-0dd4-0310-85a5-a0d7c8717e4f
I couldn't test mDNSResponder support on Linux, as Debian doesn't include it - but should work as well.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6453 09075e82-0dd4-0310-85a5-a0d7c8717e4f
playback is stopped completely. This means the player process will no
longer have to wake up 100 times per second to see if it's been told to
start playing (the main process will just spawn a new player process when
it needs to). On the downside, this means an extra pair of forks() and the
re-initializing of the player and decode processes each time playback is
restarted.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6446 09075e82-0dd4-0310-85a5-a0d7c8717e4f
one now, and trying to call NULL was causing a segfault at exit.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6398 09075e82-0dd4-0310-85a5-a0d7c8717e4f
uninitialized variables and non-returning functions that return. Let's
tell it to stfu.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6277 09075e82-0dd4-0310-85a5-a0d7c8717e4f
because lsr may return less than the input buffer size, and the rest of the
audio code needs to know the new size. This fixes the clicking that was
introduced with recent changes to the lsr code. A huge thanks to remiss
for figuring this out.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6273 09075e82-0dd4-0310-85a5-a0d7c8717e4f
relative paths in the DB or URLs. The main functional difference is that
playlistmove and playlistdelete will rewrite playlists in the correct
encoding and remove invalid lines instead of potentially modifying them.
The specific changes are:
appendSongToStoredPlaylist:
* Don't convert to FS charset
* Don't prepend music_directory if saving absolute paths
writeStoredPlaylistToPath:
* Convert to FS charset
* Prepend music_directory if saving absolute paths
loadStoredPlaylist:
* Make sure each line is either in the DB or a URL
loadPlaylist:
* Don't bother checking paths, since it's done in loadStoredPlaylist now
git-svn-id: https://svn.musicpd.org/mpd/trunk@6266 09075e82-0dd4-0310-85a5-a0d7c8717e4f
audio at once, so it won't work for us. The old full API code was still
heavily broken, as each call to pcm_convertSampleRate() used the same
state, even if it was processing two streams of audio. The new code keeps
a separate state for each audio stream that's being converted.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6255 09075e82-0dd4-0310-85a5-a0d7c8717e4f
number of channels is specified when the converter state is created.
Previously this was only done once, thus breaking horribly when the input
audio suddenly had a different channel count. A new state could be created
every time the number of channels changes, but this could happen many times
a second if resampling to two different formats at once. The simple API
doesn't have this problem, as channel count is an argument to the
conversion function itself.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6229 09075e82-0dd4-0310-85a5-a0d7c8717e4f
and samplerate conversion. This makes the code much easier to read, and
fixes a few bugs that were previously there.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6224 09075e82-0dd4-0310-85a5-a0d7c8717e4f
playlistadd, playlistdelete, etc. and would've caused the playlist to be
rewritten only up to the line with the error.
git-svn-id: https://svn.musicpd.org/mpd/trunk@6133 09075e82-0dd4-0310-85a5-a0d7c8717e4f
returning a list of matching songs, the number of results and total play
time of the results are returned.
git-svn-id: https://svn.musicpd.org/mpd/trunk@5950 09075e82-0dd4-0310-85a5-a0d7c8717e4f
db file is written. So don't try to set directory_dbModTime to the mtime
of the db file, since it will be incorrect.
git-svn-id: https://svn.musicpd.org/mpd/trunk@5945 09075e82-0dd4-0310-85a5-a0d7c8717e4f
NULL. Doing so would mean future calls to commandError with a socket as an
argument will still write the error message to the error log, and not the
socket.
git-svn-id: https://svn.musicpd.org/mpd/trunk@5891 09075e82-0dd4-0310-85a5-a0d7c8717e4f
message and trailing new line to STDERR_FILENO along with the ACK, instead
of sending them over the socket.
git-svn-id: https://svn.musicpd.org/mpd/trunk@5890 09075e82-0dd4-0310-85a5-a0d7c8717e4f
