aboutsummaryrefslogtreecommitdiffstats
path: root/src (unfollow)
Commit message (Collapse)AuthorFilesLines
2010-09-25conf: Whitespace cosmeticsThomas Jansen1-9/+9
2010-09-23playlist: fix "queued" check in playlist_sync()Max Kellermann1-1/+1
The check was meant to fix an assertion failure, but it was the wrong way around. This broke cross-fading most of the time.
2010-09-23decoder/mp4ff: support more variations of "album artist"Max Kellermann1-0/+4
According to the mantis bug report 2847, there are several possible variations of the "album artist" tag: - "album artist" - "album_artist" - "albumartist" This patch adds support for the latter two.
2010-09-23listen: fix "unused parameter" warning on WIN32Max Kellermann1-0/+2
2010-09-23main: Add Windows Service supportDenis Krjuchkov5-1/+228
I've added PIPE_EVENT_SHUTDOWN because calling g_main_loop_quit() do not work when called from another thread. Main thread was sleeping in g_poll() so I needed some way to wake it up. By some strange reason call close(event_pipe[0]) in event_pipe_deinit() hangs. In current implementation that code never reached so that was not a problem :-) I've added a conditional to leave event_pipe[0] open on Win32.
2010-09-23playlist_state: Fix the "state" line in the outputThomas Jansen1-1/+1
An '\n' was erroneously inserted in the line containing the state, e.g. "state: \nplay" instead of "state: play". Fix for bug #2992.
2010-09-23zeroconf-bonjour: use g_htons() instead of htons()Max Kellermann1-1/+1
htons() is not available if netinet/in.h is not included.
2010-09-23directory: cast DEVICE_INARCHIVE, DEVICE_CONTAINER to dev_tMax Kellermann1-2/+2
Fix gcc warning.
2010-09-23playlist_database: initialize pm.mtimeMax Kellermann1-1/+3
2010-09-23fd_util: work around aliasing warning in recvmsg_cloexec()Max Kellermann1-2/+2
2010-09-23output/httpd: access sockaddr_storage object directlyMax Kellermann1-3/+2
Work around aliasing warning.
2010-09-23decoder_control: use g_free() to manage mixramp allocationsMax Kellermann3-20/+10
Be consistent with the rest of MPD, and don't use the non-portable header "malloc.h".
2010-09-07update_walk: update existing playlist entryMax Kellermann1-1/+3
Fixes duplicate playlist entries.
2010-09-07playlist_vector: update_or_add() returns boolMax Kellermann2-2/+10
False if the vector was not modified.
2010-08-31output/httpd: implement "pause"Max Kellermann1-0/+19
Send silence to all connected clients while paused, to avoid connection interruption.
2010-08-31output/httpd: forced flush after 32 kB of input dataMax Kellermann2-0/+23
Avoid buffer underruns on the streaming client, if the encoder is "too efficient" (e.g. when encoding silence while paused).
2010-08-23ReplayGain filter: allow gain > 100 %Johan Kiviniemi1-3/+11
The ReplayGain filter clamped the gain to max. 100 % even if the algorithm determined the signal needed a boost. That would result in any such tracks being played with too low volume, effectively defeating the purpose of the filter.
2010-08-03fd_util: add function dup_cloexec()Max Kellermann2-0/+17
Unfortunately, there's no "optimized" implementation here. We can't use Linux's proprietary system call dup3(), because it would require us to specify the new descriptor.
2010-08-03fd_util: add function recvmsg_cloexec()Max Kellermann2-0/+49
2010-08-03fd_util: add function socketpair_cloexec()Max Kellermann2-0/+35
2010-07-25playlist_song: fix memory leakMax Kellermann1-0/+2
Free the temporary path string in apply_song_metadata().
2010-07-25directory: free playlist vector in directory_free()Max Kellermann1-0/+2
2010-07-25queue_save: save tags and range of non-database songsMax Kellermann5-21/+63
Use the functions song_save() and song_load() to use the same format as in the database file for those songs which need the tags.
2010-07-25song_save: save start_ms and end_msMax Kellermann1-0/+11
While this is not useful for the database, it may become useful for reusing this code for the state file.
2010-07-25queue_save: simplify get_song()Max Kellermann1-10/+3
Don't try db_get_song() if the URI has a scheme.
2010-07-25{queue,song}_print: print relative paths if possibleMax Kellermann4-1/+22
If a song with an absolute path points inside the music directory, print only the relative part. This happens when partial songs from a playlist file were loaded.
2010-07-25queue_print: hide HTTP password in the "playlist" commandMax Kellermann1-5/+2
I've already changed the "playlistinfo" command to hide HTTP passwords, but forgot to do the same for the simpler "playlist" command. This patch changes queue_print_uris() to use the code from song_print_uri().
2010-07-25queue_save: move code to queue_save_song()Max Kellermann1-7/+11
2010-07-25queue_save: queue_load_song() returns voidMax Kellermann3-11/+7
The only caller doesn't use its return value, and the value isn't useful anyway.
2010-07-25song_save: song_load() returns NULL instead of falseMax Kellermann1-2/+2
2010-07-25state_file: use the text_file libraryMax Kellermann3-38/+42
Don't use a large stack buffer.
2010-07-25playlist_state: simplify printf() callsMax Kellermann1-22/+18
Let the C compiler concatenate string constants.
2010-07-25database, ...: remove EINTR checks after stdio callsMax Kellermann5-18/+18
MPD doesn't have child processes anymore, and thus we're not expecting to receive SIGCHLD very often. Since hard disk access isn't interrupted by signals anyway, we don't need those excessive checks.
2010-07-25directory_save: duplicate the playlist nameMax Kellermann1-2/+9
The function playlist_metadata_load() will overwrite the input buffer before using the "name" parameter; since "name" points to the same buffer, we'll get a corrupted string.
2010-07-23Added album tag support for libgme decoderRasi1-0/+2
2010-07-21update: store playlist files in databaseMax Kellermann9-31/+353
Don't open the music directory for each "lsinfo" call. Get the list of playlist files from the memory database.
2010-07-21song_save: pass const songvec to songvec_save()Max Kellermann2-2/+3
2010-07-21song_save: include cleanupMax Kellermann1-1/+0
2010-07-21database: remove outdated commentMax Kellermann1-1/+0
2010-07-20tag_rva2: move code to rva2_apply_data()Max Kellermann1-11/+16
2010-07-20tag_rva2: add struct rva2_dataMax Kellermann1-18/+39
.. and several accessor inline functions, to simplify the calling code.
2010-07-20tag_rva2: name the CHANNEL_* enumMax Kellermann1-12/+12
2010-07-20decoder/mad: move RVA2 code to tag_rva2.cMax Kellermann3-89/+151
2010-07-20replay_gain_info: fix indentationMax Kellermann1-10/+7
2010-07-20input/curl: remove assertion after curl_multi_fdset()Max Kellermann1-2/+0
Some users reported that MPD crashes when using a new CURL version with the threaded DNS resolver enabled. It seems that curl_multi_fdset() returns no file descriptor when the DNS resolver runs in another thread, so MPD does not have any event to wait for. On the CURL mailing list, somebody suggested to sleep for a fixed amount of time. This is not an elegant solution, because daemons should never have to sleep without waiting for an event. I hope the CURL developers will review the API and remove the threaded DNS resolver. Meanwhile, I'm removing the assertion in question, to allow those unfortunate users running the latest CURL version to continue using MPD.
2010-07-20input/curl: query timeout from CURLMax Kellermann1-1/+18
Use curl_multi_timeout() to determine the select() timeout, instead of hard-coding one second.
2010-07-20tag_rva2: set "gain", not "peak"Max Kellermann1-2/+2
RVA2 tags only store the "gain" value, there is no "peak" attribute.
2010-07-20decoder/mad: parse_rva2() returns boolMax Kellermann1-9/+8
2010-07-20decoder/wildmidi: support version 0.2.3Max Kellermann1-0/+4
In libwildmidi 0.2.3, the function WildMidi_SampledSeek() was removed, without changing the SO name. This patch adds an autoconf check for that function. Fall back to WildMidi_FastSeek() if WildMidi_SampledSeek() is not available anymore.
2010-07-19decoder/ffmpeg: fix libavformat 0.6 by using av_open_input_stream()Max Kellermann1-89/+65
libavformat 0.6 does not pass the original URI pointer to the "open" method, which leads to a crash because MPD was using a dirty hack to pass a pointer to that method. This patch switches to av_open_input_stream() with a custom ByteIOContext class, instead of doing the URI string hack with av_open_input_file(). Loosely based on a patch from Jasper St. Pierre.