aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* conf: handle fatal errors with GErrorMax Kellermann2009-09-247-58/+136
| | | | | | | Don't call g_error(), which will abort the process and dump core. This patch does not affect all the config_get_X() functions. These need some more refactoring.
* conf: splitted function config_param_free()Max Kellermann2009-09-241-4/+11
|
* cmdline: handle fatal errors with GErrorMax Kellermann2009-09-243-5/+30
| | | | Don't call g_error(), which will abort the process and dump core.
* listen: handle fatal errors with GErrorMax Kellermann2009-09-243-10/+29
| | | | Don't call g_error(), which will abort the process and dump core.
* update: splitted update.c into several sourcesMax Kellermann2009-09-246-770/+991
|
* update: pass const string to update_enqueue()Max Kellermann2009-09-243-14/+20
| | | | | Duplicate the path string within update.c, do not expect an allocated string as parameter.
* update: renamed directory_update_init() to update_enqueue()Max Kellermann2009-09-244-4/+4
|
* decoder/sidplay: free songlength data blob in error handlerMax Kellermann2009-09-241-3/+4
| | | | | When parsing the songlength database fails, the code forgot to free the memory allocated by the raw file data.
* decoder/sidplay: moved code to sidplay_load_songlength_db()Max Kellermann2009-09-241-29/+36
|
* decoder/sidplay: free GError objectsMax Kellermann2009-09-241-0/+2
| | | | | The caller is responsible fro freeing GError objects. That g_error_free() call was missing in two places.
* decoder/sidplay: pass GError** to g_file_get_contents()Max Kellermann2009-09-241-1/+2
| | | | | The error handler dereferences GError*, but did not retrieve that object from g_file_get_contents().
* cmdline: removed options --create-db and --no-create-dbMax Kellermann2009-09-206-35/+7
| | | | | Both options are deprecated, and should not be used anymore. Many users get confused by their presence.
* output/osx: fix the OS X 10.6 buildPatrik Weiskircher2009-09-203-6/+3
| | | | Include CoreServices/CoreServices.h.
* Merge branch 'v0.15.x'Max Kellermann2009-09-104-36/+36
|\ | | | | | | | | | | | | Conflicts: NEWS configure.ac doc/mpdconf.example
| * doc/protocol: clarify "update"Max Kellermann2009-09-101-13/+8
| | | | | | | | | | The update job id is positive. When used in a command list, MPD prints one job id for each "update" command.
| * doc/mpdconf.example: additional comment signs for some parametersMax Kellermann2009-09-101-18/+18
| | | | | | | | | | Add a second column of comment signs on some parameters which users shouldn't usually set, unless they know what they're doing.
| * decoder/vorbis: revert "faster tag scanning with ov_test_callback()"Max Kellermann2009-09-102-1/+3
| | | | | | | | | | This patch made ov_time_total() unusable, and MPD did not know the duration of songs.
| * Modify version string to post-release version 0.15.4~gitAvuton Olrich2009-08-292-1/+4
| |
* | INSTALL: mention libwavpack and OpenALMax Kellermann2009-09-091-0/+6
| |
* | output/openal: support OpenAL plugin on Mac OS XSerge Ziryukin2009-09-072-3/+19
| |
* | doc: documented the "openal" output plugin settingsSerge Ziryukin2009-09-062-0/+34
| |
* | output/openal: fix default device nameSerge Ziryukin2009-09-061-3/+3
| |
* | configure.ac: use /Developer/SDKs/MacOSX10.5.sdkPatrik Weiskircher2009-09-061-0/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | First, this is not a nice fix. I am also not sure why the error happens in the first place. I assume Apple deprecated some stuff in 10.6 + x86_64. My patch simply uses the 10.5 SDK if compiling on 10.6 Snow Leopard - which is installed by default if you install XCode 3.2 that comes with Snow Leopard. The reason this is not a nice fix is, of course, that this doesn't fix the problem. It just "postpones" it to the next release of MacOS X. But, some people may need it, and its better than nothing.
* | openal output pluginSerge Ziryukin2009-09-064-0/+300
| |
* | Merge branch 'master' of git://git.musicpd.org/avuton/mpdMax Kellermann2009-09-021-7/+4
|\ \
| * | doc: Unify position of footnotes and cleanup wording.Avuton Olrich2009-08-301-4/+4
| | |
| * | doc: Remove superfluous comment already described in footnotes.Avuton Olrich2009-08-301-3/+0
| | |
* | | decoder/sidplay: support seekingMike Dawson2009-08-302-3/+35
| | |
* | | decoder/sidplay: implemented songlength databaseMike Dawson2009-08-304-2/+117
| | | | | | | | | | | | [mk: added autoconf test; fixed songlen_data_size type]
* | | decoder/sidplay: subtunesMike Dawson2009-08-302-8/+126
| | |
* | | Merged release 0.15.3 from branch 'v0.15.x'Max Kellermann2009-08-305-2/+30
|\ \ \ | |/ / |/| / | |/ | | | | | | Conflicts: NEWS configure.ac
| * mpd version 0.15.3release-0.15.3Avuton Olrich2009-08-292-2/+2
| |
| * Document nextsong and nextsongid.Anton Khirnov2009-08-241-0/+18
| |
| * decoder/vorbis: faster tag scanning with ov_test_callback()Rasmus Steinke2009-08-242-1/+3
| | | | | | | | | | | | | | | | | | using ov_test_callback with function CALLBACKS_STREAMONLY will cause scanning to stop after the comment field. ov_open (and ov_test) default to CALLBACKS_DEFAULT which scans the file structure causing a huge slowdown. The speed improvement is huge: It scanned my files around 10x faster This procedure has been recommended by monthy (main vorbis developer) and was said to be safe for scanning files.
| * update: don't re-read unchanged container filesIgor Kuzmin2009-08-192-1/+3
| | | | | | | | | | | | | | MPD checks if every flac (possibly other types as well) file contains cuesheet on every update, which produces unneeded I/O. My music collection is on NFS share, so it's quite noticeable. IMHO, it shouldn't re-read unchanged files, so I wrote simple patch to fix it.
| * output_init: initialize the "pause" flagMax Kellermann2009-08-182-0/+3
| | | | | | | | Fix stuttering due to uninitialized variable.
| * Modify version string to post-release version 0.15.3~gitAvuton Olrich2009-08-152-1/+4
| |
* | decoder/mpg123: new decoder plugin based on libmpg123Max Kellermann2009-08-266-0/+243
| | | | | | | | | | | | | | | | Still missing: - seeking - tags - streaming - encodings other than MPG123_ENC_SIGNED_16
* | command: add "findadd" command.Anton Khirnov2009-08-255-0/+69
| |
* | decoder/vorbis: open file in "binary" modeGunnar Roth2009-08-251-1/+1
| | | | | | | | | | fopen change for win32 in tag dup of vorbis win32 adaption for head changes
* | output/recorder: new output plugin for recording radio streamsMax Kellermann2009-08-247-2/+338
| | | | | | | | | | | | | | | | | | The recorder plugin writes audio played by MPD to a file. This may be useful for recording radio streams. This implementation is incomplete, because support for tags is missing, and MPD should be able to record each track to a different file.
* | conf: removed the deprecated "error_file" optionMax Kellermann2009-08-243-2/+1
| | | | | | | | | | This option was deprecated by the 0.15 release. This patch makes this option invalid.
* | Merged release 0.15.2 from branch 'v0.15.x'Max Kellermann2009-08-1511-73/+159
|\| | | | | | | | | | | | | Conflicts: NEWS configure.ac
| * mpd version 0.15.2release-0.15.2Avuton Olrich2009-08-152-2/+2
| |
| * output/shout: minimize the unpause latencyMax Kellermann2009-08-142-0/+8
| | | | | | | | | | During the pause loop, manually sleep for 500ms if shout_delay() returns a value greater than that. Don't exhaust libshout's buffer.
| * output: fixed shout stuck pause bugMax Kellermann2009-08-144-0/+22
| | | | | | | | | | | | Explicitly make the output thread leave the ao_pause() loop. This patch is a workaround, and the "pause" flag is not managed in a thread-safe way, but that's good enough for now.
| * directory: free empty directories after removing them (memleak)Max Kellermann2009-08-142-3/+9
| | | | | | | | | | dirvec_delete() does not free the object, we have to call directory_free() afterwards.
| * update: free temporary string in container scan (memleak)Max Kellermann2009-08-142-2/+7
| | | | | | | | The return value of map_directory_child_fs() must be freed.
| * decoder/flac: don't allocate cuesheet twice (memleak)Max Kellermann2009-08-142-4/+7
| | | | | | | | | | | | | | The function flac_cue_track() first calls FLAC__metadata_object_new(), then overwrites this pointer with FLAC__metadata_get_cuesheet(). This allocate two FLAC__StreamMetadata objects, but the first pointer is lost, and never freed.
| * update: free empty path string (memleak)Max Kellermann2009-08-142-1/+3
| | | | | | | | | | When you pass an empty string to directory_update_init(), it was not freed by update_task().