aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* pcm_convert: return PCM buffer from pcm_convert()Max Kellermann2009-01-177-86/+43
| | | | | | Removed yet another superfluous buffer layer: return the PCM buffer from pcm_convert() instead of copying PCM data into the caller-supplied buffer.
* music_pipe: removed music_pipe_append()Max Kellermann2009-01-172-39/+0
| | | | | The function is unused, since decoder_api() uses music_pipe_write() / music_pipe_expand() now.
* decoder_api: use music_pipe_write() instead of music_pipe_append()Max Kellermann2009-01-171-14/+33
| | | | | Copy PCM data to the music_pipe_write() buffer, and apply replay gain / normalization to it, instead of manipulating the source buffer.
* music_pipe: added functions music_pipe_write() and music_pipe_expand()Max Kellermann2009-01-172-32/+64
| | | | | | | This new API gives the caller a writable buffer to the music pipe chunk. This may allow the caller to eliminate several buffer copies, because it may manipulate the returned buffer, until it calls music_pipe_expand().
* oggvorbis: disable seeking on remote songsMax Kellermann2009-01-172-5/+28
| | | | | | | | | | | | When libvorbis knows that a song is seekable, it seeks around like crazy in the file before starting to decode it. This is very expensive on remote HTTP resources, and delays MPD for 10 or 20 seconds. This patch disables seeking on remote songs, because the advantages of quickly playing a song seem to weigh more than the theoretical ability of seeking for most MPD users. If users feel this feature is needed, we will make a configuration option for that.
* decoder_api: added assertion on partial framesMax Kellermann2009-01-161-0/+1
| | | | Decoder plugins must not send partial frames.
* pcm_resample_fallback: corrected the sample calculationMax Kellermann2009-01-162-6/+9
| | | | | Due to rounding errors, it was possible that the fallback resampler returned partial frames.
* playlist: don't store getBoolConfigParam() in a boolMax Kellermann2009-01-161-3/+5
| | | | | | getBoolConfigParam() returns an int. It is not possible to check for CONF_BOOL_UNSET after it has been assigned to a bool; use a temporary int value for that.
* listen: explicitly include sys/socket.hMax Kellermann2009-01-161-0/+1
|
* player_control: fix shadow warning with older gccMax Kellermann2009-01-161-1/+1
| | | | Older gcc versions complained about shadowed parameters in prototypes.
* oss: evaluate the oss_open() return value properlyMax Kellermann2009-01-161-1/+1
| | | | It returns bool, not int.
* mixer: check param==NULL in the alsa and oss mixer codeMax Kellermann2009-01-162-0/+7
| | | | | | When MPD starts without audio output configuration, the "param" variable is NULL. This triggers a segmentation fault in both mixer plugins.
* Enable wav file streaming for ffmpeg input pluginQball Cow2009-01-161-0/+1
|
* oggvorbis: define HAVE_OGGVORBIS when tremor is enabledMax Kellermann2009-01-152-1/+4
| | | | When tremor (libvorbisidec) is used, HAVE_OGGVORBIS was not defined.
* flac: map "Album Artist" to "AlbumArtist"Rasmus Steinke2009-01-151-1/+3
| | | | | This patch allows mpd to recognise the albumartist tag in the way foobar2000 and others write it to files.
* flac: splitted flac_copy_vorbis_comment()Max Kellermann2009-01-151-37/+38
| | | | | Splitted flac_copy_vorbis_comment() into flac_copy_comment() and flac_copy_comment().
* flac: moved code to flac_comment_value()Max Kellermann2009-01-151-5/+26
| | | | | Simplify flac_copy_vorbis_comment() by moving the comment identification code out.
* flac: always allocate tag objectMax Kellermann2009-01-154-31/+43
| | | | | Free the tag object when it turns out to be empty. This simplifies several functions and APIs.
* tag: added tag_is_defined()Max Kellermann2009-01-151-0/+9
| | | | | tag_is_defined() checks whether there is any information in the tag object.
* flac: use bool instead of intMax Kellermann2009-01-152-42/+37
|
* flac: removed "vorbis_comment_found" flagMax Kellermann2009-01-151-10/+4
| | | | Use tag_is_empty() instead.
* flac: no CamelCaseMax Kellermann2009-01-155-190/+208
| | | | Renamed types, functions, variables.
* input_curl: don't call input_curl_select() when already at EOFMax Kellermann2009-01-151-1/+3
| | | | | | Calling input_curl_select() after EOF has been reached causes an assertion failure. This can happen if the HTTP response is empty. Check c->eof before calling input_curl_select().
* input_curl: set "ready" flag on EOFMax Kellermann2009-01-152-0/+6
| | | | Set the "ready" flag for empty resources.
* input_curl: call curl_multi_info_read() in constructorMax Kellermann2009-01-152-0/+7
| | | | | To check for early connect failures, call curl_multi_info_read() in the constructor input_curl_open(). This fixes an assertion failure.
* NEWS: merge from v0.14.xMax Kellermann2009-01-151-3/+19
| | | | Copy changelog items from the upcoming 0.14.1 release.
* input_curl: added fallback for g_queue_clear()Max Kellermann2009-01-151-0/+5
| | | | | g_queue_clear() was introduced in GLib 2.14. Add a macro hack for older GLib versions to emulate it.
* log, pcm_convert: added return statements after g_error()Max Kellermann2009-01-152-1/+4
| | | | | gcc doesn't know that g_error() never returns. Work around the gcc warning.
* mapper: check if g_get_user_special_dir() is supported by GLibMax Kellermann2009-01-151-0/+2
| | | | | g_get_user_special_dir() was introduced with GLib 2.14. Don't use it in older versions.
* pcm_convert: removed unused variable "dest_size"Max Kellermann2009-01-151-4/+2
| | | | | dest_size is only used in an assertion. Remove its declaration and move the formula into the assertion.
* decoder_api: added G_GNUC_UNUSED attribute to decoder_initialized()Max Kellermann2009-01-151-1/+1
| | | | In NDEBUG, the parameter "decoder" is not used.
* listen: don't compile ipv6Supported() if !HAVE_IPV6Max Kellermann2009-01-151-4/+2
| | | | | The function ipv6Supported() is not used at all when IPv6 support was disabled at compile time.
* oggvorbis: map "Album Artist" to "AlbumArtist"Rasmus Steinke2009-01-151-1/+3
| | | | | This patch allows mpd to recognise the albumartist tag in the way foobar2000 and others write it to files.
* songvec: sort songs by disc and track numberMax Kellermann2009-01-151-0/+46
| | | | | Sorting songs by file name does not make much sense. Most of the time, users want to add songs in track order to the playlist.
* tag: added tag_get_value()Max Kellermann2009-01-152-3/+16
|
* dirvec, songvec: sort using g_utf8_collate()Max Kellermann2009-01-152-2/+2
| | | | | | Path names in the directory and song structs are always encoded in UTF-8. Don't use strcmp(), it cannot handle UTF-8 characters properly. Use GLib's UTF-8 aware g_utf8_collate() function for that.
* shout: enlarge buffer size to 32 kBAntoine Beaupré2009-01-141-1/+1
| | | | | | | I was having problems with shoutcast stream outputs before applying the attached patch, which enlarges the shoutcast output buffer. Ideally, this should be configurable, but this resolves the issue for my needs.
* oggvorbis: use g_ascii_strncasecmp() instead of strncasecmp()Max Kellermann2009-01-141-3/+3
| | | | | Don't depend on the daemon's locale settings. Comment names are ASCII.
* oggvorbis: moved tag look into vorbis_parse_comment()Max Kellermann2009-01-141-23/+13
| | | | | | vorbis_parse_comment() should be a function which converts one comment to a tag item. It should do everything required to do the conversion, including looping over all possible tag types.
* oggvorbis: moved code to vorbis_copy_comment()Max Kellermann2009-01-141-9/+21
|
* oggvorbis: use vorbis_comment_value() in vorbis_parse_comment()Max Kellermann2009-01-141-5/+4
| | | | Eliminate some duplicate code.
* oggvorbis: always allocate a tag objectMax Kellermann2009-01-141-8/+14
| | | | | | Always allocate a new tag object before parsing the vorbis comments; free it when it turns out to be empty. This simplifies the code a bit.
* oggvorbis: no CamelCaseMax Kellermann2009-01-142-61/+65
| | | | Renamed functions and variables.
* oggvorbis: use boolMax Kellermann2009-01-141-5/+5
| | | | Make ogg_parseCommentAddToTag() return bool instead of unsigned int.
* tag_id3: added support for the "album artist" tagDaniele Sluijters2009-01-141-0/+12
| | | | Read the id3 tags "TPE2" and "TSO2" into the "album artist" tag.
* allow ~/.mpd/mpd.conf as alternate config fileJérôme Quelin2009-01-143-12/+20
| | | | | | | | | | | | mpd uses some additional files to work, such as pid_file, state_file, db_file, etc. when running mpd as non-root user, it is often that those files end in ~/.mpd in that case, we end up with 2 entries in a user's home, .mpdconf and .mpd - which clutters homedirs. this patch allows ~/.mpd/mpd.conf as an alternative to ~/.mpdconf, allowing for a cleaner homedir
* song_save: don't fail on empty tag valuesMax Kellermann2009-01-141-2/+2
| | | | | | If a tag value is an empty string, the space after the colon was removed by g_strchomp(). Fix this by removing the space check and using g_strchug() on the return value.
* song_save: return value pointer from matchesAnMpdTagItemKey()Max Kellermann2009-01-141-8/+8
| | | | | | | The matchesAnMpdTagItemKey() API becomes more powerful and flexible if the return value is the value pointer instead of a boolean. It also removes (invalid and dangerous) assumptions about the string from its caller.
* song_save: itemType is "enum tag_type", not "int"Max Kellermann2009-01-141-2/+2
|
* playlist: safely search the playlist for deleted songMax Kellermann2009-01-142-1/+8
| | | | | | | | When a song file is deleted during database update, all pointers to it must be removed from the playlist. The "for" loop in deleteASongFromPlaylist() did not deal with multiple copies of the deleted song properly, and left instances of the (to-be-invalidated) pointer in. Fix this by reversing the loop.