Don't split the buffer conversion loop. When libmpcdec returns a
chunk, convert and send the whole chunk at a time. This moves several
checks out of the loop, and greatly improves performance.
Parse ID3 tags, even when they are in the middle of the stream. Very
few streams provide embedded ID3 tags. Most of them send only
Shoutcast "icy" tags, which limits the practical usefulness of this
patch.
When a command is received, decode_next_frame_header() and
decodeNextFrame() return DECODE_BREAK. This is already checked by
both callers, which means that we can eliminate lots of
decoder_get_command() checks.
When a tag is updated, the old tag was freed before the new one was
created. Reverse the order to be sure that other threads always see a
valid pointer.
This still leaves a possible race condition, but it will be addressed
later.
The stream_decode() and file_decode() methods returned a boolean,
indicating whether they were able to decode the song. This is
redundant, since we already know that: if decoder_initialized() has
been called (and dc.state==DECODE), the plugin succeeded. Change both
methods to return void.
The currently replay_gain_apply() implementation duplicates code from
pcm_volume(), except that it uses a floating point scale. Eliminate
all duplicated code from and make it utilize the pcm_volume() library
function. This introduces replay gain support for 24 bit audio.
It may be desirable to change the range of integer volume levels
(e.g. to 1024, which may utilize shifts instead of expensive integer
divisions). Introduce the constant PCM_VOLUME_1 which describes the
integer value for "100% volume". This is currently 1000.
The function simplifies wavpack_replaygain(), because it already
contains the float parser, and it works with a fixed buffer instead of
doing expensive heap allocations.
The assertion on dc.state in decoder_read() was too strict: when a
decoder tried to call decoder_read() from tag_dup(), the decoder state
was NONE. Allow this special case.
The flac plugin wasn't initialized properly when an OGG file was being
decoded. For some reason, flac_process_metadata() was explicitly not
called for OGG files. Since that seems to fix the issue, make it
always call flac_process_metadata().
Since decoder_list.c does not include the libflac headers, it cannot
know whether to add the oggflac plugin to the decoder list. Solve
this by always enabling the oggflac sub-plugin, even with older
libflac versions. When the libflac API cannot support oggflac,
disable the plugin at runtime by returning "false" from its init()
method.
The "oggflac" plugin was enabled only if HAVE_FLAC_COMMON was
defined. HAVE_FLAC_COMMON however is only an automake variable, and
is never available in decoder_list.c. Make decoder_list.c depend on
HAVE_FLAC||HAVE_OGGFLAC instead.
The player did not care about the exact error value, it only checked
whether an error has occured. This could fit well into
decoder_control.state - introduce a new state "DECODE_STATE_ERROR".
At this moment the wavpack lib doesn't use the return value of the
push_back function, which has an equivalent meaning of the return
value of ungetc(). This is a lucky situation, because so far it
simply returned with 1 as a hard coded value. From now on the
function will return EOF on error. (This function makes exactly one
byte pushable back.)
There are some functions in the wavpack-mpd input streams wrapper
which had too commonly used names (especially can_seek). I prefixed
these with "wavpack_input_".
The listen.c module breaks the build because the variable name used
("sun") for the Unix domain socket part collides with something else
on an OpenSolaris system, likely Sun specific. Renaming it to _sun
(or something else of choice) fixes the build.
[mk: renamed to "s_un"]
I had this option enabled during development, but at some point, it
must have gotten lost. FAILONERROR makes the curl stream fail when
the server returns a status code 400 or higher. We are not interested
in the server's error document.
Initialize libc's locale functions. Currently, we are only interested
in LC_CTYPE (character classification), because this is what is used
by GLib's g_get_charset().
GLib provides the function g_get_filename_charsets() which determines
the file system character set. This changes MPD's fallback: GLib
prefers UTF-8 as a fallback. MPD used to fall back to ISO Latin 1.
libwavpack expects the read_bytes() stream method to fill the whole
buffer, and fails badly when we return a partial read (i.e. not enough
data available yet). This caused wavpack streams to break.
Re-implement the buffer filling loop.
Instead of manually waiting for the input stream to become ready (to
catch server errors), just read the first byte. Since the
wavpack_input has the capability to push back one byte, we can simply
re-feed it. Advantage is: decoder_read() handles everything for us,
i.e. waiting for the stream, polling for decoder commands and error
handling.
The API of mp4_load_tag() was strange: it always returned a tag
object, no matter if a tag was found in the file; the existence of a
tag was indicated with the tag_found integer reference. This flag is
superfluous, since we can simply check whether the tag is empty or
not.
Allocate the mp4ff_callback_t object on the stack. This is easier to
handle, since we don't have to free it. Incidentally, this fixes a
memory leak in mp4_load_tag().
The function decoder_read() already cares about the decoder command,
and loops until data is available. Reduced mpd_ffmpeg_read() to no
more than the decoder_read() call.
The variable "next_song" is already protected by a memory barrier.
"total_time" is not important for synchronization, and we don't need
"volatile" here.
If an input stream provides tags (e.g. from an icecast server), send
them in the decoder_data() and decoder_tag() methods. Removed the
according code from the mp3 and oggvorbis plugins - decoders shouldn't
have to care about stream tags.
This patch also adds the missing decoder_tag() invocation to the mp3
plugin.
MPD used to have a copy of the mp4ff library. Since that has been
removed, AAC suport was disabled when there was no libmp4ff. Separate
the libmp4ff test, and enable AAC support no matter if libmp4ff is
available.
The "mod" decoder plugin was being initialized lazily, but was
deinitialized unconditionally. That led to segmentation faults.
Convert mod_initMikMod() to be the global module initialization
method. The MPD core should care about lazy initialization.
Non-local songs used to have no tags. If the decoder sends us a tag,
we should incorporate it into the song struct. This way, clients can
always show the correct song name (if provided by the server).
The try_decode() method may have read some data from the stream, which
is now lost. To make this data available to other methods, get it
back by rewinding the input stream after each try_decode() invocation.
The ogg and wavpack plugins did this manually and inconsistently; this
code can now be removed.
If the source chunk has a tag, merge it into the destination chunk.
The source chunk gets deleted after that, and this is our last chance
to grab the tag.
Ogg and ffmpeg detection was disabled when the stream was not
seekable, because the detection was too expensive. Since the curl
input stream can now rewind the stream cheaply, we can re-enable
detection on streams.
During codec detection, the beginning of the stream is consumed. This
is a common operation, which takes a lot of time when handling remote
resources. To optimize this, remember the first 64 kB of a stream.
This way, we can rewind the stream without actually fetching the start
of the stream again.
Since the aac and mod plugins have told MPD that they cannot seek, MPD
will never send a SEEK command to them. Removed the SEEK comand
checks from both plugins.
Don't pass the "seekable" flag with every decoder_data() invocation.
Since that flag won't change within the file, it is enough to pass it
to decoder_initialized() once per file.
Replace all direct music_pipe struct accesses with wrapper functions.
The compiled machine code is the same, but this way, we can change
struct internals more easily.
.. and rename dc.audioFormat to dc.in_audio_format. The music pipe
does not need to know the audio format, and its former "audioFormat"
property indicated the format of the most recently added chunk, which
might be confusing when you are reading the oldest chunks.
