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.
No CamelCase in the file name. The output_buffer struct is going to
be renamed to music_pipe. There are so many buffer levels in MPD, and
calling this one "output buffer" is wrong, because it's not the last
buffer before the music reaches the output devices.
Commit 1a4a3e1f moved decoders into a static array, but failed to
enable those plugins who did not have an init() method at all.
This patch corrects the "enabled" check.
Make map_directory_child_fs() refuse the names "." and "..". This is
currently the interface where an attacker may inject a manipulated
path (through the "update" command).
