transcode_needed() was getting called needlessly in http_daapd.c,
because 1) once it is determined that a given codec needs transcoding
for a given client there is no reason to call and check again, 2)
transcoding is irrelevant for remotes. Also some cleaning up of
user_agent_filter().
- Try to not return items which a client can't play
- Remove inotify subscription to IN_MODIFY and IN_CREATE
- Fix crash on unknown codec type in transcode.c
- Probably added some new bugs...
Clang produced interesting results without this (or casting ms to int64_t),
as the seek target got mis-computed and fell short of the requested seek
target in ms (ex. wanted 18569 ms -> got 555 ms).
Assign start_time right at the start, making the target_pts computation more
obvious wrt start_time and showing the symmetry of the target_pts and got_pts
computations.
Older versions of ffmpeg did not support raw FLAC streams properly and needed
to be fed the raw stream manually; looks like it's been fixed in ffmpeg 0.5.
It is now clear that multi-library support will not happen, so remove whatever
provisions were in the code for that.
It comes with a small change to the configuration file, too.
With this, DB schema version went to 9.
Transcoded (decoded) files will now always come out in signed, little endian,
16bit, 44100 Hz, stereo format regardless of the format of the input file.
This in effect fixes transcoding (and playback on some devices) for files that
do not match this format.
There's probably a discussion to be had regarding handling of 48 kHz and 96 kHz
content, though, as downsampling to 44.1 kHz to have the client or final output
device upsample again is clearly not an optimal solution.
Add the setup_fail_codec label and jump to it if an error occurs once the
codec has been opened. In the raw input codepath, don't use this label until
the file is properly opened, as it also closes the fd and frees the raw
buffer.
This also fixes a file descriptor leak in the case where an error happened
after the file was opened in the raw input codepath.
lseek() returns an off_t and not an int, using an int to store and
test the return value means we'll error out when the position in the file
gets past INT_MAX.
Try to be a bit more strict about integer types, use off_t or int64_t for
file size and file offsets.
Replace safe_ato*() by safe_atoi32() and safe_atoi64(), fix integer types
at call sites to match.
ejurgensen
chme
Justin Maggard
Julien BLACHE