Better error handling.
Move the checking of the shell command to start so a proper error can be thrown if it's None.
Use write instead of write_all for finer grained error handling and the ability to attempt a restart on write errors.
Use try_wait to skip flushing and killing the process if it's already dead.
Stop the player on shutdown to *mostly* prevent write errors from spamming the logs during shutdown. Previously Ctrl+c always resulted in a write error.
This sets the name displayed by PulseAudio to Librespot - Instance Name if a name is given otherwise Librespot (the default name).
This also sets the correct "role" as per the docs:
https://www.freedesktop.org/wiki/Software/PulseAudio/Documentation/Developer/Clients/ApplicationProperties/
PA_PROP_MEDIA_ROLE
"This is a property of the actual streamed data, not so much the application"
Roles are used for policies, things like automatically muting a music player when a call comes in and whatnot.
For bonus points this also sets PULSE_PROP_application.icon_name to audio-x-generic so that we get a nice icon in the PulseAudio settings by our name instead of a missing icon placeholder.
Dynamically set the alsa buffer and period based on the device's reported min/max buffer and period sizes. In the event of failure use the device's defaults.
This should have no effect on devices that allow for reasonable buffer and period sizes but would allow us to be more forgiving with less reasonable devices or configurations.
Closes: https://github.com/librespot-org/librespot/issues/895
This makes `--device ?` only show compatible devices (ones that support 2 ch 44.1 Interleaved) and it shows what `librespot` format(s) they support.
This should be more useful to users as the info maps directly to `librespot`'s `--device` and `--format` options.
* Don't panic when parsing options. Instead list valid values and exit.
* Get rid of needless .expect in playback/src/audio_backend/mod.rs.
* Enforce reasonable ranges for option values (breaking).
* Don't evaluate options that would otherwise have no effect.
* Add pub const MIXERS to mixer/mod.rs very similar to the audio_backend's implementation. (non-breaking though)
* Use different option descriptions and error messages based on what backends are enabled at build time.
* Add a -q, --quiet option that changed the logging level to warn.
* Add a short name for every flag and option.
* Note removed options.
* Other misc cleanups.
* Make error messages more consistent and concise.
* `impl From<AlsaError> for io::Error` so `AlsaErrors` can be thrown to player as `io::Errors`. This little bit of boilerplate goes a long way to simplifying things further down in the code. And will make any needed future changes easier.
* Bonus: handle ALSA backend buffer sizing a little better.
* Improve error handling
* Harmonize `Seek`: Make the decoders and player use the same math for converting between samples and milliseconds
* Reduce duplicate calls: Make decoder seek in PCM, not ms
* Simplify decoder errors with `thiserror`
* Reuse the buffer for the life of the Alsa sink
* Don't depend on capacity being exact when sizing the buffer
* Always give the PCM a period's worth of audio even when draining the buffer
* Refactoring and code cleanup
* More meaningful error messages
* Use F32 if a user requests F64 (F64 is not supported by PulseAudio)
* Move all code that can fail to `start` where errors can be returned to prevent panics
* Use drain in `stop`
Better error handling in Alsa backend
* More consistent error messages
* Bail on fatal errors in player
* Capture and log the original error as a warning when trying to write to PCM before trying to recover
* Go back to 4 periods at 125ms.
* Deal strictly in period time and periods to set ALSA buffer.
* Rename `buffer` to `period_buffer`.
* Add comments and change some other var names to add clarity.
* Let ALSA calculate the size of `period_buffer`.
* Remove deprecated use of std::u16::MAX
* Use `FromStr` for fallible `&str` conversions
* DRY up strings into constants
* Change `as_ref().map()` into `as_deref()`
* Use `Duration` for time constants and functions
* Optimize `Vec` with response times
* Move comments for `rustdoc` to parse
Dithering lowers digital-to-analog conversion ("requantization") error, linearizing output, lowering distortion and replacing it with a constant, fixed noise level, which is more pleasant to the ear than the distortion.
Guidance:
- On S24, S24_3 and S24, the default is to use triangular dithering. Depending on personal preference you may use Gaussian dithering instead; it's not as good objectively, but it may be preferred subjectively if you are looking for a more "analog" sound akin to tape hiss.
- Advanced users who know that they have a DAC without noise shaping have a third option: high-passed dithering, which is like triangular dithering except that it moves dithering noise up in frequency where it is less audible. Note: 99% of DACs are of delta-sigma design with noise shaping, so unless you have a multibit / R2R DAC, or otherwise know what you are doing, this is not for you.
- Don't dither or shape noise on S32 or F32. On F32 it's not supported anyway (there are no integer conversions and so no rounding errors) and on S32 the noise level is so far down that it is simply inaudible even after volume normalisation and control.
New command line option:
--dither DITHER Specify the dither algorithm to use - [none, gpdf,
tpdf, tpdf_hp]. Defaults to 'tpdf' for formats S16
S24, S24_3 and 'none' for other formats.
Notes:
This PR also features some opportunistic improvements. Worthy of mention are:
- matching reference Vorbis sample conversion techniques for lower noise
- a cleanup of the convert API
fe37186 added the restriction that `Sink`s must be `Send`. It turned
out later that this restrictions was unnecessary, and since some
`Sink`s aren't `Send` yet, this restriction is lifted again.
librespot-org/librespot#601 refactored the `RodioSink` in order to make
it `Send`. These changes are partly reverted in favour of the initial
simpler design.
Furthermore, there were some compile errors in the gstreamer backend
which are hereby fixed.
- DRY-ups
- Remove incorrect optimization attempt in the libvorbis decoder,
that skewed 0.0 samples non-linear
- PortAudio and SDL backends do not support S24 output. The PortAudio
bindings could, but not through this API.
