This change collects all those audio fetch parameters that were defined as
static constants into a dedicated struct, AudioFetchParams.
This struct can be read and set statically, allowing a user of the library to
modify those parameters without the need to recompile.
- Use variables directly in format strings.
As reported by clippy, variables can be used directly in the
`format!` string.
- Use rewind() instead of seeking to 0.
- Remove superfluous & and ref.
Signed-off-by: Petr Tesarik <petr@tesarici.cz>
- Improve responsiveness by downloading the smallest possible chunk
size when seeking or first loading.
- Improve download time and decrease CPU usage by downloading the
largest possible chunk size as throughput allows, still allowing
for reasonable seek responsiveness (~1 second).
- As a result, take refactoring opportunities: simplify prefetching
logic, download threading, command sending, and some ergonomics.
- Fix disappearing controls in the Spotify mobile UI while loading.
- Fix handling of seek, pause, and play commands while loading.
- Fix download rate calculation (don't use the Mercury rate).
- Fix ping time calculation under lock contention.
- Ensure there is enough disk space for the write file
- Switch streaming mode only if necessary
- Return `Err` on seeking errors, instead of exiting
- Use the actual position after seeking
- Configure the decoder according to Spotify's metadata, don't probe
- Return from `AudioFile::open` as soon as possible, with the smallest
possible block size suitable for opening the decoder, so the UI
transitions from loading to playing/paused state. From there,
the regular prefetching will take over.
- Switch from `lewton` to `Symphonia`. This is a pure Rust demuxer
and decoder in active development that supports a wide range of
formats, including Ogg Vorbis, MP3, AAC and FLAC for future HiFi
support. At the moment only Ogg Vorbis and MP3 are enabled; all
AAC files are DRM-protected.
- Bump MSRV to 1.51, required for `Symphonia`.
- Filter out all files whose format is not specified.
- Not all episodes seem to be encrypted. If we can't get an audio
key, try and see if we can play the file without decryption.
- After seeking, report the actual position instead of the target.
- Remove the 0xa7 bytes offset from `Subfile`, `Symphonia` does
not balk at Spotify's custom Ogg packet before it. This also
simplifies handling of formats other than Ogg Vorbis.
- When there is no next track to load, signal the UI that the
player has stopped. Before, the player would get stuck in an
infinite reloading loop when there was only one track in the
queue and that track could not be loaded.
This should fix remaining cases of a client connecting, and failing
to start playback from *beyond* the beginning when `librespot` is
still loading that track.
This undoes the `suppress_loading_status` workaround from #430,
under the assumption that the race condition reported there has
since been fixed on Spotify's end.
Further changes:
- Improve some debug and trace messages
- Default to streaming download strategy
- Synchronize mixer volume on loading play
- Use default normalisation values when the file position isn't
exactly what we need it to be
- Update track position only when the decoder reports a
successful seek
* 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
Some of the feature flags librespot uses are not really additive but
rather mutual exclusive. A previous attempt to improve the situation
had other drawbacks, so it's better to postpone a decision and restore
the old behaviour.
