* Receive: Config of decode sample rate/channels
This PR allows for dynamic configuration of the output sample rate and
channel count of received Opus audio. Users who rely on supported
formats should no longer need to manually resample & downmix audio
decoded from SSRCs in a call.
Opus exposes tuples of (Mono, Stereo) x (8, 12, 16, 24, 48)kHz.
Changing this at runtime (mid-call) may cause some audio glitches, as
decoder state must be reconstructed from scratch for all affected SSRCs.
* Fix doc typo, consistent naming with MixMode.
This changes HlsRequest to return true in
Compose::should_create_async.
This would otherwise cause it to be spawned on a thread without a
executor causing it to panic.
This would only cause issues if used directly since we otherwise use
Compose::create in a context where we have access to the executor.
This PR adds support for the new AEAD cryptosystems advertised by Discord, AES256-GCM and XChaCha20Poly1305. These schemes will shortly become mandatory, and provider stronger integrity/authentication guarantees over the cleartext portions of any voice packet by correctly specifying additional authenticated data.
To provide smooth switchover, we've added basic negotiation over the `CryptoMode`. This ensures that any clients who are manually specifying one of the legacy modes will automatically migrate to `Aes256Gcm` when Discord cease to advertise their original preference.
Closes#246.
---------
Co-authored-by: Kyle Simpson <kyleandrew.simpson@gmail.com>
A few tests are routinely failing on CI because they explicitly
hit YouTube URLs, even if the content in question is CC-licensed.
This PR simply adds the `#[ignore]` attribute, and adds the
`--include-ignored` flag to `cargo make test` so that correct WebM
parsing behaviour can be locally verified.
We could go a bit further and conditionally ignore the tests based
on CI-specific ENV vars, but that's probably overkill given the
scope of the issue.
Bots joining calls with users seem to provoke large runs of packets
with identical timestamps -- the existing logic was intended to handle this catchup case in addition to the normal (+=960) at all
times.
However, we were checking that a packet was modulo greater-than
the next ts, rather than modulo less-than. Simple enough to fix.
Ringbuf 0.4.2 onwards adds in a new storage type for `SharedRb`.
Unfortunately, this is handled in such a way that we fail to
compile because an explicit type parameter is required to
disambiguate the method in question.
Tokio-tungstenite have made some runtime-breaking changes to how
crypto setup is init'd -- i.e., users must do it themselves.
Reverting due to the huge (breaking) inconvenience this poses.
This will cause FreeBSD to fail setting up the socket. It may also be true of some other operating systems, but these are the ones I have been able to test.
It works on Windows and Linux.
The `Config` object provided to `Call`s and `Driver`s allows setting a `DisposalThread`, but since it is unreachable from outside the crate, the only way to properly set it is using a `Songbird` instance, which may not be adequate for all use cases. This PR just makes `DisposalThread` reachable from the outside.
Fixes behaviour where a Driver which was asked to leave an active call would receive the disconnect event several times: once when we started the disconnect, and once again when Discord killed the WS client.
A little hacked together, but it is close enough to our primary format.
Will still require manual sorting of commits by function since we're not
doing conventional commits, but its lower friction for filling out
github usernames etc.
Previously, we were only skipping zero-packet frames when we needed to
resample because the source sampling rate was not set to 48kHz. This
check should have also been applied in the case that a packet did not
need a resampler to be built.
Fixes#224.
This PR fixes a case where a call which changes channel or gracefully
reconnects would have been stuck in the Idle state. SetConn events will
now allow a transition straight back to Live if any tracks are found
attached to a mixer.
* Add ytdl search
* fix fmt
* Remove compose, add tests, return AuxMetadata
* fix parsing of AuxMetadata and better test
* Fix playability of `YoutubeDl::new_search`
Refactors such that parsing of (ND)JSON is handled in only one location
now, which allows us to greatly simplify the actual `search` method. The
main change is that any `new_search` is now instantly playable.
---------
Co-authored-by: Kyle Simpson <kyleandrew.simpson@gmail.com>
As of v0.9.1, `xsalsa20poly1305` has been deprecated. This is a mostly seamless replacement, as it appears to be the same crate authors / code / etc.
Co-authored-by: Kyle Simpson <kyleandrew.simpson@gmail.com>
* chore(ci): Update rust setup and cache action
* chore(ci): Remove legacy-tokio from test matrix
This was removed as it doesn't seem to be used any more.
This PR changes the following things in the docs workflow:
- Move from the deprecated actions-rs action to the maintained alternative of dtolnay https://github.com/dtolnay/rust-toolchain
- Instead of manually defining the cache paths and the cache keys, I moved to an action that handles this for us: https://github.com/Swatinem/rust-cache
- As github decided at some point that the generated `GITHUB_TOKEN` secret will only have read-only permissions (for new repos), unless you tell them to be different, I added the `permissions` key to the job. This will not affect this repository, but will affect e.g. new forks (the workflow will not throw an permissions error any more - Read more here: https://github.blog/changelog/2023-02-02-github-actions-updating-the-default-github_token-permissions-to-read-only/ )
