Voice fixes, better API adherence, bitrate control, documentation

* Fixing silence frame.

* Messed that one up while fighting with the borrow checker. Sorry!

* Initial machinery for playback position tracking

* Mix multiple input AudioType::Opus streams

* Encode for stereo, use nicer "soft clip"

* First stab at docs for Audio.

* Better-er docs for voice::Audio etc.

* Bitrate control.

* Fix #270, Better handling of the voice api

We were mostly doing the voice API wrong, I've changed OpCode names to
be correct. We now listenfor both Ready and Hello at connection init,
and do soft checks for Heartbeat ACKs.

* Adding missing voice opcodes, related structs

* Also removes events for messages which cannot be received.

* @Zeyla's recommended changes.

* New docstrings now have correct style for referring to functions.
* Docstrings also have room to breathe (!)
* Rand dependency now properly moved behind the `voice` feature.
* Slightly cleaner checks at voice connection initialisation.
* Various idiomatic changes throughout.

* Prevent compilation of Audio docs example.

Likely too much machinery in the background to get a working Handler, I
think.

* Re-fixing the docstrings.

* Fixing travis for Audio docs.

1 files changed, 2 insertions, 0 deletions

diff --git a/src/lib.rs b/src/lib.rs
index 196a3e0..adc8283 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -131,6 +131,8 @@ extern crate multipart;
 extern crate native_tls;
 #[cfg(feature = "opus")]
 extern crate opus;
+#[cfg(feature = "rand")]
+extern crate rand;
 #[cfg(feature = "sodiumoxide")]
 extern crate sodiumoxide;
 #[cfg(feature = "threadpool")]