In many tests we assert that encoding and decoding gives the same
object. Let's call that a "roundrip", and also provide a proc macro to
generate such tests.
Remove `fn deserialize` and instead provide a trait `fn decode_exact`
for testing.
Refactor errors, so that we can now cleanly distinguish between
`Error::UnexpectedEnd` which means that decoding might work given
more data, and `Error::Invalid`, which means that no amount of
additional data will decode.
While at it, cleanup the variants of `Error::Invalid`, and provide `ControlType`.
What made the issue fixed in the parent commit harder to spot
for me was the multitude of concepts and terms involved in
encoding: What's *serializing* vs. *encoding*? Why is there
this odd `Frame::to_bytes`?
So I decided to clean up a bit. I removed `fn serialize` so that the
logic stays close to the data, and instead of it provide
`fn encode_vec` in `trait Encode` so that we don't add up with
one-offs like `Frame::to_bytes` again.
Also we make use of `encode_vec` in the tests.
In 3c5668edd2 I mistakenly moved the check
the length of an encoded `wire::Message` from its `impl wire::Encode` to
the `fn serialize`, so that it would not apply only to messages (as
originally intended), but to *any* argument of `serialize`.
This is not a problem for gossip messages, but catastrophic for Git
streams, which tend to send a lot of data.
The fix is simple: Move the check back into `impl Encode for Message`.
Mostly changes of `use`s and moving components to `radicle-protocol`,
with the goal of just getting `radicle-node` to work on top of the new
`radicle-protocol` crate.
Lorenz Leutgeb