GitLab

The API is a bit yucky, since it has to be compatible with tokio and
async_std.

(Also, this patch fixes some async_std clippy warnings)

Public function so an application can use to build a circuit using a
custom path.
Signed-off-by: David Goulet <dgoulet@torproject.org>

Since we moved to a non-async mutex(*), we no longer need to worry that
this function might need to suspend.

(*) This is _not_ safe in general, but it's okay in this case, since
we never suspend while holding that mutex: see shared_ref.rs.

Doing this will make it easier to use it from other parts of the
crate, and will make it more obvious that it's safe to use a regular
(not async) rwlock here.

This will let us use these types both as client and server-side
implementations.

Making this change required me to change the download code to take
requests by reference.  (Sorry, David)

Also, start on a larger refactoring where we ask for documents by
their ID.

These were stored in the database before, but they were hardwired in
the API.

Previously we were "stringly typing" this field.  Now it's time to stop.

Previously we used NetDir::extend here, which used try_unwrap.  But
it's much simpler to use the intended object for this purpose.

Signed-off-by: David Goulet <dgoulet@torproject.org>

Rationale: the name of the request should match the name of the type
that you parse from it.

Having "no descriptors" mean "all of them" is kind of an accident
waiting to happen, and had wrong behavior for partial_docs_ok() and
max_response_len().

Signed-off-by: David Goulet <dgoulet@torproject.org>

Signed-off-by: David Goulet <dgoulet@torproject.org>

Signed-off-by: David Goulet <dgoulet@torproject.org>

This function sits behind an `experimental-api` feature so that we
don't need to worry about exposing the entire surface of DirMgr to
our API consumers.

This and the other recent patches are based on work from dgoulet.