This makes using the `PreferredRuntime` the first-class option inside
`arti-client`, freeing users who don't want to think about runtimes from
having to do so.
`TorClient::create_unbootstrapped` and `builder` now automatically
use this runtime, leaving only `builder_custom` for users who wish to
manually specify a runtime.
This lets us clean up the docs a lot: mentions of using custom runtimes
are now relegated to nearer the end of the crate-level documentation,
and we mostly just link to `tor_rtcompat`'s docs to explain more there.
Instead, we take some more time to explain how you use the builder API
to create clients synchronously.
Other doc cleanups included getting rid of the explanation of `TorAddr`
in the main crate-level doc; this is already well-documented elsewhere,
and is something users should discover organically later.
fixes arti#326
Add tls_conn field to ArtiHttpConnector (and argument to constructor).
Introduce MaybeHttpsStream and use it in ArtiHttpConnection.
Have the example program pass the native TLS connector.
Currently the TLS connector and the HTTPS variant are not used, but
this commit is very noisy and fomrulaic, so I have split out the code
to use them into a separate commit for easier preparation and review.
arti-hyper wants to be able to have a kind for TLS failure.
Given that arti-hyper is above arti-client, this shows that callers
above arti-client might need to invent kinds for their own errors.
Possibly this means we need other Other errors for other locations.
If we have pluggable components we might even want OtherTorError.
Fixes#365
Inspection of the code and logs shows that:
* One of the plan futures' oneshots must be returning Cancelled
* This means that the corresponding sender must have been dropped
* The sender is owned by the task spawned by spawn_launch
Presumably that entire task gets dropped as part of executor shutdown,
or something.
The correct response in this situation is to declare that we are
shutting down, and stop trying to do stuff.
Unfortunately, despite trying quite hard by putting sleeps in various
strategic places, I have not been able to reproduce the problem. So I
can't be 100% sure that the new behaviour is correct.
But I am reasonably confident that this ought not to be able to occur
unless either 1. the task from spawn_launch is dropped, or 2. that
task somehow panics despite its attempts to trap panics and report
them as errors through the oneshot.
So this "burn it all down" action ought only to occur in actually
serious situations.
I observe that
3ff9b187ea
Handle panics from circuit construction.
changed the EK for PendingCanceled to EK::ReactorShuttingDown,
and there's From impl. I think, therefore, that it is right
to reuse this Error variant.
I don't quite understand why when take_action gets an actual error it
doesn't push it, but just logs it. But I am not changing that for
now.
Arguably the two instances of retry_error.push are a sign of an
inferior flow control pattern - maybe the loop body including the code
I am adding ought to be an IEFE returning
`Result<Option<circ>, crate::Error>`.
Ian Jackson
Nick Mathewson
trinity-1686a
eta