DID Exchange Protocol

[mirgee]

Initial version of DID Exchange protocol implementation with support for peer DIDs (peer:did:2 for now), new DidDocument crate, and JWS signing / verification.

The primary aim is not to achieve perfection at this stage but to roll out a first iteration prototype.

Correctness has been tested (so far only) using AATH tests running against AcaPy peer:did branch (or against aries-vcx itself), where all didx tests are passing whether the requester or responder role is assumed by either side.

Additionally

• Changes made later on to revive this PR, merged into this: Did-exchange PR revival #1046

[Patrik-Stas]

had a chance to have a short look only so far, but left some comment and thouhts

[Patrik-Stas]

I'd suggest we split this into

GenericDidExchangeRequester
GenericDidExchangeResponder

The value proposition of the generic layer is that you don't have to care about what state the you are currently in. In current state, it technically allows you to be even more agnostic - not only in terms of states, but also in terms of roles. But in practice, that property would never be used.

If we proceed with proposed split, there's hardly any practical disadvantage for consumers, but code will be a lot simpler, as we get rid of the outer level of match statements.

And basically, you will end up with place where you actually don't have to implement handle_response for GenericDidExchangeResponder at all, because it doesn't make sense.

This will have some implications, because on libvcx layer we've previously never distinguished between Inviter/Initee, but such approach was only used for connections, not other protocols. Since we are building on greenfield here, I think we should get rid of this discrepancy.

[mirgee]

I agree in principle with it not making sense in terms of methods shared between the roles, but this is a quick-and-dirty implementation and the current approach allows for braindead simple handling on the aries-vcx-agent level which was the entire point of the enum to begin with. (I have no issues with getting rid of the discrepancy, though.)

[Patrik-Stas]

Got it, I'd like to go ahead with that

[codecov-commenter]

Codecov Report

Attention: 979 lines in your changes are missing coverage. Please review.

Comparison is base (61b7cd1) 0.05% compared to head (875519b) 0.05%.

Files	Patch %	Lines
...rotocols/did_exchange/state_machine/generic/mod.rs	0.00%	130 Missing ⚠️
aries_vcx/src/utils/mod.rs	0.00%	114 Missing ⚠️
...rc/protocols/did_exchange/state_machine/helpers.rs	0.00%	111 Missing ⚠️
...ge/state_machine/requester/request_sent/helpers.rs	0.00%	75 Missing ⚠️
...hange/state_machine/responder/response_sent/mod.rs	0.00%	69 Missing ⚠️
aries_vcx/tests/test_did_exchange.rs	1.56%	63 Missing ⚠️
...change/state_machine/requester/request_sent/mod.rs	0.00%	56 Missing ⚠️
aries_vcx/src/protocols/oob/mod.rs	0.00%	55 Missing ⚠️
aries_vcx/src/protocols/did_exchange/mod.rs	0.00%	40 Missing ⚠️
aries_vcx/src/protocols/connection/invitee/mod.rs	0.00%	38 Missing ⚠️
... and 39 more

Additional details and impacted files

@@           Coverage Diff            @@
##            main    #928      +/-   ##
========================================
- Coverage   0.05%   0.05%   -0.01%     
========================================
  Files        388     471      +83     
  Lines      21076   24009    +2933     
  Branches    3871    4307     +436     
========================================
+ Hits          12      13       +1     
- Misses     21063   23995    +2932     
  Partials       1       1              

Flag	Coverage Δ
unittests-aries-vcx	0.05% <0.20%> (-0.01%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[mirgee]

@gmulhearn @bobozaur I don't know why @Patrik-Stas requested a review, this PR is not ready for review.

The code is relatively independent on the rest of the codebase so there is a limited value in merging this prematurely, before the implementation is decent, signature conventions are defined, the other implementations are flashed out and tested against...

[mirgee]

We should be able to group the messages crates after this is merged