Add feature for renaming and deleting aids

[rodolfomiranda]

PR that follows KERIA PR #218

[codecov]

Codecov Report

Attention: Patch coverage is 95.00000% with 1 line in your changes missing coverage. Please review.

Project coverage is 82.94%. Comparing base (4c0072f) to head (c67f41b).
Report is 28 commits behind head on main.

Files with missing lines	Patch %	Lines
src/keri/app/credentialing.ts	80.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #241      +/-   ##
==========================================
+ Coverage   82.88%   82.94%   +0.06%     
==========================================
  Files          46       46              
  Lines        4212     4217       +5     
  Branches     1040     1041       +1     
==========================================
+ Hits         3491     3498       +7     
- Misses        692      694       +2     
+ Partials       29       25       -4     

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

[rodolfomiranda]

integrations test will fail until KERIA PR is merged

[lenkan]

Nice!

Open question... since we are changing the API anyways. Is there currently any benefit of using the event type as an object field key rather than just relying on the event payload t property? I'll illustrate with an example what I mean:

Currently:

POST /identifiers/name/events
{
  "rot": { ...rot payload },
  "sigs": [...sigs]
}

POST /identifiers/name/events
{
  "ixn": { ...ixn payload },
  "sigs": [...sigs]
}

Rather than:

POST /identifiers/name/events
{
  "event": { t: "rot/ixn", ...rot/ixn payload },
  "sigs": [...sigs]
}

The switch on backend would be to check body.event.t === "ixn" or body.event.t === "rot".

[lenkan]

@rodolfomiranda There is something wrong after the upgrade to 0.2.0-dev0. I tried to debug the multisig-vlei test and found that after the client of GAR1 resolves the oobi for the multisig QVI aid, they still don't have the QVI aid in the contact list.

To get this merged we need to figure out what the problem is there. Perhaps it would be better to create a separate PR for just the upgrade to 0.2.0-dev0.

[lenkan]

@rodolfomiranda Did a little bit more digging, and what we get back from the long running operation for resolving the QVI AID is just ({ oobi: "oobi url" }), rather than the expected response of the current state of the contact. This only happens here:

https://github.com/WebOfTrust/keria/blob/7ee5ba84c9aed5d2d21fa48d8241d39b06aadb93/src/keria/core/longrunning.py#L239, which only happens if the obr.cid is not in kevers. Any ideas?

[rodolfomiranda]

Any ideas?

I've been debugging a lot and saw the same but also found that the problem starts to happens after a new ixn or rot event is applied to the multisig. In the multisig-vlei-issuanse test, that happens after the ixn for the delegation in GEDA aid; in the multisig test that happens after rotating the multisig aid. In both cases, if you comment out the event (the delegation in the vlei test or the rotation in the multisig test), the problem not longer appears. If not, seems that the habs got corrupted.

I think it's related to the changes that were applied in keripy and keria to track multisig states, such as the smids and rmids as states. It works well at inception, but something is wrong after and ixn or rot event. I'm still troubleshooting...

[lenkan]

Any ideas?

I think it's related to the changes that were applied in keripy and keria to track multisig states, such as the smids and rmids as states. It works well at inception, but something is wrong after and ixn or rot event. I'm still troubleshooting...

Cross posting for visibility.

There are probably more issues then. In #252, I managed to reproduce a similar situtation for single sig only, where the problem is that a delegator cannot resolve an oobi of the delegatee.

[lenkan]

@rodolfomiranda I believe this has been superseded by a bunch of other merged PRs now. Should we close this one?