Update DID DHT Key Type to EdDSA

[nitro-neal]

This bug fixes this issue and changes the index 0 key type from Ed25519 to EdDSA (pending change coming to did-dht.com)

Here was the issue that was happening which this pr fixes:

When creating a portable did with a jwk like this:

        privateKeys: [
          {
            crv : 'Ed25519',
            d   : 'hdSIwbQwVD-fNOVEgt-k3mMl44Ip1iPi58Ex6VDGxqY',
            kty : 'OKP',
            x   : 'VYKm2SCIV9Vz3BRy-v5R9GHz3EOJCPvZ1_gP1e3XiB0',
            kid : 'cyvOypa6k-4ffsRWcza37s5XVOh1kO9ICUeo1ZxHVM8',
            alg : 'EdDSA',
          }
        ]

(notice Alg is EdDSA)

the resolver gives:

    "verificationMethod": [
      {
        "id": "did:dht:ksbkpsjytbm7kh6hnt3xi91t6to98zndtrrxzsqz9y87m5qztyqo#0#0",
        "type": "JsonWebKey",
        "controller": "did:dht:ksbkpsjytbm7kh6hnt3xi91t6to98zndtrrxzsqz9y87m5qztyqo#0",
        "publicKeyJwk": {
          "kty": "OKP",
          "crv": "Ed25519",
          "x": "VYKm2SCIV9Vz3BRy-v5R9GHz3EOJCPvZ1_gP1e3XiB0",
          "kid": "cyvOypa6k-4ffsRWcza37s5XVOh1kO9ICUeo1ZxHVM8",
          "alg": "Ed25519"
        }
      }
    ],

(notice alg is Ed25519)

The solution was to change the actual did dht spec so that this does not happen and allows use to verify jwts with did dht

[changeset-bot]

🦋 Changeset detected

Latest commit: 62a0a73

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
@web5/credentials	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[nitro-neal]

is this fine?

[github-actions]

TBDocs Report

✅ No errors or warnings

@web5/api

• Project entry file: packages/api/src/index.ts

@web5/crypto

• Project entry file: packages/crypto/src/index.ts

@web5/crypto-aws-kms

• Project entry file: packages/crypto-aws-kms/src/index.ts

@web5/dids

• Project entry file: packages/dids/src/index.ts

@web5/credentials

• Project entry file: packages/credentials/src/index.ts

---

TBDocs Report Updated at 2024-06-11T22:50:23Z 62a0a73

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 90.90%. Comparing base (2fc32ba) to head (62a0a73).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #653      +/-   ##
==========================================
+ Coverage   90.81%   90.90%   +0.09%     
==========================================
  Files         119      119              
  Lines       30101    30106       +5     
  Branches     2243     2249       +6     
==========================================
+ Hits        27337    27369      +32     
+ Misses       2729     2702      -27     
  Partials       35       35              

Components	Coverage Δ
agent	80.85% <ø> (ø)
api	94.01% <ø> (ø)
common	98.68% <ø> (ø)
credentials	94.88% <ø> (ø)
crypto	93.81% <ø> (ø)
dids	97.77% <100.00%> (+0.44%)	⬆️
identity-agent	96.70% <ø> (ø)
crypto-aws-kms	100.00% <ø> (ø)
proxy-agent	96.70% <ø> (ø)
user-agent	96.70% <ø> (ø)

[thehenrytsai]

Not a fan of opaque inputs to tests. Can you update and enable vector tests accordingly for the js implementation and call out in #638 that the kid changes must be reverted when 638 is fixed.

I think they should cover this change. Let me know if I am not making sense!

[nitro-neal]

This test tests the Jwt.sign / verify with did dht, (the reason we had to change the spec a bit in the first place)

[thehenrytsai]

Got it, assuming the main intent of this test is to test sign and verify, can we avoid using the opaque hex string in the test since it is besides the point?

[nitro-neal]

I have turned the test vectors back on, they can pass by either:

1. what i've done here in the pr, changing the kid in the test vector
   OR
2. I can add this line to our fromDnsPacket parsing:

          if(dnsRecordId === 'k0') {
            publicKey.kid = '0';
          }

Not sure which way is right at this point! but feel free to let me know which way we want to go

[nitro-neal]

I have updated,

added this line:

   if(dnsRecordId === 'k0') {
     publicKey.kid = '0';
   }

And chnaged the test vectors to have kid as 0 again

[thehenrytsai]

🚀