Add DidDht create, publish and resolve, add unit tests

[KendallWeihe]

No description provided.

[nitro-neal]

so we don't publish by default? Should maybe talk to gabe, I think a while back we defaulted to published by default?

I'm 95% sure we will run into user error on this a lot. people will do default did dht create and wonder why it is not working for resolve

[diehuxx]

people will do default did dht create and wonder why it is not working for resolve

I would argue it is good for people to notice it is not working as they might expect and wonder why, causing them to learn that did:dht is a published did. If they do not understand that aspect of did:dht, they cannot make an informed choice of which did method to use.

[KendallWeihe]

it actually is set to publish by default as you can see here

I was following the web5-js design

but perhaps here I should reinforce that experience by seeing = true? what do we think? Rust doesn't have the equivalent syntax so I was following Rust

[KendallWeihe]

alright, that's changed in the latest commit. I think we should set it as true at that layer, even though the default is true inside the rust core code (as I linked to above), because developers often look at the function signature to determine default behavior. good catch!

[decentralgabe]

FWIW I have both separated in the server impl

create - https://github.com/TBD54566975/did-dht/blob/main/impl/internal/did/did.go#L119
publish - https://github.com/TBD54566975/did-dht/blob/main/impl/internal/did/client.go#L65

I'd create them as two different methods, but perhaps better to keep this as defaulting to true

[KendallWeihe]

oh I just realized the way I had it before, there are code paths where the default value you be false actually, even tho I'm using the default value of true inside the rust core code. jumping across languages is hard! thanks for the spot @nitro-neal

[nitro-neal]

looks great!

[decentralgabe]

are these test vectors implemented? https://did-dht.com/#test-vectors

[KendallWeihe]

are these test vectors implemented? https://did-dht.com/#test-vectors

@nitro-neal is following behind with test vectors

it's an interesting question tho, because those specific test vectors are distinct from our web5-spec test vectors, but maybe they shouldn't be?

[decentralgabe]

@KendallWeihe yes ... there has been discussion about this for a while 😄 ^{1 2 3 4 5 6 7}

Footnotes

1. https://github.com/TBD54566975/web5-go/issues/142 ↩

2. https://github.com/TBD54566975/web5-kt/issues/303 ↩

3. https://github.com/TBD54566975/web5-js/issues/497 ↩

4. https://github.com/TBD54566975/web5-kt/pull/334#discussion_r1714610864 ↩

5. https://github.com/TBD54566975/web5-rs/issues/105#issuecomment-2073648662 ↩

6. https://github.com/TBD54566975/web5-spec/issues/167 ↩

7. https://github.com/TBD54566975/web5-spec/issues/11#issuecomment-1821932704 ↩

[KendallWeihe]

@decentralgabe brought the receipts! 😆

We'll get 'er done here #310