feat(golem-network): added getIdentity method to GolemNetwork object

[mgordel]

No description provided.

[grisha87]

We shouldn't blindly forward the data without any documentation this way. It's not clear for the user of this API what glm.getIdentity().identity means as it's identity->identity and there's not tsdoc that explains particular fields to the user.

I've tested the code and get the following result from the getIdentity():

{
  identity: '0x46b1...',
  name: 'eth-warsaw-workshop-test',
  role: 'manager'
}

This is roughly similar to output of yagna app-key show eth-warsaw-workshop-test --json:

{
  "allowOrigins": [],
  "createdDate": "2024-08-30T08:29:07.361270875",
  "identity": "0x46b1...",
  "key": "ec...e3",
  "name": "eth-warsaw-workshop-test",
  "role": "manager"
}

I guess that the naming of things is broken on yagna's API and in fact we have a mixture of identity and app-key concepts, where an app-key belongs to an identity.

As far as I'm aware, "identity" of the "Golem User" is established by his ETH address. This is reflected by yagna id show:

{
  "Ok": {
    "alias": null,
    "deleted": false,
    "isDefault": true,
    "isLocked": false,
    "nodeId": "0x46b1..."
  }
}

So actually, I'd rather expect from glm.getIdentity(): string -> "0x46b1..." instead of the object which melts together the identity and app-key information.

[mgordel]

Yes, it's a bit confusing. I wanted to be consistent with yagna api. And the entity should probably contain fields: {nodeId, name, role} -> https://github.com/golemfactory/ya-client/blob/master/model/src/identity.rs

but for our purposes I agree that a single nodeId string might be better