Configurable schemaId about business-partner-agent

[frank-bee]

@domwoe @etschelp
Thanks for the fruitful discussion.

For the time being we will impl. the following

1. schemes can be added via configuration or via API
2. the ones added via configuration are read-only and cannot be modified via API but only configuration
3. adding via configuration is not additive but overwrites ("set behavior").
4. configuration format is like this (map of "schema id : label")

oagent:    
  schemes:
    M6Mbe3qx7vB4wpZF4sBRjt:2:bank_account:1.0 : "Bank Account"
    M6Mbe3qx7vB4wpZasdfsdaf4sBRjt:2:foo:1.0 : "Foo"

5. labels can contain any chars , incl whitespaces
6. schema versions are ignored (they are implicitly part of the schema id, that's it)

[domwoe]

Needs more explanation.

• Should this allow to add additional schemas/credential types?
• Should this only allow to set schema ids for different ledgers?

[domwoe]

I'd like to register new schemas in this way. These new schemas should be usable as templates for documents.
We can even add them to the public profile by generating an ad hoc context if no explicit context mapping is provided.

We could do it like this:

{
          "@context": { 
            "id": "@id",
			"type": "@type", 
            "sc": "did:sov:iil:1242415245#", //schemaId
            "attribute": {                   // for each attribute in schema 
              "@id": "sc:attribte"  
            }
          }
        }
        

Example : AL3qjjhFYKgChRmWweMibV:2:eid_min:0.1

Attributes:

• familyName
• name

{
          "@context": { 
            "id": "@id",
            "type": "@type", 
            "sc": "did:sov:iil:AL3qjjhFYKgChRmWweMibV:2:eid_min:0.1#", 
            "familyName": {
              "@id": "sc:familyName"  
            },
            "name": {
              "@id": "sc:name"  
            }
          }
        }
        

This explicit context would be added in addition to the IndyCredential context

[frank-bee]

For simplicity, I thought of the following impl ( at least for the time being) :
the schemes added via configuration will be added to the database as well - with a flag "read-only".
During init phase of the app, all these "read-only" schemes will be purged in the DB and re-written.
We can later refactor this...