JSON Schema

[PrettyGoodFreedomTech]

This is a thread to discuss the role that json schema may play in the grapevine.

I anticipate using ajv validator to validate schemas. I've used it before and it's very widely used. Other options can be found at the json-schema website under tools.

Because of the grapevine's modular structure, there's going to be a lot of back and forth communication over APIs. Messages need to be formatted properly in json and validated. As of right now:

• each interpretationProtocol will require its own json schema for the object used to encode and communicate adjustable interpretation protocol parameters over the Calculation / Interpretation API. These schemas will be stored in coreTable1 (interpretationProtocols) of the (interpretation engine db)[https://github.com/Pretty-Good-Freedom-Tech/grapevine-modular-structure/blob/main/interpretation-engine/nostr/database-initialization-core.sql]. These SAME schemas must also be stored in coreTable5_nostr (interpretationProtocols_nostr) of the calculation ending db.
• each graperankProtocol in coreTable6 (grapeRankProtocols) of the calculation ending db will be associated with its own json schema. These schemas will eventually be communicated over the Endpoint API.

Who decides on the proper format? Ultimately, your grapevine will decide. Until then, we will do it.

Where will we store the json schemas? They're going to change a lot, at least in the early days while we hammer out the protocols. I'm thinking the best solution is to store them in kind 3xxxx notes and address them by naddr. There will be one for each grapeRank Calculation Protocol and one for each Interpretation Protocol. (Any others?) So for starters, I'll probably use this discussion thread to maintain a list of all relevant naddrs. Ultimately, I will create a concept for json schemas using the concept graph protocol (once I get brainstorm.ninja refactored using the PGFT NRD Template and organize the relevant naddr's using that concept, which means I won't need this discussion thread anymore, you'll just look for the concept of json schemas and find the list there.

interpretation Protocol json schemas

One naddr is needed for each of the following:

• basicFollowsInterpretationProtocol
• basicMutesInterpretationProtocol
• basicReportsInterpretationProtocol
• expandedReportsInterpretationProtocol
• brainstormNotSpamInterpretationProtocol

These naddrs will be stored in the tables:

• interpretation engine coreTable1 (interpretationProtocols) parametersJsonSchemaNaddr
• also calculation engine coreTable5_nostr (interpretationProtocols_nostr), col: parametersJsonSchemaNaddr

grapeRank Protocol json schemas

One naddr is needed for each of the following:

• basicGrapevineNetwork
• basic5StarProductCalculation

These naddrs will be stored in the table:

• calculation engine coreTable6 (grapeRankProtocols), col: parametersJsonSchemaNaddr

[PrettyGoodFreedomTech]

One more class of json schemas that I did not mention in the above post: each API, the Calculation / Interpretation API, the Endpoint API, and the Client API, will have a request and a response which must be properly formatted and so will require a reference json schema.

[PrettyGoodFreedomTech]

I am gathering all json schemas here.