readme.md

graphql to sql

** NOTICE **

This pacakge is NOT MAINTAINED ANYMORE.

Turn your Prisma GraphQL models into SQL tables without Prisma server.

Hook up adapter and do more magic.

Currently only Hasura is supported, more will come if needed.

Documentation

To generate local documentation run yarn docs in root of this module.

Online docs are located here

Debug && Development

For convenience .vscode folder is provided for easier debugging and development.

This will run the ./src/run.ts which is the way how I test the module. THis should not be imported or used directly, it is a dev/test file.

yarn
yarn start

Open VSCode and hit F5 to attach to the current running process. Enjoy debugging. :)

Logging

Debug npm package is used to do the logging.

Each of the Adapters and Parsers has it's own logger for convenience in debugging. To activate ALL logger set the environmental variable to DEBUG=GQL2SQL* and be amazed amount of logs you get :)

You can specify to log only parsers like DEBUG=GQL2SQL:parser* or just prisma parser like DEBUG=GQL2SQL:parser:prisma

Adapters are similar, instead of parser use adapter like this DEBUG=GQL2SQL:adapter*

Some Adapters like hasura have multiple loggers defined, adapter related and api related. If you wish to see them all, do following DEBUG=GQL2SQL:adapter:hasura* or just api DEBUG=GQL2SQL:adapter:hasura:api

Tasks

• DateTime must be correctly handled. especially createdAt and updatedAt
• ID! default CID, currently autoincrement

Notes

• Prisma uses CID for ID! type, implementation here is to use autoincrement INTEGER.
• UUID is defaulted to UUIDV4 and excepcts that DB supports uuid_generate_v4()

On Prisma definition models

relatedField is populated when relation is set on both types:

Example:

type A {
  id: ID!
  b: A @relation(name: "ABConnected")
}
type B {
  id: ID!
  a: A @relation(name: "ABConnected")
}

relationName is a name of the @relation directive, in above case is ABConnected