GrAMPS GraphQL Data Source Base
A boilerplate and minimal example for a GrAMPS data source.
Quickstart
Set up a local data source in seconds with:
npx graphql-cli create -b gramps-graphql/data-source-base data-source-mydata
cd $_
yarn dev
NOTE: We recommend prefixing data source projects with data-source-
for clarity and the eventual support of CLI tools to add data sources to your gateway. So if you’re creating a user management data source for the Acme company, we recommend data-source-acme-users
or data-source-acmeusers
as a directory/repo name.
ALSO NOTE: $_
is a handy shortcut for using the last argument passed to the previous command. It also does other stuff, but that's a rabbit hole for another time.
After running yarn dev
, you’ll see a message with URLs for the GraphQL gateway and the GraphQL Playground. Open the Playground link (usually http://localhost:8080/playground if you don’t already have something running on port 8080), then run a query:
{
getById(id: 123) {
id
name
lucky_numbers
}
}
To Develop with Mock Data
Add the --mock
flag to enable mock data, which is helpful for working offline.
yarn dev --mock
See src/mocks.js
to modify your mock resolvers.
NOTE: For more information on the GrAMPS CLI and its available options, check out the docs.
To Run the Tests
GrAMPS data sources start you off with 100% test coverage so you can build high-reliability GraphQL servers without a bunch of setup work.
Run the tests with:
yarn test
What's Inside?
Inside, you’ll find:
- Schema — type definitions for GraphQL to interpret the data (see the
GraphQL docs on schemas)
- Resolvers — functions to map the results of calls to model methods to
the schema
- Mock Resolvers — mock functions for offline development
- Context — an object with methods to interact with data that is passed to resolver functions
Each file contains a TODO
comment explaining the changes you’ll need to make to create a working data source.
The goal of this repo is to provide enough code to allow a working example of a data source and its related tests, but to limit how much boilerplate needs to be edited to get your own data source implemented.
Code Quality and Continuous Integration
To help ensure a reliable, easy-to-maintain data source, this example also includes:
Notes for Developers
Currently, there is no watch capability (PRs welcome!), so the service needs to be stopped (control
+ C
) and restarted (yarn dev
) to reflect new changes to the data source.