Skip to content

Latest commit

 

History

History
97 lines (71 loc) · 3.95 KB

README.md

File metadata and controls

97 lines (71 loc) · 3.95 KB
Raw

Conduits backend services

Backend consists of a resource server and a gateway server.

The resource server exposes a REST api to manage users and their conduits. A conduit is an endpoint serviced by a gateway server.

The gateway server provides a unified and normalized API to the conduit. Requests to the conduit are transformed as necessary and forwarded to the down stream service provider. The response from the service provider is likewise transformed as necessary.

Development

Getting started

Two .env files are required to get started. Both of these files should be present in the backend folder relative to the root of the repository.

Copy .env-example as .env
Copy .env.conduit-user.example as .env.conduit-user

  1. A global .env file containing client credentials for the gateway-server. For expediency, the gateway is treated as a special user. In Oauth terminology, the gateway is a confidential client!

    • GATEWAY_SERVER_EMAIL
    • GATEWAY_SERVER_PASSWORD

    The above values will be used to authenticate the gateway
    Refer to the .env-example for more accurate information on what variables are supposed to be in the actual global .env file.
    Notes:

    1. Create and populate this file before starting the crud-server; otherwise, it will fail to start.
    2. This is a workaround until we have proper user management in place.

  2. A .env.conduit-user that contains credentials related to the functioning of the conduit service. This includes information about the external service and user details. This is required for testing the gateway server.

       CONDUIT_SERVICE_TYPE=one of {airtable|googleSheets|email}
   CONDUIT_SERVICE_API_KEY=do not share your secrets
   CONDUIT_SERVICE_OBJECT_KEY=variable portion that identifies the object

  3. Integration tests require bootstrap data created by unit tests. So, npm run test-model first before testing the REST api. Alternatively you can also run npm run createdb.

Running the servers

  1. Populate test data by running test-model
  2. Start the resource server by running start-resource-server
  3. Start the gateway server by running start-gateway-server

Debugging

Error responses and stack traces can be logged to the console by setting the DUMP_ERROR_RESPONSE and DUMP_STACK_TRACE environment variables. The features can be enabled by prepending the environment variable with the npm task command.

DUMP_ERROR_RESPONSE=1 DUMP_STACK_TRACE=1 npm run <task-name>

Developer Tasks

task command line
Install dependencies npm install
Run linter npm run lint
Fix lint errors npm run lint:fix
Run data layer tests npm run test-model
Run data layer tests with code coverage npm run test-model-with-coverage
Run REST api tests npm run test-rest
Run REST api tests with code coverage npm run test-rest-with-coverage
Run gateway tests npm run test-gateway
Run resource server npm run start-resource-server
Run gateway server npm run start-gateway-server
Init db with user: admin@praas.com, pwd: praas npm run createdb

Further reading

Resource server API documentation

Run the script npm run api-doc, browser will open resource-server.html page.