proxied/saved/mocked environment to test and develop microservices

You can install it globally

npm -g autarchy

or as a dependency

npm -D autarchy

Browse the recent requests

Allow other to access your local services

mkdir .autarchy

cd .autarchy

echo module.exports = {} > config.js

The file .autarchy/config.js will eventually allow to set default configuration for all the services. Currently it's not used.

Create a folder for each service, with a config.js inside

.autarchy/foo/config.js

module.exports = {
  type: 'REST',
  remote: 'https://example.com/api',
  local: {
    ip: '127.0.0.1',
    port: 5000,
  }
}

The type of the service. The provided types are REST and graphql

Custom types can be added.

It defaults to the folder name.

It's used to determine the collection's name ${type}-${name}

In the example the name is 'foo' and the collection name is 'REST-foo'

The url of the service that you want to proxy

It is not required if you are mocking all the endpoints from the database or a pre or post function.

Where to bind the local service.

• local.ip defaults to '0.0.0.0
• local.port defaults to 80

(context) => context

it's executed just after matching the endpoint.

It can be used to modify the request that will be used to search on the database or to be sent to the proxy

It also can set a response, or modify the configuration.

Please refer to the service type documentation to further instructions regarding the context and the response

[(context) => ({ some mongo query }), ...] | false

Requests are cached so after a first fetch you don't need the remote service to be up and running.

You can use pre to set context.conf.toQuery to false to disable the cache search

Otherwise it will loop the array running query after query until it returns a response.

Please refer to the service type documentation to further instructions regarding the defaults and the context schema regarding the request.

(context) => document

You can use pre to set context.conf.toDocument to false to skip inserting the request and response in the database

Please refer to the service type documentation to further instructions regarding the defaults and the context schema regarding the request.

(context) => context

it's executed just before sending the response.

It can be used to modify the response that will be sent

Please refer to the service type documentation to further instructions regarding the context and the response

Requests are also logged, you can disable that setting skipLogs thuthy.

If you want an specific configuration for a endpoint, you need to create a file whose name is the http method (or all), and the file path is the endpoint/ + the endpoint path

For instance: This file .autarchy/foo/endpoints/some/endpoint/GET.js will override .autarchy/foo/config.js when the request is a GET to /some/endpoint

You can use parameters between brackets.

For instance this file .autarchy/foo/endpoints/some/endpoint/[id]/all.js will override .autarchy/foo/config.js when a POST is sent to /some/endpoint/bar or when a GET is sent to /some/endpoint/buz

{
  "request": {
    "path": "",
    "params": {},
    "headers": {},
    "method": "",
    "body": {},
    "query": {}
  },
  "conf": {
    // ...foo/config.js,
    // ...foo/endoints/path/METHOD.js 
  },
  "response": {
    "status": 200,
    "headers": {},
    "body": {}
  }
}

If you want to override .autarchy/foo/config.js in some queries you can create any file at any path inside .autarchy/foo/queries/

For instance

.autarchy/foo/queries/bar.js

module.exports = {
  match: (context) => true // false 
}

It will loop over the files under queries/ until some match returns truthy

First one that matches, will override the configuration

{
  "request": {
    "headers": {},
    "method": "",
    "body": {},
    "query": {}
  },
  "conf": {
    // ...foo/config.js,
    // ...foo/queries/baz.js 
  },
  "response": {
    // the graphql response
  }
}