abought / jam-setup

A basic script for bootstrapping JamDB namespaces and collections.

Description and Usage

This is a pretty minimal set of scripts for creating new namespaces and collections in JamDB.

Getting a token

Creating a new namespace requires a special token that needs to be generated via the jam CLI on the JamDB server itself. Running: jam token system-system-system will create a jwt (expires in 3 hours) that can be used in the steps below.

Run cp config/default.yml config/local.yml to create a local settings file. This gives something like:

stage:
  JAM_TOKEN: CHANGEME
  JAM_URL: https://staging-metadata.osf.io
  COLLECTIONS_PATH: ./collections.js

prod:
  JAM_TOKEN: CHANGEME
  JAM_URL: https://metadata.osf.io
  COLLECTIONS_PATH: ./collections.js

You'll need to replace the value of JAM_TOKEN with the token generated above (for the appropriate environment).

collections.js

The exports of this module will be module be used to determine which collections will be created under the bootstrapped namespace. This file should be structured like:

module.exports = [{
    id: 'experiments',
    attrs: {
         // optional: state: 'mongo',
         // optional: schema: {},
         permissions: {
              '*': READ
         }
    }
}, ...
];

Adding the state field will allow you to specify which backend JamDB will use for a given collection (generally this is not needed!).

Adding the schema attribute will allow you to provided a JSON Schema that JamDB will use to validate all new and updated documents.

Running

First: npm install

Open a node.js shell (node) run:

> var Client = require('./client.js');
undefined
> var client = new Client(<your_environment>);
undefined
> client.bootstrap(<your_namespace>, 'user-osf-<your-osf-guid>');