This repo contains code for Runway, a software system for integrating data to Ed-Fi. It can be deployed locally via Docker or in AWS.
Runway is a friendly UI over our open source tools Earthmover and Lightbeam for loading flat files (particularly assessments) to an Ed-Fi ODS. But note that Runway itself is not open source, but non-commercial. It is free to use for Education Agencies and non-profits, but requires a license for for-profit entities. See our license for details.
Design
The system consists of several components:
- the Application frontent and backend (see
app/), with which users interact to set up Ed-Fi connections and run integration jobs
- the Executor (see
executor/), which executes Jobs, transforming input data and sending it to an Ed-Fi API
The Executor uses earthmover to transform input data and lightbeamto send it to an Ed-Fi API.
Usage
You can deploy this app in AWS or locally. Once deployed, see application features for tips on how to use the app.
AWS deployment
Prerequisites:
- You must have an AWS account
- You must have deployed a custom VPC, 2 public subnets, and 2 private subnets before deploying the Runway templates. It is not advisable for users to use the default VPC initially deployed by AWS.
- You must have created a Hosted Zone and NS record for the domain at which you intend to deploy your Runway environment(s).
- You must have created a GitHub Connection in AWS CodePipeline between your GitHub and AWS accounts.
If you would like more guidance and support with deploying Runway in AWS, please reach out to our team using this form.
AWS Console Deployment Steps:
- In your AWS account, navigate to S3 and create an S3 bucket with a unique name that will be used to hold the Runway CloudFormation templates. Example bucket name:
{OrgName}-{ProductName}-{EnvironmentType}-cloudformationeducationanalytics-runway-dev-cloudformation
- Fork the Runway repository under your own Github account and use the
main branch.
- Upload the contents inside of the cloudformation folder into your S3 bucket. Make sure the S3 bucket contains both
lambdas and templates folders with their respective files inside.
- Copy the S3 URL for the
templates > 1-main.yml file.
- Navigate to the CloudFormation console in AWS. Create a new stack with new resources.
- Select
Choose an existing template > Amazon S3 URL and paste your copied URL for the 1-main.yml file into the field. Click Next.
- Enter stack parameter values. Please read the doc here for more information on parameter values and descriptions.
- Navigate through the next screens on the console until you are able to start the deployment.
Local deployment
You can run Runway locally and use it to load to your ODS without any dependence on AWS.
See app/README.md for full local instructions
External API
Runway exposes an API for programmatic job submission, allowing external systems to create and start jobs without using the UI. The API requires OAuth2 bearer tokens from a configured identity provider. See the External API README for configuration and usage instructions.
Application features
Coming soon...
This repo also contains a folder bundleSchemas/ which contains several examples of JSON schemas (which a bundle could provide) that would be turned into form elements on the "create job" part of this app, to collect configuration settings to be passed into the bundle.
Such schemas can be turned into form elements using (for React) react-jsonschema-form or (for jQuery) jsonform, jsonToForm, or json-editor.
