*This repo is highly-dated and dependencies have known-security vulnerabilities.

Bluecard is a a prototype ID scanning system.

It's designed to replace manual registration workflows, where a clerk is handed a license, and hand-types information contained on it.

This project was created to:

1. Learn Django
2. Explore Amazon's AWS Lambda processing framework
3. Experiment with the Captricity document processing API

Typically registration processes involve a clerk reading a license, and typing the information into an EHR or other proprietary system.

Instead the general workflow is:

1. The license is scanned or photographed by the clerk, or even by a provider or other care-giver
2. The license image is stored (in this case in an Amazon S3 bucket)
3. A program notices the new image and sends it to Captricity (Amazon Lambda)
4. Data is extracted and stored (Amazon RDS)

Ideally, the data would be extracted and stored by a seperate Lambda process, currently that is partially implemented in the Django layer.

The compute intensive portions of the app are almost entirely offloaded to Amazon and Captricity.

Django expects a config file in config/config.cnf.

Here's an example:

[django]
secret_key = mydjangosecret

[mysql]
database = bluecard
user = bluecard
password = mypassword
default-character-set = utf8
host = my-datastore-or-rds
port = 3306

[aws]
keyid = myamazon
key = key
bucket = 'bluecard'

[captricity]
apitoken = myapitoken

The AWS lambda worker lives in the aws-lambda directory.

Deploying the lambda worker requires building and uploading a ZIP of the aws-lambda directory. See the shell script create-aws-package.sh. This would ideally be automated to avoid having to use the AWS GUI.

The script depends on aws-lambda/config/config.cnf, which can be identical to whats in the main config directory.