bewest / smart_hv_merge_app

FIXME

• URL paths for running locally (via "My App" in the SMART sandboxes)

  • /smartapp/index.html (route) ** needs to be the same as in the manifest

• af logs smart-hv-merge --all sometimes works, wait a few minutes if not and try to restart from AF control panel.

• Create a second "mappings" table and don't overload requests

SMART Glucose Logbook: Clinican App

Arjun Sanyal [email protected]

Implements a simple SMART REST app that establishes a connection to a user using the HealthVault [Patient Connect][] API. Each request is stored in a persisent SQLite database and is later updated with the HealthVault patient and record ids by a script that polls HealthVault for authorized requests. The script is run manually, but could be automated.

Once the connection is established it shows a set of patient-reported data (weights) and clinical data from the SMART Container in one interface.

Patient Connect

• Requirements

  • Python 2.7 or above
  • web.py
  • SQLite3

• To setup run python setup.py to create the SQLite3 request database in /data

• To run python merge.py 8000

• Then log in to the reference container: http://sandbox.smartplatforms.org

• Launch the "Developers Sandbox" app, which will point at your locally hosted version of this app

---

SMART Glucose Logbook Patient App

Arjun Sanyal [email protected] Josh Mandel [email protected]

This is the SMART patient-facing glucose logbook app. It runs as a standalone Python WSGI app and includes metadata to make running on AppFog simple but it can be run easily on other systems. It integrates with Microsoft HealthVault for authentication and storage of the patient's personal health data via the healthvault_py Python library included as a git submodule. Both reading and writing of data to the patient's HealthVault account is supported.

The app is written with the support of the AngularJS "framework" and the code is organized using a stripped-down version of the angular-seed sample repository. The visualization are written using the D3 JavaScript visualization and interaction library.

TODO

• Document HealthVault setup

Directory Structure

• The /app, /config, /logs, /scripts, /test directories are from https://github.com/angular/angular-seed

AppFog Setup

• requirements.txt

  • This file describes the dependecies (including versions) for this app. Without it AppFog's virtual server's wouldn't have the libraries installed that this app needs.

    flask==0.8 pycrypto==2.6 lxml==2.3.4

• manifest.yml

  • Defines the configuration parameters for the desired AppFog instance
  • Change the file for your preferred settings

Running Locally

You must have root permissions to run the app locally since it is configured by default to listen on port 80. You can change this in the wsgi.py script. To start on localhost port 80:

$ sudo python wsgi.py

Running on AppFog

Assuming you have the AppFog commadline app installed and are logged in using it, updating and starting the app should be simply:

$ af apps
$ af push <your-app-name>
$ af update <your-app-name> --debug
$ af logs <your-app-name> --all