Parflow Performance Benchmark API

Description

This API is built using Flask with PyMongo and queries a MongoDB database to serve endpoints to the various React Apps.

Quick Start

Requirements:

• Python virtual environment (VENV/Conda)
  • Install necessary packages
• Store MongoDB connection link (When Developing):
  • ENV Variable

Follow these instructions for running while developing

Variables to note

• [env name]
  • Name of the virtual environment folder than python will create
• [Mongo Connection Link]
  • Link with valid credentials for MongoDB connection

1. Open terminal and move to project directory

• $ cd /path/to/parflow-performance-benchmark-flask-api

2. Create python virtual environment in root project directory

• Using VENV
  • Create a python virtual environment, this will create a folder to contain the environment
    • $ python3 -m venv [env name]
  • Activate the virtual environment
    • $ source [env name]/bin/activate
  • Install dependencies
    • $ python3 -m pip install -r requirements.txt

3. Store your provided MongoDB connection link

• ENV Variable
  • Export variable
    • $ export MONGO_CONNECTION_LINK="[Mongo Connection Link]"

4. Run the Flask API

• $ python3 app.py

Debugging

To run the API in debugging mode:

1. Open app.py in an editor
2. Add debug=True as a parameter in app.run([parameters])

Deployment Notes

• Deployment on Tuolumne uses Apache with mod_wsgi so using an environment var for the Mongo Connection Link will not work. Steps for prepping for deployment are below:
  1. Clone the repository onto Tuolumne
  2. Open mongo/mongo_client.py
  3. Find the line containing: connection_string = None
  4. Replace None with your Mongo Connection Link encased in quotations
     • The line should now look like: connection_string = 'mongodb://...'
  5. Copy repo files into the mod_wsgi folder, replacing the old deployment files

TODO

• Restructure to use a package like flask_restful for authentication and better endpoint management
• Improve endpoint definitons to follow a pattern such as /adapt/api/documents or /princeton/api/documents to differentiate between requests
  • Restructure file system to also differentiate with subfolders
• POSSIBLE: Use something like flask_mongoengine to restrict queries and to automatically escape POST data that may be malicious