Technical Implementation

Project Structure

  • docs - folder contains all the md files for GitHub Pages
  • receipt_analyzer - main project folder
    • api - contains code for controllers for different operations
    • service - contains code to connect to azure form analyzer API and return the JSON results
  • sample_receipts - contains sample jpg files for testing


Setup

Prerequisite:

  • Python 3
  • pip

Setup on Local Windows Machine:

  • Clone the code from GitHub Repo
  • Run the following commands: 
     cd receipt_analyzer
 pip install -r requirements.txt

Setup on Digital Ocean:

  • Clone the code from GitHub Repo on local machine
  • Create a new CentOS Droplet
  • Transfer the code via ssh by executing following command: 
     scp -r ./receipt_analyzer root@<Droplet's I.P.>:~
  • SSH into the Droplet by executing following command: 
     ssh root@<Droplet's I.P.>
  • Prepare the droplet to run python code by executing following commands: 
     sudo dnf update -y
 sudo dnf install python3 -y
 sudo dnf -y groupinstall development
 sudo dnf install epel-release
 sudo yum install screen

Start the REST API:

If running on local windows machine

  • Replace IP in settings.py file with localhost
  • Run the following command: 
     python receipt_analyzer/app.py

If running on digital ocean droplet

  • Run the following command: 
     screen -S session1
 python3 receipt_analyzer/app.py
  • Use Ctrl+A+D to detach the screen session before closing command prompt. This will ensure the python server is up and running even if command prompt is closed.