Feedback Collection System

Overview

This is a minimilistic web application that provides the user to provide feedback.

Features

1. Submit feedback
2. View all feedbacks

Stack

1. Python Tornado as Web server and framework (Python 3.7, Tornado latest)
2. VueJS for frontend scripting (VueJS latest)
3. MySQL as database service (MySQL 5.7)
4. Minio for log backups (Minio latest)

How to run the application

If you are using docker, it becomes really simple to bring up the application and its required components.

git clone https://github.com/gbakthavatchalam/feedback-system.git
cd feedback-system/
docker-compose up

How to view the application

• For application, Open the browser by pointing to http://localhost:8888
• For Minio, Open the browser by pointing to http://localhost:9000. Creds from docker-compose file can be used for logging in.

Frontend Endpoints

"/" GET Lets you to submit feedback
"/list" GET Lets you to view all feedbacks

Backend APIs

"/api/feedback" GET - Gets list of all feedbacks
"/api/feedback" POST - Saves the feedback

Warning

1. MySQL & Minio credentials are stored in the docker-compose.yml
2. Bucket names for Minio can be configurated in settings.py
3. Log settings can be configured in settings.py

Future Work

1. Credentials can be moved and passed in as environment variables to the docker-compose.yml file
2. Optimize the frontend to be a SPA(Single Page Application) by using client routing (Vue Router)
3. For tornado webapp, use a config file and parse options from config file rather than settings.py
4. Application needs some fine-tuning for better exception handling and reporting.

Log Extraction

1. In backend, logs are backed up in 2 different ways.
2. One is by implementing a custom TimedRotatingFileHandler which saves the log file to Minio before it tries to clean up the old files.
3. Other way, we are storing metadata related to individual requests as JSON files in a separate bucket on Minio