Redis JSON + Search Restaurant Menu Demo

Contents

1. Summary
2. Features
3. Prerequisites
4. Installation
5. Scenario 1
6. Scenario 2
7. Scenario 3

Summary

This is a Jupyter notebook that loads a Kaggle dataset into Redis in two manners: string and JSON. Comparisons of string and JSON operations are made for three scenarios.

Features

• Loads Kaggle restaurant data
• Executes Redis operations on that data to solve various hypothetical business problems

Prerequisites

• Redis Enterprise database with Search and JSON enabled
• Kaggle account
• Python
• Jupyter

Installation

1. Clone this repo.
2. Edit the .env_template file with your credentials.
3. Copy or rename that file to .env
4. Follow notebook steps

Scenario 1

Business Problem

Given a known restaurant ID and a menu item name, find its price.

String

restaurant_str = client.get(f'restaurant_str:{restaurant_id}')
restaurant_json = json.loads(restaurant_str)
menu = restaurant_json['menu']
for item in menu:
    if item['name'] == menu_item:
        price = item['price'] 
        break

JSON w/Search

query = Query(f'@id:[{restaurant_id}, {restaurant_id}]')\
    .return_field(f'$.menu[?(@.name=="{menu_item}")].price', as_field='price')
result = client.ft('restaurant_idx').search(query)

Scenario 2

Business Problem

Find the number of Papa Johns restaurants within a 100 mi radius of Madison WI

String - Exercise left for the Reader

• Loop thru the Redis key space
• Deserialize each string to JSON
• Perform a string comparison on the restaurant name element, covering letter case variations
• For restaurant name string matches, perform a Haversine calculation
• Update local counter for Haversine matches

JSON w/Search

query = Query(f'@name:"Papa Johns" @coords:[{madison} 100 mi]').paging(0, 0)
count = client.ft('restaurant_idx').search(query).total 

Scenario 3

Business Problem

Find the Top 3 menu items by count in the State of Texas

String - Exercise left for the Reader

• Loop thru the Redis key space
• Deserialize each string to JSON
• Perform a string comparison on the restaurant address element
• For restaurant address string matches, initiate a inner loop on the menu elements - cover letter case + stem variations
• Maintain counters for every possible menu item. Increment applicable counter per menu element
• Sort counters, display Top 3 items

JSON w/Search

request = AggregateRequest('@address:TX')\
    .group_by('@menu_item_name', reducers.count().alias('item_count'))\
    .sort_by(Desc('@item_count'))\
    .limit(0,3)
result = client.ft('restaurant_idx').aggregate(request)