Skip to content
Project for Google Apprenticeship Interview
Python HTML JavaScript
Branch: master
Clone or download

Google Project Challenge - Data Display and Search

Find and view data about pets!


  • Search through the adoptable pets database
  • Autocomplete
  • Address geocoding


  • Tested on Python 3.7, so let's go with that
  • Deps in requirements.txt (see below)
  • This was developed on both Windows 10 & Debian 9 Stretch, so presumably any Unix system will be fine
  • Network connection if geocoding enabled

Installing dependencies

sudo apt-get install python3 python3-pip pip3 install -r requirements.txt


Running the server (debug mode)

Simply execute the program like so. Uses the Flask debug server by default.


Make sure you execute in the same CWD as the ingest command, otherwise the pets.db won't be in the expected location.

Extracting the data into the database

This will likely not be required if I have included pets.db in the repo, however for brevity I have detailed my approach. I've written a series of scripts to actually get the information into a database. Simply execute them as follows:


This will output the pets database to pets.db. Make sure the CWD is the root of the repo.

The geocoding flag requires network access and vastly increases processing time, so disable if needed.

For more help, see:

python3 -h

Running tests

Each file can be tested simply by running their test_*.py counterparts.


See relevant files for more detailed comments. This project is made of several parts, detailed below:

  • data_*.py - Takes the raw CSV file provided by Google, converts it into something nicer. Several decisions on filtering/processing are made here. Will optionally populate the SQL database.
  • - Entry point for the Flask-based webapp. This defines the routes and frontend API surface.
  • /templates - Contains all Jinja2-based templates
  • /static - Contains static assets, most notably CSS and client-side scripts


  • Only one breed extracted per animal listing. This was to prevent a many-to-many relationship and a lot of complexity.
  • Full autocomplete keyword dictionary is stored in memory - may fail with a large number of keywords.
You can’t perform that action at this time.