Steam Sales Analysis

Overview

Welcome to Steam Sales Analysis – an innovative project designed to harness the power of data for insights into the gaming world. We have meticulously crafted an ETL (Extract, Transform, Load) pipeline that covers every essential step: data retrieval, processing, validation, and ingestion. By leveraging the robust Steamspy and Steam APIs, we collect comprehensive game-related metadata, details, and sales figures.

But we don’t stop there. The culmination of this data journey sees the information elegantly loaded into a MySQL database hosted on Aiven Cloud. From this solid foundation, we take it a step further: the data is analyzed and visualized through dynamic and interactive Tableau dashboards. This transforms raw numbers into actionable insights, offering a clear window into gaming trends and sales performance. Join us as we dive deep into the data and bring the world of gaming to life!

steamstore CLI

Setup

Installing the package

For general use, setting up the environment and dependencies is straightforward:

# Install the python distribution from PyPI
pip install steamstore-etl

Setting up the environment variables

• Create an .env file in a directory.

# Database configuration
MYSQL_USERNAME=<your_mysql_username>
MYSQL_PASSWORD=<your_mysql_password>
MYSQL_HOST=<your_mysql_host>
MYSQL_PORT=<your_mysql_port>
MYSQL_DB_NAME=<your_mysql_db_name>

• Open a terminal at the specified location

  For Ubuntu (or other Unix-like systems)

  1. Load .env Variables into the Terminal

     To load the variables from the .env file into your current terminal session, you can use the export command along with the dotenv command if you have the dotenv utility installed.

     Using export directly (manual method):

     export $(grep -v '^#' .env | xargs)
     

     • grep -v '^#' .env removes any comments from the file.
     • xargs converts the output into environment variable export commands.

     Using dotenv (requires installation):

     If you prefer a tool, you can use dotenv:

     • Install dotenv if you don't have it:

     sudo apt-get install python3-dotenv
     

     • Then, use the following command to load the .env file:

     dotenv
     

     Using source (not typical for .env but useful for .sh files):

     If your .env file is simple, you can use source directly (this method assumes no special parsing is needed):

     source .env
     

     Note that source works well if your .env file only contains simple KEY=VALUE pairs.

  2. Verify the Variables

     After loading, you can check that the environment variables are set:

     echo $MYSQL_USERNAME
     

  For Windows

  1. Load .env Variables into PowerShell

     You can use a PowerShell script to load the variables from the .env file.

     Create a PowerShell script (e.g., load_env.ps1):

     Get-Content .env | ForEach-Object {
        if ($_ -match "^(.*?)=(.*)$") {
           [System.Environment]::SetEnvironmentVariable($matches[1], $matches[2], [System.EnvironmentVariableTarget]::Process)
        }
     }
     

     • This script reads each line from the .env file and sets it as an environment variable for the current PowerShell session.

     Run the script:

     .\load_env.ps1
     

     Verify the Variables:

     echo $env:MYSQL_USERNAME
     

  2. Load .env Variables into Command Prompt

     The Command Prompt does not have built-in support for .env files. You can use a batch script to achieve this.

     Create a batch script (e.g., load_env.bat):

     @echo off
     for /f "tokens=1,2 delims==" %%A in (.env) do set %%A=%%B
     

     Run the batch script:

     load_env.bat
     

     Verify the Variables:

     echo %MYSQL_USERNAME%
     

CLI for Steam Store Data Ingestion ETL Pipeline

Usage:

$ steamstore [OPTIONS] COMMAND [ARGS]...

Options:

• --install-completion: Install completion for the current shell.
• --show-completion: Show completion for the current shell, to copy it or customize the installation.
• --help: Show this message and exit.

Commands:

• clean_steam_data: Clean the Steam Data and ingest into the Custom Database
• fetch_steamspy_data: Fetch from SteamSpy Database and ingest data into Custom Database
• fetch_steamspy_metadata: Fetch metadata from SteamSpy Database and ingest metadata into Custom Database
• fetch_steamstore_data: Fetch from Steam Store Database and ingest data into Custom Database

Detailed Command Usage

steamstore clean_steam_data

Clean the Steam Data and ingest into the Custom Database

Usage:

$ steamstore clean_steam_data [OPTIONS]

Options:

• --batch-size INTEGER: Number of records to process in each batch. [default: 1000]
• --help: Show this message and exit.

steamstore fetch_steamspy_data

Fetch from SteamSpy Database and ingest data into Custom Database

Usage:

$ steamstore fetch_steamspy_data [OPTIONS]

Options:

• --batch-size INTEGER: Number of records to process in each batch. [default: 1000]
• --help: Show this message and exit.

steamstore fetch_steamspy_metadata

Fetch metadata from SteamSpy Database and ingest metadata into Custom Database

Usage:

$ steamstore fetch_steamspy_metadata [OPTIONS]

Options:

• --max-pages INTEGER: Number of pages to fetch from. [default: 100]
• --help: Show this message and exit.

steamstore fetch_steamstore_data

Fetch from Steam Store Database and ingest data into Custom Database

Usage:

$ steamstore fetch_steamstore_data [OPTIONS]

Options:

• --batch-size INTEGER: Number of app IDs to process in each batch. [default: 5]
• --bulk-factor INTEGER: Factor to determine when to perform a bulk insert (batch_size * bulk_factor). [default: 10]
• --reverse / --no-reverse: Process app IDs in reverse order. [default: no-reverse]
• --help: Show this message and exit.

Setup Instructions

Development Setup

For development purposes, you might need to have additional dependencies and tools:

1. Clone the repository:

   git clone https://github.com/DataForgeOpenAIHub/Steam-Sales-Analysis.git
   cd steam-sales-analysis
   

2. Create a virtual environment:

   • Using venv:

     python -m venv game
     source game/bin/activate  # On Windows use `game\Scripts\activate`
     

   • Using conda:

     conda env create -f environment.yml
     conda activate game
     

3. Install dependencies:

   • Install general dependencies:

     pip install -r requirements.txt
     

   • Install development dependencies:

     pip install -r dev-requirements.txt
     

4. Configuration:

   • Create an .env file in the root directory of the repository.
   • Add the following variables to the .env file:

     # Database configuration
     MYSQL_USERNAME=<your_mysql_username>
     MYSQL_PASSWORD=<your_mysql_password>
     MYSQL_HOST=<your_mysql_host>
     MYSQL_PORT=<your_mysql_port>
     MYSQL_DB_NAME=<your_mysql_db_name>
     

Database Integration

The project connects to a MySQL database hosted on Aiven Cloud using the credentials provided in the .env file. Ensure that the database is properly set up and accessible with the provided credentials.

Running Individual Parts of the ETL Pipeline

To execute the ETL pipeline, use the following commands:

1. To collect metadata:

   steamstore fetch_steamspy_metadata
   

2. To collect SteamSpy data:

   steamstore fetch_steamspy_data --batch-size 1000
   

3. To collect Steam data:

   steamstore fetch_steamstore_data --batch-size 5 --bulk-factor 10
   

4. To clean Steam data:

   steamstore clean_steam_data --batch-size 1000
   

This will start the process of retrieving data from the Steamspy and Steam APIs, processing and validating it, and then loading it into the MySQL database.

Dashboard

• Explore the interactive Tableau dashboard.

Authors

1. Kayvan Shah | MS in Applied Data Science | USC
2. Sudarshana S Rao | MS in Electrical Engineering (Machine Learning & Data Science) | USC
3. Rohit Veeradhi | MS in Electrical Engineering (Machine Learning & Data Science) | USC

References:

API Used:

• Steamspy API
• Steam Store API - InternalSteamWebAPI
• Steam Web API Documentation
• RJackson/StorefrontAPI Documentation
• Steamworks Web API Reference

Repository

• Nik Davis's Steam Data Science Project

LICENSE

This repository is licensed under the MIT License. See the LICENSE file for details.

Disclaimer