Joshua Doucet

Software and Computer Enthusiast

Code Portfolio

Josh's GitHub

Wandering Threads Storefront

The Wandering Threads Nesting Feathers (WTNF) Storefront is an online web application that allows users to log in to their user account, browse products for sale, add them to a shopping cart, and place orders.

• Developed using TypeScript, HTML, PostgreSQL languages
• Backend built using PostgreSQL, Node.js, and Express
• Frontend build using Angular
• Application hosted on Amazon Web Services (EC2, S3, RDS, EB)
• Uses CircleCI for CI/CD

Products Page

Product Details Page

More on Wandering Threads Storefront Less on Wandering Threads Storefront

Sign In

Sign Up

Shopping Cart

Wandering Threads Storefront on GitHub
Visit the WTNF Storefront

Jimmy's Lefties

A web application for finding unpaired and mismatched shoes. Jimmy's Lefties scrapes shoe listings from multiple sites and centralizes them in one location. Inspired by amputees wanting to easily find a single shoe

• Built using Python and the Django framework.
• Uses the BeautifulSoup module for web scraping to acquire shoe listings on other sites.
• Uses the plotly module for bar chart data visualization
• Uses Django's built-in functionality for user authentication
• User search logging using a Django model

Home page

Search for shoes with phrase "New Balance"

More on Jimmys Lefties Less on Jimmys Lefties

Search for all shoes

Administrator page that shows all user shoes searches

Login page

Register page

Jimmys Lefties on GitHub
A walkthrough of Jimms Lefties on YouTube

Musiccharts MySQL Database

The music charts database project aimed to take a .CSV file of music chart data and convert it into a third normal form relational database using MySQL. The input data in the CSV file contained rows which identify a specific music track that ranked on a Billboard 100 chart for a given year in the range years 1964-1975. The columns represented details about a track such as title, artist, writers, year, record label, date the track entered the Top 100 charts, and the ranking(s) the track had on the top 100 charts. The problem with the input data was that it was not in third normal form which is desirable in a relational database.

• Built using MySQL Server 8
• An entity relationship diagram was built to organize the data
• The project focused on normalizing the input data so that it is better fit for a relational database

ERD for the musiccharts DB

More on Musiccharts DB Less on Musiccharts DB

Table descriptions

Table descriptions (cont.)

Using dynamic SQL to insert chart ranking data from 76 repeating columns (week1-week76) into the normalized DB

A query for finding song titles that appeared in the 1969 top 100 charts that were performed by at least 2 different artists. Show title, artist, and highest track rank for that year.

Pseudo PL20 Language Compiler

An application that compiles "PL20" source code into "quads" which are executed using a pseudo interpreter built using C++ code. The application takes a "PL20" source code file as input. Then, the source code is tokenized using lexical analysis that is based on a deterministic finite automata (DFA). After tokenization, the tokens are parsed using syntax analysis based on context-free grammars (CFGs) that ensure proper code syntax. Once the code is verified to be syntactically correct, "quad" codes are generated based on the sequence of tokens. The quads resemble assembly code instructions with three arguments. Lastly, the quads are fed into the pseudo interpreter which reaads the quads and executes the corresponding C++ operations.

• Built using C++ and Visual Studio 2019
• Focuses on the fundamentals of compiler design
• Utilizes deterministic finite automaton (DFA) and context-free-grammars (CFG) structures to "compile" PL20 source code into runnable C++ code.

A PL20 program for input

An ouput of the PL20 program

More on Pseudo PL20 Compiler Less on Pseudo PL20 Compiler

Another example program written in PL20

Tokenization of the above triangle area PL20 source program

Initial SymbolTable state, variables and constants found in the triangle area source

Quad codes generated for the triangle area source. OP1, OP2, and OP3 represents indicies in the SymbolTable

The execution trace of the above quad executions. Red arrows indicate user input values. Blue arrow represents the program output.

The final values in the SymbolTable after the quads have finished executing.

Pet Finder App

A web application for getting pets into new homes. Individuals and organizations are able to list pets that are available for adoption, and other individuals who are looking to adopt can search the web application for pets based on several kinds of pet search criteria. Furthermore, individual are able to take a "pet quiz" which helps them determine what kind of pet may be an ideal fit for them

• Built using the Ruby on Rails framework within a Docker container running a Cloud9 IDE image
• Built using Agile, TDD, and BDD methodologies in a team of 5 developers while using Git and GitHub for version control
• Uses rails models to store pet and user information in a SQLite3 database
• Uses the Google Maps API to display static maps of a listed pet's location

Home page

More on Pet Finder App Less on Pet Finder App

Taking a pet quiz	Pet quiz results	Adding a pet to pet finder
Pet search criteria page	Pet search results page	Pet details and contact page

DungeonBrawl

A Windows Forms turn-based fighting game where HEROES are put up against ENEMIES. You play as the heroes trying to defeat never-ending waves of enemies. Your objective is to defeat as many waves of enemies as possible before your three fighters have been downed and become unable to battle. Each time you defeat all three enemies, you will proceed to the next level. Upon entering a new level you will be presented with a new wave of enemies that have full health HP and full special points SP.

• Built using C# and Windows Forms within Visual Studio
• The application is event driven and updated the UI when the player performs game actions such as attacking and defending
• Each character type utilizes class inheritance to ensure all character types have the same base fields and actions. Each character class type is given different stats such as health points, special attack points, strength, and intellect.

A game in progress

How to play the game

More on DungeonBrawl Less on DungeonBrawl

Another game in progress

Fields for the base Character class

DungeonBrawl on GitHub

Minesweeper Clone

A Window Forms game that mimics the classic Minesweeper game. Find and mark all 10 mines to win the game.

• Built using C# and Windows Forms within Visual Studio
• Tracks lifetime game stats such as win loss ratio and average game time

An in progress game

Winning a game

More on Minesweeper Less on Minesweeper

Losing a game

How to play

Showing lifetime game stats

Game menu

Movie Journal

An Android app that utilizes TMDb to help you find and keep track of your favorite films. Search and view popular movies, read reviews, watch trailers, and make a list of your favorites!

• Built using Java and Android Studio
• Fetches nmovie data from The Movie Database API in an asynchronous manner
• Stores user favorite movies in a SQLite3 database
• Optimized for devices of different screen sizes

Searching for movies with "star" in the title

Showing user favorited movies

Showing the most popular TMDB movies

More on Movie Journal Less on Movie Journal

Showing the main app menu

Showing a movie details page

Trailers and reviews on movie details page

Showing most popular on a tablet

Showing movie details on a tablet/span>

Searching for "star" on a tablet in landscape orientation

Get Movie Journal on Google Play

Snap and Store

An Android app that allows users to create an inventory of objects. The database can be used to inventory personal belongings, items in storage, or a list of items that will be for sale. The name of an item, how many you have, the name of manufacturer, as well as the value/cost of the item, and an image of the object is all useful information that Snap and Store keeps track of for users.

• Built using Java and Android Studio
• Stores user items in a local SQLite3 database
• Allows users to store photos in the local DB
• Calculates the total value of all items in inventory

Show all items in storage

Adding a new item (1 of 2)

Adding a new item (2 of 2)

More on Snap and Store Less on Snap and Store

Deleting an item

Clean slate, deleting all items

Get Snap and Store on Google Play

Snap and Store on GitHub

Blackjack

A clone of the Blackjack card game where players compete against the dealer to see if they can get as close to the card value of 21 without going over and busting. A gamified betting system is imcluded in the game.

• Built using Python within the CodeSkulptor browser IDE
• Utilizes the simplegui module
• Displays card images by drawing a subset of pixels from a larger image that contains all the card images

Betting $325 on a game

Player losing the bet

More on Blackjack Less on Blackjack

Betting $177 on a game

Player winning the bet

Play Blackjack here