java-algorithms-playground

📚 Learning and exploring algorithms (simple ones!) in Java.

Overview

This codebase is for me to explore basic algorithms using Java. I started the beginnings of this codebase years ago (2015) without version control. Later, I formalized it as a Git repository, added "run instructions" and did other clean ups. Today (2023), I want to get even more leverage out of it as I continue to implement algorithms in Java as a way to learn language features like Java's Pattern Matching for switch and in general to enrich my intuition about computational power in the same spirit as Computers are Fast.

A tenet of a codebase like this is to reduce the noise. That's why the project does not use a build tool like Gradle or Maven and instead relies on a direct invocation of javac to compile the program. A project like this is for "early programmers" who are learning principles of computation and just enough programming language knowledge (e.g. Java, javac, and java; or honestly a non-compiled language would be best) to exercise their knowledge.

NOTE: This project was developed on macOS. It is for my own personal use.

Instructions

Follow these instructions to build and run the program.

1. Use Java 19
2. Compile the program:

   • ./build.sh

3. Run the program:

   • java -cp out/ --enable-preview dgroomes.Main

   • The program should output something like this:

     Question 1.1
     testing: charsAreAllUnique_v_1
     input string|expected result|actual result
     a|true|true
     ab|true|true
     ba|true|true
     aa|false|false
     |true|true
     testing: charsAreAllUnique_v_1
     input string|expected result|actual result
     a|true|true
     ab|true|true
     ba|true|true
     aa|false|false
     |true|true
     
     
     Question 1.3
     testing: isAPermutationOfB
     input string|expected result|actual result
     baa|true|true
     aba|true|true
     a|false|false
     testing: isAPermutationOfB
     input string|expected result|actual result
     baa|true|true
     aba|true|true
     a|false|false
     
     ... snip ...
     

Wish List

General clean-ups, TODOs and things I wish to implement for this project:

• DONE Organize everything. Delete the unused code. Finish the unfinished code. Use consistent package names between the main and source test sets.
  • DONE Fix the classpath scanning stuff. Guava recommends not using their own utility (com.google.common.reflect.ClassPath) and recommends using https://github.com/classgraph/classgraph.
• DONE Consider a "zero-weight project". Meaning, consider removing all dependencies, and Gradle and instead only having Java source in this project. The project would be built with a direct invocation of javac. Why? It's rare to have Java projects without a build tool like Gradle or Maven but in an academic context, we just want to focus on the task at hand: algorithms. As much as I want the convenience of Gradle and JUnit, the inclusion of those things multiplies the complexity of the codebase and ultimately is a risk for losing the reader. They may try to run the project and give up because of environmental issues or worse software rot (old versions of libraries and tools that don't work after some years). That would be a shame.
  • DONE Replace JUnit tests with public static void main tests
  • DONE Remove Gradle. In place of it, there will be a build script that looks like this https://github.com/dgroomes/jshell-playground/blob/d2a0674b78c17493b0e9c79bede05ee6d8c4fca6/basic/build.sh#L11
• Consistent test log output. Right now it's a mismash of "printing test cases" or printing nothing at all. I don't mean for this to be a "test framework" but instead just conventions for describe what is being tested, and the test result (success or failure).
• DONE Port over the simple search algorithms from my other playground repo: https://github.com/dgroomes/arrow-playground. Those algorithms should live here and that repo should be focused on the mechanics of the Arrow library.