alexis779 / adventofcode

Solutions to advent of code challenge in Rust

How to learn a new programming language

Let's solve problems in Advent of code 2023 challenge to learn Rust.

Each day of the Advent period includes a pair of challenges to solve.

Create a dedicated commit for each day. Pushing code regularly serves a a SMART (Specific, Measurable, Achievable, Relevant, Time-bound) goal that helps measuring your own progress.

Having a regular pace will indicate you're making consistent progress during the learning process.

Environment setup

Runtime

Install Rust programming language.

IDE

Install Visual Studio Code. It will serve as an IDE to edit, compile and run the code. In the extension marketplace in Visual Code, install Rust & AWS toolkit extensions.

Git

Install git. Create a new repository in github, like AdventOfCode.

Dev setup

Create a new workspace

$ cargo new advent_of_code_2023

Dev process

Product Requirements

Read the problem statement, say day1. This initial step is equivalent to gathering the product requirements.

Problem Solving

Work out the outline of a possible solution. Think of the data structures and algorithms you need to solve the problem.

Unit test

• create a new input file for the test sample in inputs/day1.txt
• add 2 new unit test functions in src/lib.rs

Writing a unit test establishes the contract the module is responsible for. Create a file containing the sample input. Then call a public function that parses this file using the file name as a String parameter. Carefully choose the function name to be self-explanatory. Add a test assertion on the actual value the function should return based the content of the input file.

Implement a solution

• create a new module in src/day1.rs
• add 2 new public functions in the new module src/day1.rs

Implement the algorithm you came up with in the problem Solving part, using adequate data structures and business models, until the tests pass.

Add all the implementation in a dedicated module under test coverage.

It should include a set of functions providing the implementation. They each take a set of inputs and return an output, providing a top-down execution of the code organized as a Russian doll game.

It should also contain the definitions of the models using a set of struct declarations.

Submit the solution

• create a new input file for the application sample in inputs/day1_input.txt
• update the main application in src/main.rs to call the public function. It will output the result you submit to solve the 2 parts of the challenge. This will grant you 2 gold starts.

Push the new commit

Make sure the tests pass

$ cargo test

Then push the code

$ git push

Rust books

When blocked, read a book documenting the key features of the programming language.

• https://github.com/Dhghomon/easy_rust?tab=readme-ov-file#introduction

Start a conversation with Chat bots

• Mistral

Great for low level syntax review. Confront the generated version with your own implementation

• Claude

Great for high level suggestions on coding best practices. It may also be able to spot your bug ?

• Amazon Q

Click the hexagon icon in the left panel in Visual Code to open it. Its answers provide external links to relevant blog articles for general questions.

Install a code assistant

Amazon CodeWhisperer is available as part of AWS Toolkit extension after signing up for an AWS builder id.

1. Code assistant will generate a one-liner or even a block of code based on what you're starting to type.

2. Write a brief description of what you intend to do as a code comment. It prompts the model prior to generating an implementation of the spec provided in the comment.

This will be hit or miss, but serves as a placeholder when stuck when trying to come up with valid syntax.

Various options are available to handle auto-completed code,

• To accept it, press Tab
• To discard it, press Escape
• To generate an alternative solution, presse left arrow