AtomToast / eve

Implementing evolution with self-learning autonomous agents roaming a virtual environment.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

EVE Build Status

This project aims to simulate basic lifeforms evolving and getting smarter, in a simplified manner.

Development

Setup

  1. Install cmake & Catch2 on your OS
  2. Clone this repository:
    HTTPS: git clone https://github.com/Xerus2000/eve
    SSH: git clone git@github.com:Xerus2000/eve.git
  3. Build 'eve' via CLion so that all tasks become available for running
  4. In CLion go to Settings > Editor > File and Code Templates and remove the "Created by" header in c/cpp files
  5. Execute this to set up the git commit message hooks:
    git config core.hooksPath .dev/githooks

Structure

All production code goes into src, separated into model, view and controller.
test contains the tests and mirrors the source file structure.

Classes and Inheritance

The folder structure should represent classes and inheritance.

Therefore,

  • every class belongs in its own folder with the same name
  • every sub-class belongs within its parents folder

Thus, as of currently we have the following directory tree:

object
├── entity
│   ├── food
│   └── individual
│       ├── species1
│       ├── species2
│       └── ...
└── obstacle

Representing the following Inheritance Diagram:

        Object
          │
   +------+------+
Obstacle       Entity
                 │
          +------+------+
          Food      Individual
                        │
                 +------+------+
                Sp1    Sp2    ...

Testing

We use Catch2 for testing.

The tests should be in BDD style, except for simple functions & data structures where it is usually more appropriate to use test cases and sections.

Git Guidelines

Commit messages use the type-scope format as seen in Karma Runner.

Branch names should roughly follow this convention as well, so a branch adding an individual called Fred should be called something like feat/fred while a branch fixing a bug in the list of individuals in the world could be named fix/world-individuals.

More detailed Collaboration Guidelines are being worked on in Notion.

About

Implementing evolution with self-learning autonomous agents roaming a virtual environment.


Languages

Language:C++ 95.6%Language:C 2.3%Language:CMake 1.1%Language:Shell 0.7%Language:Objective-C 0.3%