TLDR with bald head
I want to create a library (I think that's the right term) for my own learning environments for reinforcement (envs for short).. Most Envs are based on self-implemented games that were created either in pure Python or in C ++ with Python bindings. What would be the best way to structure this project so that it is easy to expand, maintain and make the most sense? I want to be able to reuse code, e.g. B. using a general board class for all my board game implementations (e.g. chess, lot, Gomoku). I plan to make it cross-platform with the help of CMake and could even dare to pack it as a Conda package.
On my first search, I found that this layout (and its variations) is popular, and decided to rely on it.
My original plan was To structure my library into projects, create a repo for each project and insert one project as a Git submodule into another. To create the environment for the game
In 2048 the structure would be as follows (CMakeLists omitted):
(The env vector is based on the env-2048, which is based on the game-2048 that uses the general board class.)
general-board
├── external
│ └── Catch2/
├── include
│ └── general-board
│ └── file.h
├── src
│ └── file.cpp
└── tests
└── tests.cpp
game-2048
├── app
│ └── manual_game.cpp
├── external
│ ├── Catch2
│ └── general-board
├── include
│ └── game-2048
│ └── file.h
├── src
│ └── file.cpp
└── tests
└── tests.cpp
env-2048
├── external
│ ├── Catch2/
│ └── game-2048/
├── include
│ └── env-2048
│ └── file.h
├── src
│ └── file.cpp
└── tests
└── tests.cpp
env-vector <---- this would be on the top, bundling the envs together
├── external
│ ├── Catch2/
│ ├── env-2048/
│ ├── env-chess/ <---- another board game
│ └── env-go/ <---- another board game
├── include
│ └── env-vector
│ └── file.h
├── python
│ └── pybind11_magic_here
├── src
│ └── file.cpp
└── tests
└── tests.cpp
After some implementation, I was worried if the number of sub-modules and redundancy got too large. With this structure, the project above would include the General Board project n times (where n is the number of games based on one board), and the catch2 would be even more included. Seems fishy and prone to errors.
My second idea was to create a large project and to include everything in the project in a "flat" way and not in a "nested" way as before. It would look like this:
(line ending with '/' depicts a folder)
environments_all_in_one
│
├── external
│ └── Catch2/
├── include
│ └── environments_all_in_one
│ └── **not_even_sure_what_to_put_here**
├── python
│ └── pybind11_magic_here
├── src
│ ├── env_vector
│ ├── envs
│ │ ├── env-2048/
│ │ ├── env-chess/
│ │ └── env-go/
│ ├── games
│ │ ├── game-2048/
│ │ ├── game-chess/
│ │ └── game-go/
│ └── general-board
│ ├── board_abc/
│ ├── board_array/
│ └── board_vector/
└── tests
└── tests.cpp
In this way, code would not be present more than once and definitely contributes to transparency. However, since I have no experience, I have to ask:
Is there a better way to do this?