danorel / NQueens

Web matrix-based n-queens quiz solver

NQueens

Project meant for Prolog subject as coursework

Hi! Welcome to my repo. Let me show you a short instruction how to launch the application.

The whole application is written in such stack:

• Backend: Python + Prolog insertions.
• Front-end: React.

Development configurations:

1. You need to install all dependencies in three core modules:

• /client: npm install --save
• /server: pip3 install -r requirements.txt
• /(root): npm install --save

2. Then return back to the root of the application and launch this command: npm run development:local. In such way you will launch simultaneously both of back-end and front-end code.

Production configurations:

1. In the root of the application run the command: npm run prod:local. This will completele install all required dependencies in all service and start the app in production mode.

What is the code structure?

Back-end structure

1. /server/src/pl/*.pl - Prolog files with back-end logic. NQueens combinations and moves are generated here.
2. /server/src/api/*.py - Application routes with back-end endpoints.
3. /server/src/entities/*.py - Flask/Prolog library connnectivity.
4. /server/src/modules/*.py - Modules for applications.
5. /server/src/modules/queens/api.py - Main components with all API routes for client-server connectivity.

Front-end structure

1. /client/public/*.(png|jpg|mp4) - Public images, videos and other stuff, used in the application.
2. /client/src/modules/*.(ts|tsx) - Game components with view based on Domain-Driven Design.
3. /client/src/routes/*.tsx - Routes meant for web-game navigation.

How to connect the application with Prolog?

For such purpose I used the PySwip Python library, which helped me to connect the Python back-end with Prolog logic.

Main components:

• prolog = pyswip.Prolog() - Basic class for Prolog-based operations.
• prolog.consult(*.some_path_to_prolog_file) - Helped to read the prolog files with logic and to get the results.
• prolog.query("n_queen(4,Y)") - Query to prolog to get the results of the predicate. Y will contain the results.

In such way I settled powerful Python-Prolog connectivity.

What I haven't done correctly?

The program calculating NQueens with backtracking algorithm. It extracts all possible moves before runtime interpretation of the program. Is such case I cannot use the API in Runtime and inline in the POST/GET requests NQueen calculations for different sizes of desk.

What could be improved?

• Find the way how to calculate the algorithm in Runtime.
• Find the way how to get only current move of the queen, not all possible combinations of queen locations on the desk.

Good luck!