Run the code below in your terminal to install the necessary modules to compile the web app.
pip install -r requirements.txt
- Make sure that your Postgres server is configured as follows and turned on:
| Host | Port |
|---|---|
| localhost | 5432 |
Run
cp .env.example .env
Insert your username and password to postgres in .env
Then you can initialize the database with (if python3 does not work replace it with python):
python3 init.py
Then you can compile the web app and start the database by running the command:
python3 run.py
The web app is built up of several underlying sites, each with different implementations and use cases within the web app.
The home site serves as the navigational hub for the other sites. You can access the subsites from the menu or directly from the URL by entering the name of the page. The URLs are case-sensitive, so remember to use lowercase.
For example, to navigate to the search page, enter:
http://127.0.0.1:5000/search
To access the home page, the endpoint path is just /.
Here, it is possible to search the database for games using the name of a player and a sequence of moves in the given player's game. Both search requirements use regex matching (for player names in PostgreSQL; for sequence of moves (such as "e2e4, e7e5, Ng1f3"), with the Python module re). This means that it is possible to search through the database for all players if no player name is supplied since the empty string regex matches all player names in this context.
It is also possible to search for similar games to a given game ID using the second search button. This uses an LCS-solver module in Python to find how similar games are. A single game is returned (if the database has more than one game in the collection). The ID's are simple numbers (1, 2, ..., 20, ...) assigned by the Python application.
Note that the database fetches random information from pgnmentor when initializing the database. As such, a previously found game might not exist after re-initializing the database.
After you have searched for a gameID in the database, you can compare the two games that are most similar. The games are compared via their opening, and you can see their stats such as where, when, and who played the match. Additionally, you can see how the two games unfolded with their moves. There is currently no way to look up a specific game's gameID before searching for similar games.
Here, you can upload your own chess games to the DIScover Chess database. The app supports either Portable Game Notation files (.pgn) or a link to a file from the website PGN Mentor.
If you have a chess.com account, you can download a PGN file from the website and upload your own games to DIScover Chess!
This site contains a small section with information about this project.