This is an opensource project which is based on data that is owned and published by boardgamegeek.com. All data that is processed by this project is owned by boardgamegeek.com.
The purpose of this project is to download all ranked game data that boardgamegeek keeps, to convert the original XML data to json format and to present this data via a small Angular single page application.
Boardgamegeek XML API v.1 is used to download the original XML data.
You need to have the following software installed
- wget
- silversearcher
- Java JDK v 11+
- awk
- sed
- tar
- gzip
You need to run
./create-data.sh
This will take about 4-5 hours to complete to download all games' XML data from boardgamegeek. In order not to generate too much load on boardgamegeek, every XML API call is performed after 5 seconds of wait. This is recommended and required by boardgamegeek.
The script will first create some necessary directories and create archives of your old runs. Then it will first download the sitemap boardgame pages from boardgamegeek.com. It will then generate a list of game ids that correspond to those games. It will then download, using the boardgamegeek XML api v.1, the xml data of these games. Finally it will run a small software program contained in the converter directory to convert all the xml data to json. It will generate 2 json files; one maps 1-1 to the xml data and another which is custom format used by the frontend application.
games-list directory contains an Angular frontend application which presents all the games. Running the script will copy for you the json data to the assets of the Angular application. In order to run the frontend application perform
cd games-list
npm install
ng serve
You will have to wait a few moments until the single page application renders the material table because of the large number of data it renders. Optimizing the single page application is not within the scope of this project.