This repository contains the code behind the list at corpus-analysis.com.
The site is written in old-school PHP and definitely not an example of great software engineering. That being said, you're more than welcome to optimize the code!
The architecture is a bit odd: The actual data is stored in a Google Sheets sheet. This sheet, which effectively works as a multiuser backend, is then being downloaded as a CSV file and processed.
If you want to run your own version, you simply need to change and rename config.default.php to config.php. Of course, you will have to create your own Google Sheets and Google Forms if you want to replicate the service. Also, make sure to remove the analytics code and change the copyright/impressum.
If you want to run this on NGINX (i.e., the .htaccess won't work), remember to add the rewrite rules such as rewrite ^/tag/([^/]*)$ /index.php?tag=$1 last; to the configuration.
If you want to develop using Docker, there is a docker-compose.yml which will set up a working environment.
First, add 127.0.0.1 corpus-analysis.local to your hosts file. Now run docker-compose up --build --force-recreate. You can now browse the website on http://corpus-analysis.local/.
For an even quicker and simpler development setup, you can use, for example, Matt Rayner's DockerLAMP: docker run -p "80:80" -v ${PWD}/app:/app mattrayner/lamp:latest-1804.
GDPR-compliant cookie consent is implemented using the fantastic Coockie Consent by Orest Bida.