- Gathers the wiki pages from the database dump
- Extracts relevant pages
- Inspects sections for links (currently manual, see old commit for other)
- using marko: markdown parsing
- wikimarkup: converts to HTML, uses beautifulsoup to extract texts
- manual: rough (but fast and robust) text processing
- Postprocesses links
- Caches information
- Generates interactive graph using pyvis network