Simple frontend for some site dump I found online ( ͡° ͜ʖ ͡°)
- In-memory index of authors and caterories
- Parallel indexing (bottlenecked by disk speed anyway)
- Plain HTML web interface using html/template
- Zero dependencies (except that ~20 loc mmap package)
# Download database. This could take a while
curl https://the-eye.eu/public/AI/pile_preliminary_components/Literotica.jsonl.zst | zstdcat > database.jsonl
# Run directly
make run
# Or use Docker
docker build -t literotigo .
docker run -v $(pwd)/database.jsonl:/var/database.jsonl -p "8080:8080" literotigo
Releases under WTFPL