Ladybug + Icebug Star Wars Notebook

This workspace contains a reproducible uv project and a notebook demo for the Ladybug nightly Arrow query path feeding Icebug graph analytics.

Notebook Index

Notebooks are listed from simpler graph-query examples to more advanced storage and analytics workflows.

1. Python Tutorial: Analyze a Social Network - introductory Ladybug graph querying from Python.
2. Ladybug: query a Star Wars graph stored in DuckDB - attach a DuckDB database and declare a relationship table backed by an attached DuckDB table.
3. Ladybug nightly + Icebug: Star Wars graph analytics - query Arrow-backed Star Wars data, persist it to native Ladybug storage, export CSR, and run Icebug analytics.
4. Ladybug nightly + Icebug: Star Wars graph analytics with Arrow CSR - build Arrow CSR tables in memory and share them between Ladybug and Icebug.
5. Ladybug nightly + Icebug: Karate Club graph analytics from icebug-disk - convert DuckDB tables to icebug-disk Parquet, load them natively in Ladybug, and run Icebug algorithms.
6. pgembed: embedded Postgres logical replication - start two embedded Postgres clusters with pgembed over Unix domain sockets, publish logical WAL changes from the primary, and read replicated rows from the secondary.

Open the notebook index directly with:

uv run jupyter notebook index.ipynb

Or open an individual notebook with:

uv run jupyter notebook <notebook>.ipynb

Or execute one from the CLI:

uv run jupyter nbconvert --to notebook --execute ladybug_icebug_starwars.ipynb --inplace

The pgembed replication notebook uses an optional dependency group:

uv sync --extra postgres