This repo demonstrates using xllify and the xllify-build action to package some simple Luau functions into an .xll. The workflows are defined in build.yaml.
You can download pre-built XLLs from the latest release.
- xllify_demo.xll - lua functions see DOCS.md; see test sheet for Black Scholes
- xllify_demo_duckdb.xll - see duckdb/
- xllify_demo_python.xll - see black_scholes.py; see test sheet for Black Scholes
- xllify_perftest_luau.xll - see perftest/luau/
- xllify_perftest_python.xll - see perftest/python/
For more information about the action, see https://github.com/xllifycom/xllify-build or the workflow in this repo.
The duckdb directory contains a comprehensive example of integrating DuckDB's analytical capabilities with Excel. This demo showcases:
- In-memory database with 1.25M rows of synthetic stock market data
- Advanced SQL window functions (moving averages, rankings, percentiles)
- Dynamic array results that spill into Excel cells
- Pre-built analytical queries as Excel functions
See the DuckDB demo README for detailed documentation. It is built as xllify_demo_duckdb.xll and is attached to releases.
Simple performance test to exercise calling a function 5000 times. The single function implementation test_func is just a simple random number generator. Luau and Python versions have been implemented.
The sheet exercise.xlsx contains 5000 rows. Change the seed cell to recalculate. This workbook will work for both the luau and Python variants, but not at the same time of course.
- Load exercise.xlsx
- Load xllify_demo_luau.xll or xllify_demo_python.xll
- Observe recalc
- Change seed value
- Observe recalc
Luau uses multithreaded recalculation. Multi-process support for Python is coming soon. (It's actually in the codebase already just not exposed.)
After downloading and installing from the releases page, Excel should show the shiny new functions.
Note: you may have to open the .xll file Properties and check the Unblock checkbox.
