From ef8104c69bd35bb419f2acc791582502d3459136 Mon Sep 17 00:00:00 2001 From: Steven Lee <20972179+alitrack@users.noreply.github.com> Date: Wed, 28 Oct 2026 16:59:58 +0800 Subject: [PATCH] Add ml, orc, and python_udf community extensions (#402) Adds three community extensions maintained under the alitrack org: - ml: columnar-native ML (34 algorithms, zero Python deps) - orc: Apache ORC reader in pure Rust - python_udf: embedded CPython UDFs via PyO3 + quack-rs All three are available via INSTALL FROM community. Co-authored-by: lihongyan --- README.md | 3 +++ 1 file changed, 3 insertions(+) diff --git a/README.md b/README.md index c614d15..698a164 100644 --- a/README.md +++ b/README.md @@ -555,12 +555,15 @@ Community-contributed DuckDB extensions, which can be installed via `INSTALL ⟨ - [`lindel`](https://query.farm/products/extensions/lindel/) - Linearization/Delinearization, Z-Order, Hilbert and Morton Curves. - [`marisa`](https://query.farm/products/extensions/marisa/) - Builds and queries compact MARISA tries for string lookup and prefix search. - [`minijinja`](https://query.farm/products/extensions/minijinja/) - Renders Jinja-compatible templates from SQL using MiniJinja. +- [`ml`](https://github.com/alitrack/duckdb-ml) - Lightweight, columnar-native ML for DuckDB — train + inference, 34 algorithms, zero Python deps. - [`netquack`](https://github.com/hatamiarash7/duckdb-netquack) - Parsing, extracting, and analyzing domains, URIs, and paths with ease. - [`observefs`](https://github.com/dentiny/duckdb-filesystem-observability) - I/O observability for DuckDB filesystems with latency statistics and external file cache access insights. - [`onager`](https://github.com/CogitatorTech/onager) - A DuckDB extension for graph data analytics. +- [`orc`](https://github.com/alitrack/duckdb_orc) - Reads Apache ORC files in pure Rust — all primitives, struct, list, map types and every compression codec (Zlib, Snappy, LZO, LZ4, ZSTD). - [`pfc`](https://github.com/ImpossibleForge/pfc-duckdb) - Read block-indexed PFC-compressed JSONL logs with timestamp filtering — 25% smaller than gzip with minimal S3 egress. - [`prql`](https://github.com/ywelsch/duckdb-prql) - Run PRQL commands directly within DuckDB. - [`pst`](https://github.com/intellekthq/duckdb-pst) - Read Microsoft PST files in-place with rich schemas for emails, contacts, appointments, tasks, and more. +- [`python_udf`](https://github.com/alitrack/duckdb-python) - Embedded CPython UDFs for DuckDB — py_eval, py_agg, py_scan, py_register, py_call, py_map. - [`query_condition_cache`](https://github.com/dentiny/duckdb-query-condition-cache) - Caches query conditions to improve performance for repeated-query workloads. - [`quickjs`](https://query.farm/products/extensions/quickjs/) - Runs JavaScript from SQL using the QuickJS-NG engine. - [`radio`](https://query.farm/products/extensions/radio/) - WebSocket and Redis Pub/Sub messaging from SQL.