fast-collections

Blazing-fast data structures and algorithms you've been missing — written in Rust, usable from Python.

Why this exists

Python's standard library lacks many fundamental data structures and algorithms (which is not necessarily bad). Whenever you need a trie, a B‑tree, a suffix array, or anything beyond what's provided, you either copy a slow pure‑Python implementation from GitHub or write your own dangerous version.

fast-collections fills those gaps with Rust implementations. You get production-ready data structures and algorithms that are orders of magnitude faster than anything you could write in pure Python — with compact memory layout and full type annotations.

Why not write it yourself in Python?

Writing such data structures from scratch is hard. Even if you know the theory, it's easy to miss edge cases or introduce subtle bugs. And pure‑Python implementations are often slow anyway.

We've already done the work for you — fast‑collections provides well‑tested, Rust‑optimized versions of these structures. Just import and use them with confidence.

Who is this for?

• Anyone who has ever needed a data structure or algorithm that isn't in the standard library.
• Anyone who wants Rust performance without leaving Python.

Quickstart

pip install fast-collections

from fast_collections import Trie

t = Trie()
t.insert("hello")
t.insert("world")

assert "hello" in t
assert t.starts_with("he")

See the full documentation for all available data structures, algorithms, and their APIs.