pybloomfilter3 is a maintained fork of pybloomfiltermmap3 by @prashnts.
The goal of pybloomfilter3 is simple: to provide a fast, simple, scalable, correct library for Bloom filters in Python.
There are a couple reasons to use this module:
- It natively uses mmapped files.
- It is fast (see benchmarks).
- It natively does the set things you want a Bloom filter to do.
After you install, the interface to use is a cross between a file interface and an ste interface. As an example:
>>> import pybloomfilter
>>> fruit = pybloomfilter.BloomFilter(100000, 0.1, '/tmp/words.bloom')
>>> fruit.update(('apple', 'pear', 'orange', 'apple'))
>>> len(fruit)
3
>>> 'mike' in fruit
False
>>> 'apple' in fruit
TrueTo create an in-memory filter, simply omit the file location:
>>> fruit = pybloomfilter.BloomFilter(10000, 0.1)
>>> fruit.add('apple')
>>> 'apple' in fruit
TrueThese in-memory filters can be pickled and reloaded:
>>> import pickle
>>> pickled_fruit = pickle.dumps(fruit)
>>> unpickled_fruit = pickle.loads(pickled_fruit)
>>> 'apple' in unpickled_fruit
TrueCaveat: it is currently not possible to persist this filter later as an mmap file.
Current docs are available at pybloomfiltermmap3.rtfd.io.
To install:
pip install pybloomfilter3and you should be set.
pybloomfiltermmap is an excellent Bloom filter implementation for Python 2 by @axiak and contributors. I, @prashnts, made initial changes to add support for Python 3 sometime in 2016 as the current pybloomfiltermmap3 on PyPI. Since then, with the help of contributors, there have been incremental improvements and bug fixes while maintaining the API from versions 0.4.x and below.
@cavoq forked pybloomfiltermmap3 in 2025 to continue development and maintenance of the library to ensure it remains compatible with the latest Python versions (pybloomfilter3 on PyPI).
Some new features and changes were first introduced in version 0.5.0. From this point on, the goal is to reach stability, as well as add a few more APIs to expand upon the use cases. While we can't guarantee that we won't change the current interface, the transition from versions 0.4.x and below should be quick one liners. Please open an issue if we broke your build!
Suggestions, bug reports, and / or patches are welcome!
When contributing, you should set up an appropriate Python 3 environment and install the dependencies listed in requirements-dev.txt.
Package installation depends on a generated pybloomfilter.c file, which requires Cython module to be in your current environment.
# Creates a virtual env called "env"
python -m venv env
# Activates the created virtual env
source env/bin/activatepython -m pip install --upgrade pip
pip install cythonpython -m buildpython -m unittest discover -s tests -p "*.py"python -m pip install --upgrade twine
python -m twine upload dist/*.tar.gzSee the LICENSE file. It's under the MIT License.