cityhash 0.3.2

Python bindings for CityHash and FarmHash

A Python wrapper around FarmHash and CityHash, two fast non-cryptographic hashing functions.

Getting Started

To use this package in your program, simply type

pip install cityhash

This package exposes Python APIs for CityHash and FarmHash under cityhash and farmhash namespaces, respectively. Each provides 32-, 64- and 128-bit implementations.

Usage Examples

Stateless hashing

Usage example for farmhash:

>>> from farmhash import FarmHash32, FarmHash64, FarmHash128
>>> FarmHash32("abc")
1961358185
>>> FarmHash64("abc")
2640714258260161385
>>> FarmHash128("abc")
76434233956484675513733017140465933893

Hardware-independent fingerprints

Fingerprints are seedless hashes which are guaranteed to be hardware- and platform-independent. This can be useful for networking applications which require persisting hashed values.

>>> from farmhash import Fingerprint128
>>> Fingerprint128("abc")
76434233956484675513733017140465933893

Incremental hashing

CityHash and FarmHash do not support incremental hashing and thus are not ideal for hashing of streams. If you require incremental hashing feature, use MetroHash or xxHash instead, which do support it.

Fast hashing of NumPy arrays

The Python Buffer Protocol allows Python objects to expose their data as raw byte arrays to other objects, for fast access without copying to a separate location in memory. Among others, NumPy is a major framework that supports this protocol.

All hashing functions in this packege will read byte arrays from objects that expose them via the buffer protocol. Here is an example showing hashing of a 4D NumPy array:

>>> import numpy as np
>>> from farmhash import FarmHash64
>>> arr = np.zeros((256, 256, 4))
>>> FarmHash64(arr)
1550282412043536862

The arrays need to be contiguous for this to work. To convert a non-contiguous array, use NumPy’s ascontiguousarray() function.

SSE4.2 support

On CPUs that support SSE4.2 instruction set, FarmHash-64 has an advantage over its non-optimized version and over vanilla CityHash-64, as can be seen below. The numbers below were recoreded on a 2.4 GHz Intel Xeon CPU (E5-2620), and the task was to hash a 512x512x3 NumPy array.

Method	Time (64-bit)	Time (128-bit)
FarmHash / SSE4.2	373 µs ± 48.3 µs	480 µs ± 15.3 µs
FarmHash	464 µs ± 19.2 µs	490 µs ± 23.0 µs
CityHashCrc / SSE4.2	N/A	377 µs ± 21.7 µs
CityHash	492 µs ± 16.7 µs	487 µs ± 22.0 µs

The SSE4 support in CityHash is available under cityhashcrc module. To use SSE4.2-optimized CityHash in a platform-independent way, you can use the following:

try:
    from cityhashcrc import CityHashCrc128 as CityHash128
except Exception:
    from cityhash import CityHash128

Development

Local workflow

For those who want to contribute, here is a quick start using some makefile commands:

git clone https://github.com/escherba/python-cityhash.git
cd python-cityhash
make env           # create a Python virtualenv
make test          # run Python tests
make cpp-test      # run C++ tests
make shell         # enter IPython shell

The Makefiles provided have self-documenting targets. To find out which targets are available, type:

make help

Distribution

The wheels are built using cibuildwheel and are distributed to PyPI using GitHub actions using this workflow. The wheels contain compiled binaries and are available for the following platforms: windows-amd64, ubuntu-x86, linux-x86_64, linux-aarch64, and macosx-x86_64.

See Also

For other fast non-cryptographic hash functions available as Python extensions, see MetroHash, MurmurHash, and xxHash.

Authors

The original Python bindings were written by Alexander [Amper] Marshalov, then were largely rewritten for more flexibility by Eugene Scherba. The CityHash and FarmHash algorithms and their C++ implementation are by Google.

License

This software is licensed under the MIT License. See the included LICENSE file for details.