A CPython extension for Vectorscan, an open source fork of Hyperscan, Intel's open source (prior to version 5.4), high-performance multiple regex matching library.
- ✅ Binary manylinux-compatible wheels
- ✅ Statically linked (no need to build Hyperscan/Vectorscan)
- ✅ Chimera support
# 🪄 Installing libhs is NOT required, because python-hyperscan is statically linked
pip install hyperscanIf you'd like to use Intel's Hyperscan rather than Vectorscan, or if you'd like to enable native CPU detection to build optimized non-FAT libraries (default off in Vectorscan), extending the manylinux-hyperscan Docker image used to build the binary wheels for this library should be fairly straightforward.
python-hyperscan currently exposes most of the C API, with the
following caveats or exceptions:
- No stream compression support.
- No custom allocator support.
hs_expression_info,hs_expression_ext_info,hs_populate_platform, andhs_serialized_database_infonot exposed yet.
See the documentation for more detailed build instructions.
![@github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=64&v=4){kind=small}
