Up to 10x faster strings for C, C++, Python, Rust, and Swift, leveraging NEON, AVX2, AVX-512, and SWAR to accelerate search, sort, edit distances, alignment scores, etc 🦖
Currently, the anti-diagonal approach for Levenshtein distance is only implemented for square matrices.
It should be easy to generalize it to rectangular ones, opening opportunities for more aggressive vectorization.
Currently, the anti-diagonal approach for Levenshtein distance is only implemented for square matrices. It should be easy to generalize it to rectangular ones, opening opportunities for more aggressive vectorization.
https://github.com/ashvardanian/StringZilla/blob/06423189e7a5b9e2882b471e98ec897d40734fc4/include/stringzilla/stringzilla.h#L2010-L2013