marhta

Fast string similarity algorithms

This library is still in development and may not be stable.

Installation

pip install marhta

Usage

from marhta import levenshtein_similarity, jaro_winkler_similarity

# Calculate string similarities
print(levenshtein_similarity("hello", "helo"))  # 0.8
print(jaro_winkler_similarity("martha", "marhta"))  # 0.961

# Find best matches
from marhta import levenshtein_match
strings = ["apple", "banana", "orange", "pear"]
matches = levenshtein_match("aple", strings)
print(matches)  # [("apple", 0.8), ("pear", 0.5)]

Features

• Levenshtein distance and similarity measures
• Jaro-Winkler distance and similarity measures
• Fuzzy string matching with customizable thresholds

Performance

Written in Rust for improved performance, while maintaining a Pythonic API.