Polyleven

0.11.0 · active · verified Sat Apr 11

Polyleven is a hyper-fast Python library for computing Levenshtein distance, implemented in C for optimal performance. It is designed to be efficient for comparing both short and long string inputs, is stand-alone with no external Python dependencies, and is distributed under the MIT License. The current version is 0.11.0, released on February 9, 2026, indicating an active release cadence.

Warnings

gotcha For optimal performance, especially with long strings or when only interested in small distances, provide an integer `k` as the third argument to the `levenshtein()` function. This `max_threshold` limits the maximum distance computed, making the operation significantly more efficient. If the true distance exceeds `k`, the function returns `k + 1`.
Fix: Call `levenshtein('string1', 'string2', k=max_threshold)` where `max_threshold` is the maximum relevant distance. For example, `levenshtein('abcde', 'xxxxx', 2)` would return `3` as the actual distance (5) is greater than the threshold (2).
gotcha The license for Polyleven changed from Public Domain to MIT License starting from version 0.7. Users of older versions relying on Public Domain terms should be aware of this change.
Fix: Verify the license (MIT for v0.7+) to ensure compatibility with your project's licensing requirements. If strict adherence to Public Domain is necessary, ensure usage of versions prior to 0.7.
gotcha While the `max_threshold` argument generally improves performance, the speed boost is most noticeable when the threshold `k` is small, typically less than 3. Larger thresholds still offer some benefit but diminish as `k` approaches the full string length.
Fix: When designing for maximum performance and a small distance is expected or acceptable, prioritize using a `max_threshold` value of 1 or 2.

Install

pip install polyleven Install with pip

Imports

levenshtein
```
from polyleven import levenshtein
```

Quickstart

This quickstart demonstrates how to import the `levenshtein` function and use it to calculate the Levenshtein distance between two strings. It also shows how to leverage the optional `max_threshold` argument to improve performance, especially when only interested in distances below a certain value.

from polyleven import levenshtein

# Calculate Levenshtein distance between two strings
distance1 = levenshtein('kitten', 'sitting')
print(f"Distance between 'kitten' and 'sitting': {distance1}")

# Calculate Levenshtein distance with a maximum threshold for efficiency
# If the actual distance exceeds the threshold, the threshold + 1 is returned.
distance2 = levenshtein('apple', 'aple', 1) # Actual distance is 1
print(f"Distance between 'apple' and 'aple' with max_threshold=1: {distance2}")

distance3 = levenshtein('banana', 'orange', 2) # Actual distance is higher than 2
print(f"Distance between 'banana' and 'orange' with max_threshold=2: {distance3}")

assert distance1 == 3
assert distance2 == 1
assert distance3 == 3 # Returns threshold + 1 because actual distance > threshold

view raw JSON →