https://github.com/rfgplk/fuzzy
an approximate string matching library written in C++
https://github.com/rfgplk/fuzzy
c-plus-plus cplusplus cpp cpp-library cpp20 edit-distance edit-distance-algorithm hamming-distance header-only levenshtein-distance
Last synced: about 1 month ago
JSON representation
an approximate string matching library written in C++
- Host: GitHub
- URL: https://github.com/rfgplk/fuzzy
- Owner: rfgplk
- License: bsl-1.0
- Created: 2024-10-09T20:29:22.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-10-10T12:16:57.000Z (over 1 year ago)
- Last Synced: 2025-05-30T18:14:18.406Z (10 months ago)
- Topics: c-plus-plus, cplusplus, cpp, cpp-library, cpp20, edit-distance, edit-distance-algorithm, hamming-distance, header-only, levenshtein-distance
- Language: C++
- Homepage:
- Size: 47.9 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Fuzzy
### a C++ Hamming/Levenshtein distance approximate string matching library
... so *fast* you'd think he was a **hash**.
Fuzzy is an extremely fast **approximate string matching** (fuzzy finding) library written in C++20.
It offers two primary modes of operation, one via the Hamming metric and the other via the Levenshtein distance (by means of the W-F algorithm). The Hamming metric is extremely efficient, being able to parse and match strings at speeds of ~3-4 GB/s per thread based on current testing (Haswell/Broadwell/Skylake). The second, more precise mode of operation, is via the Levenshtein metric, which is slower but offers more accurate hits. Requires AVX2.
## Usage
```cpp
similar(string, string); // gauge approximate similarity of two strings
metric_hamming(token, string); // retrieve a raw Hamming metric
metric_lsh(token, string); // retrieve a raw Levenshtein metric
rough_find(token, strings); // find an approximate string match by using the Hamming distance
precise_find(token, strings); // find a string match more precisely by using the Levenshetin distance
fuzzy_find(token, strings); // search for occurance of token in strings
fuzzy_exists(token, strings); // check if an exact match of a string exists
```
### Example
```cpp
#include "src/fuzzy.hpp"
std::string to_find = "Hello";
std::vector words = { "apples", "ello", "chameleon", "asphalt", "behemoth" };
auto f = fuzzy::fuzzy_find(to_find, words, 3); // get top three results
std::string best_match = f(); // best match string by name
long double likelihood = f.likelihood(); // get likelihood of string match
std::string likelihood = f[n]; // get name of string at index, ordered from highest likelihood hit
```
## Installation
Fuzzy is a header only library. Just copy all files from `src/` and include `src/fuzzy.hpp` into your project.
## TODO
- [ ] add fuzzy_blob/fuzzy_bin
- [ ] add full SSE support
- [ ] expand AVX support via AVX-128 \& AVX-512
- [ ] more comprehensive unit test suite
- [ ] add functionality for string streaming
- [ ] C bindings
## Misc
no dependencies, requires C++20, tested with g++.
## License
Licensed under the Boost Software License.