py_regex

A small Rust library providing a convenient wrapper around the Python regex module via PyO3.
It allows you to compile fuzzy (approximate) regular expressions with per-pattern thresholds and use Python’s advanced regex engine directly from Rust.

Features

Compile Python regex patterns (including fuzzy matching) from Rust
Set individual error thresholds per pattern (insertion, deletion, substitution)
Call search, finditer, sub, and extract match groups, start/end positions
Thread-safe (after calling pyo3::prepare_freethreaded_python())
Minimal dependencies

Requirements

Python 3.x with the regex package installed
pyo3 0.24.0

Installation

Add this to your Cargo.toml:

[dependencies]
pyo3 = "0.24.0"
py-regex = "0.1"

Ensure that you have the Python regex module:

pip install regex

Quick Start

Call pyo3::prepare_freethreaded_python() once at program start:

fn main() -> pyo3::PyResult<()> {
    // Initialize the Python interpreter for multi-threaded use
    pyo3::prepare_freethreaded_python();

    // Your code here...
    Ok(())
}

Compile a Pattern

use py_regex::PyRegex;

let re = PyRegex::new(r"(?i)\bhello\b")?;
assert!(re.is_match("Hello, world!")?);

Fuzzy Matching

// Automatically uses fuzzy thresholds per pattern
let fuzzy = PyRegex::new(r"(rust){e<=2}")?;
let m = fuzzy.find_iter("ruxy").pop();
assert!(m.is_some());

Find All Matches

let re = PyRegex::new(r"\d+")?;
for m in re.find_iter("123 abc 456")? {
    println!("Match: {:?}", m.group(0)?);
}

Replace (Substitution)

let re = PyRegex::new(r"\d+")?;
let result = re.replace("There are 123 apples", "NUM")?;
assert_eq!(result, "There are NUM apples");

Extract Groups

let re = PyRegex::new(r"(\w+)@(\w+\.\w+)")?;
if let Some(m) = re.find_iter("user@example.com").pop() {
let parts = m.groups()?; // Vec<Option<String>>
// parts[0] is the full match, parts[1] username, parts[2] domain
}

API

`PyRegex::new(pattern: &str) -> PyResult<PyRegex>`

Compile a Python regex pattern.

`PyRegex::is_match(text: &str) -> PyResult<bool>`

Return true if search(text) finds a match.

`PyRegex::find_iter(text: &str) -> PyResult<Vec<PyRegexMatch>>`

Return all non-overlapping matches as PyRegexMatch.

`PyRegex::replace(text: &str, replacement: &str) -> PyResult<String>`

Perform substitution (sub) on the input text.

`PyRegexMatch`

group(idx: usize) -> PyResult<Option<String>>
groups() -> PyResult<Vec<Option<String>>>
start(idx: usize) -> PyResult<isize>
end(idx: usize) -> PyResult<isize>

License

This project is licensed under the MIT License. See LICENSE for details.

py-regex 0.1.1