Multithreading

Due to Global Interpreter Lock (GIL) implemented in Python interpreter, multithreading in Python can‘t be truly executed in parallel. In matx, we implement matx.pmap to support multithreading in native.

import matx
from typing import Any, List

def lower_impl(s: str) -> str:
    return s.lower()

def MyParallelLower(inputs: List[str]) -> List[str]:
    return matx.pmap(lower_impl, inputs)

p_lower = matx.script(MyParallelLower)
print(p_lower(["Hello"]))

Regular Expression

Matx implements a builtin regular expression engine based on PCRE.

import matx
from typing import Any, List

class Spliter:
    def __init__(self) -> None:
        self.regex: Any = matx.Regex(
            r'(([\v\f\r\n]+)|(?<=[^？。；，！!?][？。；，！!?])(?=[^？。；，！!?]))')

    def __call__(self, ss: str) -> List[str]:
        return self.regex.split(ss)

spliter = matx.script(Spliter)()
print(spliter("hELLO \v WORLD"))