data_juicer.ops.mapper.clean_ip_mapper 源代码

from typing import Optional

import regex as re

from ..base_op import OPERATORS, Mapper




@OPERATORS.register_module("clean_ip_mapper")
class CleanIpMapper(Mapper):
    """Cleans IPv4 and IPv6 addresses from text samples.

    This operator removes or replaces IPv4 and IPv6 addresses in the text. It uses a regular
    expression to identify and clean the IP addresses. By default, it replaces the IP
    addresses with an empty string, effectively removing them. The operator can be
    configured with a custom pattern and replacement string. If no pattern is provided, a
    default pattern for both IPv4 and IPv6 addresses is used. The operator processes samples
    in batches.

    - Uses a regular expression to find and clean IP addresses.
    - Replaces found IP addresses with a specified replacement string.
    - Default replacement string is an empty string, which removes the IP addresses.
    - Can use a custom regular expression pattern if provided.
    - Processes samples in batches for efficiency."""

    _batched_op = True



    def __init__(self, pattern: Optional[str] = None, repl: str = "", *args, **kwargs):
        """
        Initialization method.

        :param pattern: regular expression pattern to search for within text.
        :param repl: replacement string, default is empty string.
        :param args: extra args
        :param kwargs: extra args
        """

        super().__init__(*args, **kwargs)
        if pattern is None:
            ipv4 = r"(?:(?:1[0-9][0-9]\.)|(?:2[0-4][0-9]\.)|"
            ipv4 += r"(?:25[0-5]\.)|(?:[1-9][0-9]\.)|(?:[0-9]\.))"
            ipv4 += r"{3}(?:(?:1[0-9][0-9])|(?:2[0-4][0-9])|"
            ipv4 += r"(?:25[0-5])|(?:[1-9][0-9])|(?:[0-9]))"
            self.pattern = ipv4 + "|"
            self.pattern += r"([\da-fA-F]{1,4}:){7}[\da-fA-F]{1,4}"  # ipv6
            # compressed ipv6 ("::"), not touching other ASCII word characters
            # or colons, so "std::cout" and partial matches are left alone
            hextet = r"[\da-fA-F]{1,4}"
            compressed = "|".join(
                [
                    rf"(?:{hextet}:){{1,7}}:",
                    rf"(?:{hextet}:){{1,6}}:{hextet}",
                    rf"(?:{hextet}:){{1,5}}(?::{hextet}){{1,2}}",
                    rf"(?:{hextet}:){{1,4}}(?::{hextet}){{1,3}}",
                    rf"(?:{hextet}:){{1,3}}(?::{hextet}){{1,4}}",
                    rf"(?:{hextet}:){{1,2}}(?::{hextet}){{1,5}}",
                    rf"{hextet}:(?::{hextet}){{1,6}}",
                    rf":(?::{hextet}){{1,7}}",
                ]
            )
            # ipv6 ending in an ipv4 address (::ffff:192.0.2.128, 64:ff9b::198.51.100.7),
            # matched whole; no match may end right before a dotted ipv4 part
            embedded = "|".join(
                [rf"(?:{hextet}:){{6}}", rf"::(?:{hextet}:){{0,5}}"]
                + [rf"(?:{hextet}:){{{n}}}:(?:{hextet}:){{0,{5 - n}}}" for n in range(1, 6)]
            )
            # the first lookahead only skips quickly past text where no address can start
            self.pattern += rf"|(?<![0-9A-Za-z_:])(?=[\da-fA-F]{{0,4}}:)"
            self.pattern += rf"(?:(?:{embedded}){ipv4}|{compressed})(?![0-9A-Za-z_:]|\.\d)"
        else:
            self.pattern = pattern
            if (len(pattern) > 2) and (
                pattern.startswith("r'") and pattern.endswith("'") or pattern.startswith('r"') and pattern.endswith('"')
            ):
                self.pattern = pattern[2:-1]
        self.repl = repl




    def process_batched(self, samples):
        for idx, text in enumerate(samples[self.text_key]):
            if not re.search(self.pattern, text, flags=re.DOTALL):
                continue
            samples[self.text_key][idx] = re.sub(pattern=self.pattern, repl=self.repl, string=text, flags=re.DOTALL)
        return samples