email.py

"""Module for the email detector."""

from typing import Any, Dict, List, Tuple

import regex as re

from src.detections import EmailDetection
from src.detectors.interface import Detector


EMAIL_REGEX = re.compile(
    r"(?P<local_part>[a-z0-9!#$%&\'*+/=?^_`{|}~-]+"
    r"(?:\.[a-z0-9!#$%&\'*+/=?^_`{|}~-]+)*@)"
    r"(?P<domain>(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+)"
    r"(?P<tld>[a-z0-9](?:[a-z0-9-]*[a-z0-9])?)",
    re.I,
)


class EmailDetector(Detector):
    """Detector for emails."""

    def detect(
        self, text: str, annotations: Dict[str, List[Tuple[int, int, Any]]]
    ) -> List[Tuple[int, int, EmailDetection]]:
        """Detects emails in the text.

        Args:
            text (str): the text to be searched

        Returns:
            List[Tuple[int, int, EmailDetection]]: a list of tuples containing
                (start, end, entity_type)

        """
        matches = EMAIL_REGEX.finditer(text)
        emails = []
        for match in matches:
            emails.append((match.start(), match.end(), EmailDetection()))

        return emails