Messenger_1Million_Stats/million/analyze/message_evaluation.py

import re
from typing import Dict
from million.model.message import Message
import million.analyze.media_count_mapper as mcm


_memoization: Dict[Message, int] = {}
_dns_solver: mcm.MediaCountMapper = mcm.MediaCountMapper()


def get(msg: Message) -> int:
    """
    Returns the estimated value counted in this message
    """
    return _memoization.get(msg, _compute(msg))


def reset(msg: Message) -> None:
    """
    Drop memorized value of this Message
    """
    if msg in _memoization:
        _memoization.pop(msg)


def reset() -> None:
    """
    Drop every memorized message value
    """
    _memoization.clear()


def _compute(msg: Message) -> int:
    value = _dns_solver.solve(msg) or _computeContent(msg) or None

    _memoization[msg] = value
    return value


def _computeContent(msg: Message) -> int:
    if not msg.content:
        return
    
    s = re.sub(r'[^\s\d.,]|[.,]{2,}',"", msg.content)
    match = re.search(r"\d+", s)

    if match:
        return int(match.group())

    return None