import re
from dataclasses import dataclass
from datetime import datetime
from typing import Optional
import locale 
LOG_PATTERN = re.compile(
    r'(?P<host>\S+) - - \[(?P<time>.+)\] "(?P<request>.*)" (?P<status_code>\d+) (?P<bytes_sent>\d+|-)'
)

@dataclass
class ApacheLogEntry:
    host: str
    timestamp: datetime
    method: str
    path: str
    status_code: int
    bytes_sent: Optional[int]
    original_line: str

    @classmethod
    def from_log(cls, log_string):
        match = LOG_PATTERN.match(log_string.strip())
        if not match:
            return None

        try:
            locale.setlocale(locale.LC_ALL, "en_US.UTF-8")

            request = match.group("request").split(" ")

            host = match.group('host')
            timestamp = datetime.strptime(match.group("time"), "%d/%b/%Y:%H:%M:%S %z")
            method = request[0]
            path = request[1]
            status_code = int(match.group('status_code'))
            bytes_sent = int(match.group('bytes_sent')) if match.group('bytes_sent') != '-' else None
        except:
            return None

        return cls(host, timestamp, method, path, status_code, bytes_sent, log_string)