"""Extracteur VidMoly (vidmoly.to / vidmoly.me) — page embed → mp4/m3u8.

Le lecteur jwplayer est configuré dans du JS parfois packé (p,a,c,k,e,d) :
`sources: [{file: "...m3u8"}]`. On dépacke si besoin puis on extrait par regex.
"""

import logging
import re

from app.scrapers.base import HosterExtractor, ScrapeError, VideoLink, register_hoster
from app.scrapers.hosters._packer import unpack_packed_js
from app.scrapers.http import fetch

logger = logging.getLogger(__name__)

_PATTERNS = (
    re.compile(r'sources\s*:\s*\[\s*\{[^}]*?file\s*:\s*["\'](?P<url>[^"\']+)["\']'),
    re.compile(r'file\s*:\s*["\'](?P<url>https?://[^"\']+?\.(?:m3u8|mp4)[^"\']*)["\']'),
    re.compile(r'["\'](?P<url>https?://[^"\']*?\.(?:m3u8|mp4)(?:\?[^"\']*)?)["\']'),
)


@register_hoster
class VidmolyExtractor(HosterExtractor):
    name = "vidmoly"
    domains = ("vidmoly.to", "vidmoly.me", "vidmoly.biz", "vidmoly.com")

    async def extract(self, embed_url: str) -> VideoLink:
        html = await fetch(embed_url)
        candidates: list[str] = []
        for content in (html, unpack_packed_js(html) or ""):
            for pattern in _PATTERNS:
                for match in pattern.finditer(content):
                    candidates.append(match.group("url"))
        video = [u for u in candidates if re.search(r"\.(mp4|m3u8)(\?|$)", u)]
        url = next((u for u in video if ".mp4" in u), None) or (video[0] if video else None)
        if not url:
            raise ScrapeError(f"vidmoly : URL vidéo introuvable dans {embed_url}")
        return VideoLink(
            url=url,
            hoster=self.name,
            headers={"Referer": embed_url},
            is_hls=".m3u8" in url,
        )