"""Extracteur SendVid (sendvid.com) — page embed → mp4 direct.

La page embed expose soit une balise `<source src="...">` ( lecteur HTML5),
soit une variable JS `video_source`. Extraction par regex, sans dépendance JS.
"""

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'<source[^>]+src=["\'](?P<url>[^"\']+)["\']'),
    re.compile(r'video_source\s*[:=]\s*["\'](?P<url>[^"\']+)["\']'),
    re.compile(r'["\'](?P<url>https?://[^"\']*?\.(?:mp4|m3u8)(?:\?[^"\']*)?)["\']'),
)


@register_hoster
class SendvidExtractor(HosterExtractor):
    name = "sendvid"
    domains = ("sendvid.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"))
        url = self._pick(candidates)
        if not url:
            raise ScrapeError(f"sendvid : URL vidéo introuvable dans {embed_url}")
        return VideoLink(
            url=url,
            hoster=self.name,
            headers={"Referer": embed_url},
            is_hls=".m3u8" in url,
        )

    @staticmethod
    def _pick(candidates: list[str]) -> str | None:
        video = [u for u in candidates if re.search(r"\.(mp4|m3u8)(\?|$)", u)]
        for url in video:
            if ".mp4" in url:
                return url
        return video[0] if video else None