iptv/M3U8/scrapers/streameast.py

import asyncio
from datetime import datetime, timedelta
from functools import partial
from pathlib import Path
from urllib.parse import urljoin

import httpx
from playwright.async_api import async_playwright
from selectolax.parser import HTMLParser

from .utils import (
    TZ,
    capture_req,
    get_base,
    get_logger,
    leagues,
    load_cache,
    new_browser,
    now,
    safe_process_event,
    write_cache,
)

log = get_logger(__name__)

urls: dict[str, dict[str, str | float]] = {}

CACHE_FILE = Path(__file__).parent / "caches" / "streameast.json"

MIRRORS = [
    "https://streameast.ga",
    "https://streameast.tw",
    "https://streameast.ph",
    "https://streameast.sg",
    "https://streameast.ch",
    "https://streameast.ec",
    "https://streameast.fi",
    "https://streameast.ms",
    "https://streameast.ps",
    "https://streameast.cf",
    "https://streameast.sk",
    "https://thestreameast.co",
    "https://thestreameast.fun",
    "https://thestreameast.ru",
    "https://thestreameast.su",
]


async def process_event(url: str, url_num: int) -> str | None:
    async with async_playwright() as p:
        browser, context = await new_browser(p, browser="brave")

        page = await context.new_page()

        captured: list[str] = []

        got_one = asyncio.Event()

        handler = partial(capture_req, captured=captured, got_one=got_one)

        page.on("request", handler)

        try:
            await page.goto(url, wait_until="domcontentloaded", timeout=15_000)

            wait_task = asyncio.create_task(got_one.wait())

            try:
                await asyncio.wait_for(wait_task, timeout=10)
            except asyncio.TimeoutError:
                log.warning(f"URL {url_num}) Timed out waiting for M3U8.")
                return

            finally:
                if not wait_task.done():
                    wait_task.cancel()

                    try:
                        await wait_task
                    except asyncio.CancelledError:
                        pass

            if captured:
                log.info(f"URL {url_num}) Captured M3U8")

                return captured[-1]

            log.warning(f"URL {url_num}) No M3U8 captured after waiting.")
            return

        except Exception as e:
            log.warning(f"URL {url_num}) Exception while processing: {e}")
            return

        finally:
            page.remove_listener("request", handler)
            await page.close()
            await browser.close()


async def get_events(
    client: httpx.AsyncClient,
    url: str,
    cached_keys: set[str],
) -> list[dict[str, str]]:
    try:
        r = await client.get(url)
        r.raise_for_status()
    except Exception as e:
        log.error(f'Failed to fetch "{url}"\n{e}')

        return []

    soup = HTMLParser(r.text)
    events = []

    start_dt = now - timedelta(minutes=30)
    end_dt = now + timedelta(minutes=30)

    for section in soup.css("div.se-sport-section"):
        if not (sport := section.attributes.get("data-sport-name", "").strip()):
            continue

        for a in section.css("a.uefa-card"):
            href = urljoin(url, a.attributes.get("href", ""))

            team_spans = [t.text(strip=True) for t in a.css("span.uefa-name")]

            if len(team_spans) == 2:
                name = f"{team_spans[0]} vs {team_spans[1]}"

            elif len(team_spans) == 1:
                name = team_spans[0]

            else:
                continue

            if not (time_span := a.css_first(".uefa-time")):
                continue

            time_text = time_span.text(strip=True)
            timestamp = int(time_span.attributes.get("data-time", 0))

            key = f"[{sport}] {name} (SEAST)"

            if cached_keys & {key}:
                continue

            event_dt = datetime.fromtimestamp(timestamp, TZ)

            if time_text == "LIVE" or (start_dt <= event_dt < end_dt):
                events.append(
                    {
                        "sport": sport,
                        "event": name,
                        "link": href,
                    }
                )

    return events


async def scrape(client: httpx.AsyncClient) -> None:
    cached_urls = load_cache(CACHE_FILE, exp=10_800)
    cached_count = len(cached_urls)
    urls.update(cached_urls)

    log.info(f"Collected {cached_count} event(s) from cache")

    if not (base_url := await get_base(client, MIRRORS)):
        log.warning("No working StreamEast mirrors")
        write_cache(CACHE_FILE, cached_urls)
        return

    log.info(f'Scraping from "{base_url}"')

    events = await get_events(
        client,
        base_url,
        set(cached_urls.keys()),
    )

    log.info(f"Processing {len(events)} new URL(s)")

    for i, ev in enumerate(events, start=1):
        url = await safe_process_event(
            lambda: process_event(ev["link"], url_num=i),
            url_num=i,
            log=log,
        )

        if url:
            sport, event = ev["sport"], ev["event"]

            tvg_id, logo = leagues.info(sport)

            if sport == "NBA" and leagues.is_valid(event, "WNBA"):
                sport = "WNBA"
                tvg_id, logo = leagues.info("WNBA")

            key = f"[{sport}] {event} (SEAST)"

            entry = {
                "url": url,
                "logo": logo,
                "base": base_url,
                "timestamp": now.timestamp(),
                "id": tvg_id or "Live.Event.us",
            }

            urls[key] = cached_urls[key] = entry

    if new_count := len(cached_urls) - cached_count:
        log.info(f"Collected and cached {new_count} new event(s)")
    else:
        log.info("No new events found")

    write_cache(CACHE_FILE, cached_urls)