iptv/M3U8/scrapers/old/volo.py

import asyncio
from functools import partial
from urllib.parse import urljoin

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

from .utils import Cache, Time, get_logger, leagues, network

log = get_logger(__name__)

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

CACHE_FILE = Cache("volo.json", exp=10_800)

HTML_CACHE = Cache("volo-html.json", exp=86_400)

BASE_URL = "http://volokit2.fun"

valid_sports = {
    # "boxing": "Boxing",
    # "college-football": "CFB",
    # "mens-college-basketball": "CBB",
    "mlb": "MLB",
    "mls": "Soccer",
    "nba": "NBA",
    "nfl": "NFL",
    "nhl": "NHL",
    # "race": "Racing",
    # "ufc": "UFC",
    "wnba": "WNBA",
}


def fix_event(s: str) -> str:
    return " ".join(x.capitalize() for x in s.split())


async def process_event(
    url: str,
    url_num: int,
    context: BrowserContext,
    timeout: int | float = 10,
) -> str | None:

    page = await context.new_page()

    captured: list[str] = []

    got_one = asyncio.Event()

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

    page.on("request", handler)

    try:
        await page.goto(
            url,
            wait_until="domcontentloaded",
            timeout=10_000,
        )

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

        try:

            iframe = page.locator("iframe").first

            src = await iframe.get_attribute("src")

            await page.goto(
                src,
                wait_until="domcontentloaded",
                timeout=10_000,
            )

            await page.click("#volokit_player")

            await asyncio.wait_for(wait_task, timeout=timeout)
        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[0]

        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()


async def refresh_html_cache(
    client: httpx.AsyncClient,
    url: str,
    sport: str,
) -> dict[str, dict[str, str | float]]:

    try:
        r = await client.get(url)
        r.raise_for_status()
    except Exception as e:
        log.error(f'Failed to fetch "{url}": {e}')

        return {}

    soup = HTMLParser(r.text)
    now = Time.clean(Time.now())

    events = {}

    for card in soup.css("#events .table .vevent.theevent"):
        if not (href := card.css_first("a").attributes.get("href")):
            continue

        name_node = card.css_first(".teamtd.event")
        time_node = card.css_first(".time")

        if not (name_node and time_node):
            continue

        name = name_node.text(strip=True)
        time = time_node.text(strip=True)

        event_sport = valid_sports[sport]
        event_name = fix_event(name)

        event_dt = Time.from_only_time(
            time,
            now.date(),
            timezone="UTC",
        )

        key = f"[{event_sport}] {event_name} (VOLO)"

        events[key] = {
            "sport": event_sport,
            "event": event_name,
            "link": href,
            "event_ts": event_dt.timestamp(),
            "timestamp": now.timestamp(),
        }

    return events


async def get_events(
    client: httpx.AsyncClient,
    sport_urls: dict[str, str],
    cached_keys: set[str],
) -> list[dict[str, str]]:

    now = Time.clean(Time.now())

    if not (events := HTML_CACHE.load()):
        tasks = [
            refresh_html_cache(
                client,
                url,
                sport,
            )
            for sport, url in sport_urls.items()
        ]

        results = await asyncio.gather(*tasks)

        events = {k: v for data in results for k, v in data.items()}

        HTML_CACHE.write(events)

    live = []

    start_ts = now.delta(minutes=-30).timestamp()
    end_ts = now.delta(minutes=30).timestamp()

    for k, v in events.items():
        if cached_keys & {k}:
            continue

        if not start_ts <= v["event_ts"] <= end_ts:
            continue

        live.append({**v})

    return live


async def scrape(client: httpx.AsyncClient) -> None:
    cached_urls = CACHE_FILE.load()
    cached_count = len(cached_urls)
    urls.update(cached_urls)

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

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

    sport_urls = {sport: urljoin(BASE_URL, f"sport/{sport}") for sport in valid_sports}

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

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

    if events:
        async with async_playwright() as p:
            browser, context = await network.browser(p, browser="brave")

            for i, ev in enumerate(events, start=1):
                handler = partial(
                    process_event,
                    url=ev["link"],
                    url_num=i,
                    context=context,
                )

                url = await network.safe_process(
                    handler,
                    url_num=i,
                    log=log,
                )

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

                    tvg_id, logo = leagues.get_tvg_info(sport, event)

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

                    entry = {
                        "url": url,
                        "logo": logo,
                        "base": BASE_URL,
                        "timestamp": ts,
                        "id": tvg_id or "Live.Event.us",
                    }

                    urls[key] = cached_urls[key] = entry

            await browser.close()

    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")

    CACHE_FILE.write(cached_urls)