iptv/M3U8/scrapers/ppv.py

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

import httpx
from playwright.async_api import async_playwright

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

log = get_logger(__name__)

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

API_FILE = Cache(Path(__file__).parent / "caches" / "ppv_api.json", exp=86_400)

CACHE_FILE = Cache(Path(__file__).parent / "caches" / "ppv.json", exp=10_800)

MIRRORS = [
    "https://ppvs.su",
    "https://ppv.to",
    "https://ppv.wtf",
    "https://ppv.land",
    "https://freeppv.fun",
]


def get_tvg(sport: str, event: str) -> str:
    match sport:
        case "American Football":
            if leagues.is_valid(event, "NFL"):
                return "NFL.Dummy.us"

            else:
                return "NCAA.Sports.Dummy.us"

        case "Basketball":
            if leagues.is_valid(event, "NBA"):
                return "NBA.Basketball.Dummy.us"

            elif leagues.is_valid(event, "WNBA"):
                return "WNBA.dummy.us"

            # NCAA

            else:
                return "Basketball.Dummy.us"

        case _:
            return leagues.info(sport)[0]


async def refresh_api_cache(
    client: httpx.AsyncClient, url: str
) -> dict[str, dict[str, str]]:
    log.info("Refreshing API cache")

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

    return r.json()


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

        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=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,
    base_url: str,
    cached_keys: set[str],
) -> list[dict[str, str]]:

    events: list[dict[str, str]] = []

    if not (api_data := API_FILE.load(nearest_hr=True, per_entry=False)):
        api_data = await refresh_api_cache(client, urljoin(base_url, "api/streams"))

        API_FILE.write(api_data)

    for stream_group in api_data.get("streams", []):
        sport = stream_group("category", [])

        if sport == "24/7 Streams":
            continue

        for event in stream_group.get("streams", []):
            name, start_ts, end_ts, logo, uri_name = (
                event["name"],
                event["starts_at"],
                event["ends_at"],
                event["poster"],
                event["uri_name"],
            )

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

            if cached_keys & {key}:
                continue

            start_dt = Time.from_ts(start_ts).delta(minutes=-30)

            end_dt = Time.from_ts(end_ts).delta(minutes=30)

            if not start_dt <= Time.now() < end_dt:
                continue

            events.append(
                {
                    "sport": sport,
                    "event": name,
                    "link": urljoin(base_url, f"live/{uri_name}"),
                    "logo": logo,
                    "timestamp": start_dt.timestamp(),
                }
            )

    return events


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")

    if not (base_url := await network.get_base(MIRRORS)):
        log.warning("No working PPV mirrors")
        CACHE_FILE.write(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 network.safe_process(
            lambda: process_event(ev["link"], url_num=i),
            url_num=i,
            log=log,
        )

        if url:
            sport, event, logo, ts = (
                ev["sport"],
                ev["event"],
                ev["logo"],
                ev["timestamp"],
            )

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

            entry = {
                "url": url,
                "logo": logo,
                "base": base_url,
                "timestamp": ts,
                "id": get_tvg(sport, event) 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")

    CACHE_FILE.write(cached_urls)