e

M3U8/scrapers/roxie.py

@@ -0,0 +1,226 @@
+import asyncio
+from functools import partial
+from pathlib import Path
+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]] = {}
+
+BASE_URL = "https://roxiestreams.cc"
+
+SPORT_URLS = {
+    sport: urljoin(BASE_URL, sport.lower())
+    for sport in ["Soccer", "MLB", "NBA", "NFL", "Fighting", "Motorsports"]
+}
+
+CACHE_FILE = Cache(Path(__file__).parent / "caches" / "roxie.json", exp=10_800)
+
+HTML_CACHE = Cache(Path(__file__).parent / "caches" / "roxie_html.json", exp=86_400)
+
+
+async def process_event(
+    url: str,
+    url_num: int,
+    context: BrowserContext,
+) -> 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=15_000,
+        )
+
+        wait_task = asyncio.create_task(got_one.wait())
+
+        try:
+            await asyncio.wait_for(wait_task, timeout=6)
+        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()
+
+
+async def refresh_html_cache(
+    client: httpx.AsyncClient,
+    url: str,
+    sport: 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)
+
+    events = {}
+
+    for row in soup.css("table#eventsTable tbody tr"):
+        a_tag = row.css_first("td a")
+
+        if not a_tag:
+            continue
+
+        event = a_tag.text(strip=True)
+
+        event_link = a_tag.attributes.get("href")
+
+        if not (span := row.css_first("span.countdown-timer")):
+            continue
+
+        data_start = span.attributes["data-start"]
+
+        event_dt = Time.from_str(f"{data_start} PST", "%B %d, %Y %H:%M:%S")
+
+        key = f"[{sport}] {event} (ROXIE)"
+
+        events[key] = {
+            "sport": sport,
+            "event": event,
+            "link": event_link,
+            "event_ts": event_dt.timestamp(),
+            "timestamp": Time.now().timestamp(),
+        }
+
+    return events
+
+
+async def get_events(
+    client: httpx.AsyncClient,
+    sport_urls: dict[str, str],
+    cached_keys: set[str],
+) -> list[dict[str, str]]:
+
+    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 = []
+
+    now = Time.clean(Time.now())
+    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}"')
+
+    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)
+
+            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.info(sport)
+
+                    key = f"[{sport}] {event} (ROXIE)"
+
+                    entry = {
+                        "url": url,
+                        "logo": logo,
+                        "base": "",
+                        "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)