e

M3U8/scrapers/streamed.py

@@ -0,0 +1,214 @@
+import asyncio
+from datetime import datetime, timedelta
+from functools import partial
+from pathlib import Path
+
+import httpx
+from playwright.async_api import async_playwright
+from selectolax.parser import HTMLParser
+
+from .utils import (
+    TZ,
+    capture_req,
+    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" / "streamed.json"
+
+HTML_CACHE = Path(__file__).parent / "caches" / "streamed_php.json"
+
+base_url = "https://streamed.site/webmaster.php"
+
+
+def get_date(s: str) -> datetime:
+    try:
+        return datetime.strptime(s, "%Y-%m-%d %H:%M %Z").astimezone(TZ)
+    except ValueError:
+        s = s.replace("ET", "").strip()
+        return datetime.strptime(s, "%Y-%m-%d %H:%M").astimezone(TZ)
+
+
+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 refresh_html_cache(client: httpx.AsyncClient, url: str) -> 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 = {}
+
+    for row in soup.css("div.wrap div.row"):
+        if not (date := row.css_first("div.date")):
+            continue
+
+        event_dt = get_date(date.text(strip=True))
+
+        if event_dt.date() != now.date():
+            continue
+
+        league = row.css_first("div.league")
+
+        title = row.css_first("div.title")
+
+        hds_a = row.css_first("div.hds a")
+
+        if not (league and title and hds_a):
+            continue
+
+        sport, event = league.text(strip=True), title.text(strip=True)
+
+        onclick = hds_a.attributes.get("onclick", "")
+
+        if not (chnl_id := "".join(s for s in onclick if s.isdigit())):
+            continue
+
+        key = f"[{sport}] {event} (STRMD)"
+
+        events[key] = {
+            "sport": sport,
+            "event": event,
+            "link": f"https://streamed.site/set.php?{chnl_id}",
+            "ts": event_dt.timestamp(),
+        }
+
+    return events
+
+
+async def get_events(
+    client: httpx.AsyncClient,
+    url: str,
+    cached_keys: set[str],
+) -> list[dict[str, str]]:
+
+    if not (events := load_cache(HTML_CACHE, exp=10_800)):
+        events = await refresh_html_cache(client, url)
+        write_cache(HTML_CACHE, events)
+
+    live = []
+    start_ts = (now - timedelta(minutes=30)).timestamp()
+    end_ts = (now + timedelta(minutes=30)).timestamp()
+
+    for k, v in events.items():
+        if cached_keys & {k}:
+            continue
+
+        if not start_ts <= v["ts"] < end_ts:
+            continue
+
+        live.append({**v})
+
+    return live
+
+
+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")
+
+    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)
+
+            key = f"[{sport}] {event} (STRMD)"
+
+            entry = {
+                "url": url,
+                "logo": logo,
+                "base": "https://streamed.site/",
+                "timestamp": now.timestamp(),
+                "id": tvg_id,
+            }
+
+            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)