e

M3U8/scrapers/streamed.py

@@ -4,7 +4,7 @@ from functools import partial
 from pathlib import Path
 
 import httpx
-from playwright.async_api import async_playwright
+from playwright.async_api import BrowserContext, async_playwright
 from selectolax.parser import HTMLParser
 
 from .utils import Cache, Time, get_logger, leagues, network
@@ -20,56 +20,56 @@ HTML_CACHE = Cache(Path(__file__).parent / "caches" / "streamed_php.json", exp=8
 BASE_URL = "https://streamed.site/webmaster.php"
 
 
-async def process_event(url: str, url_num: int) -> str | None:
-    async with async_playwright() as p:
-        browser, context = await network.browser(p, browser="brave")
+async def process_event(
+    url: str,
+    url_num: int,
+    context: BrowserContext,
+) -> str | None:
+    page = await context.new_page()
 
-        page = await context.new_page()
+    captured: list[str] = []
 
-        captured: list[str] = []
+    got_one = asyncio.Event()
 
-        got_one = asyncio.Event()
+    handler = partial(network.capture_req, captured=captured, got_one=got_one)
 
-        handler = partial(network.capture_req, captured=captured, got_one=got_one)
+    page.on("request", handler)
 
-        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 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}")
+            await asyncio.wait_for(wait_task, timeout=10)
+        except asyncio.TimeoutError:
+            log.warning(f"URL {url_num}) Timed out waiting for M3U8.")
             return
 
         finally:
-            page.remove_listener("request", handler)
-            await page.close()
-            await browser.close()
+            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) -> dict[str, str]:
@@ -168,29 +168,38 @@ async def scrape(client: httpx.AsyncClient) -> None:
 
     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,
-        )
+    async with async_playwright() as p:
+        browser, context = await network.browser(p, browser="brave")
 
-        if url:
-            sport, event, ts = ev["sport"], ev["event"], ev["event_ts"]
+        for i, ev in enumerate(events, start=1):
+            url = await network.safe_process(
+                lambda: process_event(
+                    ev["link"],
+                    url_num=i,
+                    context=context,
+                ),
+                url_num=i,
+                log=log,
+            )
 
-            tvg_id, logo = leagues.info(sport)
+            if url:
+                sport, event, ts = ev["sport"], ev["event"], ev["event_ts"]
 
-            key = f"[{sport}] {event} (STRMD)"
+                tvg_id, logo = leagues.info(sport)
 
-            entry = {
-                "url": url,
-                "logo": logo,
-                "base": "",
-                "timestamp": ts,
-                "id": tvg_id or "Live.Event.us",
-            }
+                key = f"[{sport}] {event} (STRMD)"
 
-            urls[key] = cached_urls[key] = entry
+                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)")