iptv/M3U8/scrapers/utils/network.py
2025-09-17 23:52:37 -04:00

150 lines
4.5 KiB
Python

import asyncio
import logging
import re
from collections.abc import Callable
from typing import Any
import httpx
from playwright.async_api import Browser, BrowserContext, Playwright, Request
from .config import UA
CLIENT = httpx.AsyncClient(
timeout=5,
follow_redirects=True,
headers={"User-Agent": UA},
)
async def check_status(client: httpx.AsyncClient, url: str) -> bool:
try:
r = await client.get(url)
r.raise_for_status()
except Exception:
return False
return r.status_code == 200
async def get_base(client: httpx.AsyncClient, mirrors: list[str]) -> str | None:
tasks = [check_status(client, link) for link in mirrors]
results = await asyncio.gather(*tasks)
try:
return [url for url, ok in zip(mirrors, results) if ok][0]
except IndexError:
return
async def safe_process_event(
fn: Callable,
url_num: int,
timeout: int | float = 20,
log: logging.Logger | None = None,
) -> Any | None:
if not log:
log = logging.getLogger(__name__)
task = asyncio.create_task(fn())
try:
return await asyncio.wait_for(task, timeout=timeout)
except asyncio.TimeoutError:
log.warning(f"URL {url_num}) Timed out after {timeout}s, skipping event")
task.cancel()
try:
await task
except asyncio.CancelledError:
pass
except Exception as e:
log.debug(f"URL {url_num}) Ignore exception after timeout: {e}")
def capture_req(
req: Request,
captured: list[str],
got_one: asyncio.Event,
) -> None:
valid_m3u8 = re.compile(r"^(?!.*(amazonaws|knitcdn)).*\.m3u8")
if valid_m3u8.search(req.url):
captured.append(req.url)
got_one.set()
async def new_browser(
playwright: Playwright,
browser: str = "firefox",
ignore_https_errors: bool = False,
) -> tuple[Browser, BrowserContext]:
if browser == "brave":
brwsr = await playwright.chromium.connect_over_cdp("http://localhost:9222")
context = brwsr.contexts[0]
else:
brwsr = await playwright.firefox.launch(headless=True)
context = await brwsr.new_context(
user_agent=UA,
ignore_https_errors=ignore_https_errors,
viewport={"width": 1366, "height": 768},
device_scale_factor=1,
locale="en-US",
timezone_id="America/New_York",
color_scheme="dark",
permissions=["geolocation"],
extra_http_headers={
"Accept-Language": "en-US,en;q=0.9",
"Upgrade-Insecure-Requests": "1",
},
)
await context.add_init_script(
"""
Object.defineProperty(navigator, 'webdriver', {get: () => undefined});
Object.defineProperty(navigator, 'languages', {
get: () => ['en-US', 'en']
});
Object.defineProperty(navigator, 'plugins', {
get: () => [1, 2, 3, 4]
});
const elementDescriptor = Object.getOwnPropertyDescriptor(HTMLElement.prototype, 'offsetHeight');
Object.defineProperty(HTMLDivElement.prototype, 'offsetHeight', {
...elementDescriptor,
get: function() {
if (this.id === 'modernizr') { return 24; }
return elementDescriptor.get.apply(this);
}
});
Object.defineProperty(window.screen, 'width', { get: () => 1366 });
Object.defineProperty(window.screen, 'height', { get: () => 768 });
const getParameter = WebGLRenderingContext.prototype. getParameter;
WebGLRenderingContext.prototype.getParameter = function (param) {
if (param === 37445) return "Intel Inc."; // UNMASKED_VENDOR_WEBGL
if (param === 37446) return "Intel Iris OpenGL Engine"; // UNMASKED_RENDERER_WEBGL
return getParameter.apply(this, [param]);
};
const observer = new MutationObserver(mutations => {
mutations.forEach(mutation => {
mutation.addedNodes.forEach(node => {
if (node.tagName === 'IFRAME' && node.hasAttribute('sandbox')) {
node.removeAttribute('sandbox');
}
});
});
});
observer.observe(document.documentElement, { childList: true, subtree: true });
"""
)
return brwsr, context