Untitled
Posted by Anonymous on Thu 13th Aug 2026 03:29
raw | new post

#!/usr/bin/env python3

import sys

from invisible_playwright import InvisiblePlaywright

INIT_SCRIPT = """
() => {
    window.__m3u8FetchState = { count: 0, secondUrl: null };

    const originalFetch = window.fetch.bind(window);

    window.fetch = function (input, init) {
        try {
            const rawUrl =
                typeof input === 'string'
                    ? input
                    : input instanceof URL
                    ? input.toString()
                    : input && input.url
                    ? input.url
                    : String(input);

            const resolved = new URL(rawUrl, window.location.href);
            const pathOnly = resolved.pathname; // ignore query string / hash

            if (pathOnly.toLowerCase().endsWith('.m3u8')) {
                window.__m3u8FetchState.count += 1;
                if (window.__m3u8FetchState.count === 2 && !window.__m3u8FetchState.secondUrl) {
                    window.__m3u8FetchState.secondUrl = resolved.href;
                }
            }
        } catch (e) {
            // If URL parsing fails for some exotic input, just let the
            // fetch proceed untouched rather than breaking the page.
        }

        return originalFetch(input, init);
    };
}
"""


def main() -> int:
    if len(sys.argv) < 2:
        print("Usage: python second_m3u8_fetch.py <url>", file=sys.stderr)
        return 1

    url = sys.argv[1]

    with InvisiblePlaywright() as browser:
        page = browser.new_page()

        # Installed before any page JS runs.
        page.add_init_script(INIT_SCRIPT)

        try:
            page.goto(url, wait_until="domcontentloaded")
        except Exception as err:
            print(f"Navigation failed: {err}", file=sys.stderr)
            return 3

        print("Waiting for the second fetch()-ed .m3u8 URL (no timeout)...", file=sys.stderr)

        # timeout=0 means wait indefinitely. If the page only ever fetches
        # one .m3u8 file, this call simply never returns — by design.
        handle = page.wait_for_function(
            "() => window.__m3u8FetchState.secondUrl", timeout=0
        )
        second_m3u8_url = handle.json_value()

    print(second_m3u8_url)
    return 0


if __name__ == "__main__":
    sys.exit(main())