Fix unmaintained pytube 15 issue (see pytube/pytube#1954 (comment))

ytdownload.py

@@ -1,12 +1,55 @@
-import os
+import os, re
 from time import sleep
-from pytube import YouTube, Playlist
+from pytube import YouTube, Playlist, cipher
 from moviepy.editor import *
 import argparse
 from multiprocessing import Process, Queue, cpu_count, freeze_support
 
+
+def get_throttling_function_name(js: str) -> str:
+    """Extract the name of the function that computes the throttling parameter.
+
+    :param str js:
+        The contents of the base.js asset file.
+    :rtype: str
+    :returns:
+        The name of the function used to compute the throttling parameter.
+    """
+    function_patterns = [
+        # https://github.com/ytdl-org/youtube-dl/issues/29326#issuecomment-865985377
+        # https://github.com/yt-dlp/yt-dlp/commit/48416bc4a8f1d5ff07d5977659cb8ece7640dcd8
+        # var Bpa = [iha];
+        # ...
+        # a.C && (b = a.get("n")) && (b = Bpa[0](b), a.set("n", b),
+        # Bpa.length || iha("")) }};
+        # In the above case, `iha` is the relevant function name
+        r'a\.[a-zA-Z]\s*&&\s*\([a-z]\s*=\s*a\.get\("n"\)\)\s*&&\s*' r"\([a-z]\s*=\s*([a-zA-Z0-9$]+)(\[\d+\])?\([a-z]\)",
+        r"\([a-z]\s*=\s*([a-zA-Z0-9$]+)(\[\d+\])\([a-z]\)",
+    ]
+    # logger.debug('Finding throttling function name')
+    for pattern in function_patterns:
+        regex = re.compile(pattern)
+        function_match = regex.search(js)
+        if function_match:
+            # logger.debug("finished regex search, matched: %s", pattern)
+            if len(function_match.groups()) == 1:
+                return function_match.group(1)
+            idx = function_match.group(2)
+            if idx:
+                idx = idx.strip("[]")
+                array = re.search(r"var {nfunc}\s*=\s*(\[.+?\]);".format(nfunc=re.escape(function_match.group(1))), js)
+                if array:
+                    array = array.group(1).strip("[]").split(",")
+                    array = [x.strip() for x in array]
+                    return array[int(idx)]
+
+    raise re.RegexMatchError(caller="get_throttling_function_name", pattern="multiple")
+
+
+cipher.get_throttling_function_name = get_throttling_function_name
+
 processes = {}
-max_proccesses = cpu_count()
+max_proccesses = max(cpu_count() / 2, 1)
 queue = Queue()