code.delx.au - youtube-cgi/blob - youtube.cgi

   1 #!/usr/bin/env python3
   2
   3 import cgi
   4 import html.parser
   5 import http.cookiejar
   6 import json
   7 import os
   8 import re
   9 import shutil
  10 import subprocess
  11 import sys
  12 import time
  13 import urllib.error
  14 import urllib.parse
  15 import urllib.request
  16
  17
  18 USER_AGENT = "Mozilla/5.0 (X11; Linux x86_64; rv:67.0) Gecko/20100101 Firefox/67.0"
  19
  20 MIMETYPES = {
  21     "video/mp4": "mp4",
  22     "video/x-flv": "flv",
  23     "video/3gpp": "3gp",
  24 }
  25
  26 QUALITIES = {
  27     "hd1080": 5,
  28     "hd720": 4,
  29     "large": 3,
  30     "medium": 2,
  31     "small": 1,
  32 }
  33
  34
  35 class VideoUnavailable(Exception):
  36     pass
  37
  38 class NotYouTube(Exception):
  39     pass
  40
  41 def print_form(url="", msg=""):
  42     script_url = "https://%s%s" % (os.environ["HTTP_HOST"], os.environ["REQUEST_URI"])
  43     sys.stdout.write("Content-Type: text/html\r\n\r\n")
  44     sys.stdout.write("""
  45 <!DOCTYPE html>
  46 <html>
  47 <head>
  48     <title>delx.net.au - YouTube Scraper</title>
  49     <link rel="stylesheet" type="text/css" href="/style.css">
  50     <style type="text/css">
  51         input[type="text"] {
  52             width: 100%;
  53         }
  54         .error {
  55             color: red;
  56         }
  57     </style>
  58 </head>
  59 <body>
  60     <h1>delx.net.au - YouTube Scraper</h1>
  61     {0}
  62     <form action="" method="get">
  63     <p>This page will let you easily download YouTube videos to watch offline. It
  64     will automatically grab the highest quality version.</p>
  65     <div><input type="text" name="url" value="{1}"/></div>
  66     <div><input type="submit" value="Download!"/></div>
  67     </form>
  68     <p>Tip! Use this bookmarklet: <a href="javascript:(function(){window.location='{2}?url='+escape(location);})()">YouTube Download</a>
  69     to easily download videos. Right-click the link and add it to bookmarks,
  70     then when you're looking at a YouTube page select that bookmark from your
  71     browser's bookmarks menu to download the video straight away.</p>
  72 </body>
  73 </html>
  74 """.replace("{0}", msg).replace("{1}", url).replace("{2}", script_url))
  75
  76 cookiejar = http.cookiejar.CookieJar()
  77 urlopener = urllib.request.build_opener(urllib.request.HTTPCookieProcessor(cookiejar))
  78 referrer = ""
  79
  80 def urlopen(url, offset=None):
  81     if url.startswith("//"):
  82         url = "https:" + url
  83     if not url.startswith("http://") and not url.startswith("https://"):
  84         url = "https://www.youtube.com" + url
  85
  86     global referrer
  87     req = urllib.request.Request(url)
  88     if not referrer:
  89         referrer = url
  90     else:
  91         req.add_header("Referer", referrer)
  92
  93     req.add_header("User-Agent", USER_AGENT)
  94
  95     if offset:
  96         req.add_header("Range", "bytes=%d-" % offset)
  97
  98     res = urlopener.open(req)
  99
 100     content_range = res.getheader("Content-Range")
 101     if content_range:
 102         tokens = content_range.split()
 103         assert tokens[0] == "bytes"
 104         start = int(tokens[1].split("-")[0])
 105         assert start == offset
 106     return res
 107
 108 def validate_url(url):
 109     parsed_url = urllib.parse.urlparse(url)
 110     scheme_ok = parsed_url.scheme == "https"
 111     host_ok = parsed_url.netloc.lstrip("www.") in ["youtube.com", "youtu.be"]
 112
 113     if scheme_ok and host_ok:
 114         return
 115     else:
 116         raise NotYouTube()
 117
 118 def parse_url(url, parser):
 119     f = urlopen(url)
 120     parser.feed(f.read().decode("utf-8"))
 121     parser.close()
 122     f.close()
 123
 124 def append_to_qs(url, params):
 125     r = list(urllib.parse.urlsplit(url))
 126     qs = urllib.parse.parse_qs(r[3])
 127     qs.update(params)
 128     r[3] = urllib.parse.urlencode(qs, True)
 129     url = urllib.parse.urlunsplit(r)
 130     return url
 131
 132 def get_player_config(scripts):
 133     player_config = None
 134     for script in scripts:
 135         for line in script.split("\n"):
 136             s = "ytplayer.config = {"
 137             if s in line:
 138                 p1 = line.find(s) + len(s) - 1
 139                 p2 = line.find("};", p1) + 1
 140                 if p1 >= 0 and p2 > 0:
 141                     return json.loads(line[p1:p2])
 142
 143 def extract_js(script):
 144     PREFIX = "var _yt_player={};(function(g){var window=this;"
 145     SUFFIX = ";})(_yt_player);\n"
 146     assert script.startswith(PREFIX)
 147     assert script.endswith(SUFFIX)
 148
 149     return script[len(PREFIX):-len(SUFFIX)]
 150
 151 def find_func_name(script):
 152     FUNC_NAME = R"([a-zA-Z0-9$]+)"
 153     FUNC_PARAMS = R"(\([a-zA-Z,\.]+\.s\))"
 154     TERMINATOR = R"[,;\)]"
 155     PATTERN = FUNC_NAME + FUNC_PARAMS + TERMINATOR
 156
 157     match = re.search(PATTERN, script)
 158     func_name = match.groups()[0]
 159     return func_name
 160
 161 def decode_signature(js_url, signature):
 162     f = urlopen(js_url)
 163     script = f.read().decode("utf-8")
 164     f.close()
 165
 166     func_name = find_func_name(script)
 167
 168     params = {
 169         "func_name": func_name,
 170         "signature": json.dumps(signature),
 171         "code": json.dumps(extract_js(script)),
 172     }
 173     p = subprocess.Popen(
 174         "node",
 175         shell=True,
 176         close_fds=True,
 177         stdin=subprocess.PIPE,
 178         stdout=subprocess.PIPE
 179     )
 180     js_decode_script = ("""
 181         const vm = require('vm');
 182
 183         const sandbox = {
 184             location: {
 185                 hash: '',
 186                 href: '',
 187                 protocol: 'http:'
 188             },
 189             history: {
 190                 pushState: function(){}
 191             },
 192             document: {},
 193             navigator: {
 194                 userAgent: ''
 195             },
 196             XMLHttpRequest: class XMLHttpRequest {},
 197             matchMedia: () => ({matches: () => {}, media: ''}),
 198             signature: %(signature)s,
 199             transformed_signature: null,
 200             g: function(){} // this is _yt_player
 201         };
 202         sandbox.window = sandbox;
 203
 204         const code_string = %(code)s + ';';
 205         const exec_string = 'transformed_signature = %(func_name)s("", "MARKER", signature);';
 206         vm.runInNewContext(code_string + exec_string, sandbox);
 207
 208         function findSignature(obj) {
 209             if (typeof obj !== 'object') {
 210                 return;
 211             }
 212             for (const [key, value] of Object.entries(obj)) {
 213                 if (key === 'MARKER') {
 214                     return value;
 215                 }
 216                 const result = findSignature(value);
 217                 if (result) {
 218                     return result;
 219                 }
 220             }
 221         }
 222         console.log(findSignature(sandbox.transformed_signature));
 223     """ % params)
 224
 225     p.stdin.write(js_decode_script.encode("utf-8"))
 226     p.stdin.close()
 227
 228     transformed_signature = p.stdout.read().decode("utf-8").strip()
 229     transformed_signature = urllib.parse.unquote(transformed_signature)
 230     if p.wait() != 0:
 231         raise Exception("js failed to execute: %d" % p.returncode)
 232
 233     return transformed_signature
 234
 235 def get_best_video(player_config):
 236     url_data_list = player_config["args"]["url_encoded_fmt_stream_map"].split(",")
 237     js_url = player_config["assets"]["js"]
 238
 239     best_url = None
 240     best_quality = None
 241     best_extension = None
 242     for url_data in url_data_list:
 243         url_data = urllib.parse.parse_qs(url_data)
 244         mimetype = url_data["type"][0].split(";")[0]
 245         quality = url_data["quality"][0]
 246
 247         if "stereo3d" in url_data:
 248             continue
 249         if quality not in QUALITIES:
 250             continue
 251         if mimetype not in MIMETYPES:
 252             continue
 253
 254         extension = MIMETYPES[mimetype]
 255         quality = QUALITIES.get(quality, -1)
 256
 257         if best_quality is not None and quality < best_quality:
 258             continue
 259
 260         video_url = url_data["url"][0]
 261         if "sig" in url_data:
 262             signature = url_data["sig"][0]
 263         elif "s" in url_data:
 264             signature = decode_signature(js_url, url_data["s"][0])
 265         else:
 266             signature = None
 267
 268         if signature:
 269             sp = url_data.get("sp", ["signature"])[0]
 270             video_url = append_to_qs(video_url, {sp: signature})
 271
 272         best_url = video_url
 273         best_quality = quality
 274         best_extension = extension
 275
 276     return best_url, best_extension
 277
 278 def sanitize_filename(filename):
 279     return (
 280         re.sub("\s+", " ", filename.strip())
 281         .replace("\\", "-")
 282         .replace("/", "-")
 283         .replace("\0", " ")
 284     )
 285
 286 def get_video_url(page):
 287     player_config = get_player_config(page.scripts)
 288     if not player_config:
 289         raise VideoUnavailable(page.unavailable_message or "Could not find video URL")
 290
 291     video_url, extension = get_best_video(player_config)
 292     if not video_url:
 293         return None, None
 294
 295     filename = sanitize_filename(player_config["args"]["title"])
 296     filename += "." + extension
 297
 298     return video_url, filename
 299
 300 class YouTubeVideoPageParser(html.parser.HTMLParser):
 301     def __init__(self):
 302         super().__init__()
 303         self.unavailable_message = None
 304         self.scripts = []
 305
 306     def handle_starttag(self, tag, attrs):
 307         attrs = dict(attrs)
 308         self._handle_unavailable_message(tag, attrs)
 309         self._handle_script(tag, attrs)
 310
 311     def handle_endtag(self, tag):
 312         self.handle_data = self._ignore_data
 313
 314     def _ignore_data(self, _):
 315         pass
 316
 317     def _handle_unavailable_message(self, tag, attrs):
 318         if attrs.get("id", None) == "unavailable-message":
 319             self.handle_data = self._handle_unavailable_message_data
 320
 321     def _handle_unavailable_message_data(self, data):
 322         self.unavailable_message = data.strip()
 323
 324     def _handle_script(self, tag, attrs):
 325         if tag == "script":
 326             self.handle_data = self._handle_script_data
 327
 328     def _handle_script_data(self, data):
 329         if data:
 330             self.scripts.append(data)
 331
 332 def write_video(filename, video_data):
 333     quoted_filename = urllib.parse.quote(filename.encode("utf-8"))
 334     sys.stdout.buffer.write(
 335         b"Content-Disposition: attachment; filename*=UTF-8''{0}\r\n"
 336         .replace(b"{0}", quoted_filename.encode("utf-8"))
 337     )
 338     sys.stdout.buffer.write(
 339         b"Content-Length: {0}\r\n"
 340         .replace(b"{0}", video_data.getheader("Content-Length").encode("utf-8"))
 341     )
 342     sys.stdout.buffer.write(b"\r\n")
 343     shutil.copyfileobj(video_data, sys.stdout.buffer)
 344     video_data.close()
 345
 346 def cgimain():
 347     args = cgi.parse()
 348     try:
 349         url = args["url"][0]
 350     except:
 351         print_form(url="https://www.youtube.com/watch?v=FOOBAR")
 352         return
 353
 354     try:
 355         page = YouTubeVideoPageParser()
 356         validate_url(url)
 357         parse_url(url, page)
 358         video_url, filename = get_video_url(page)
 359         video_data = urlopen(video_url)
 360     except VideoUnavailable as e:
 361         print_form(
 362             url=url,
 363             msg="<p class='error'>Sorry, there was an error: %s</p>" % cgi.escape(e.args[0])
 364         )
 365     except NotYouTube:
 366         print_form(
 367             url=url,
 368             msg="<p class='error'>Sorry, that does not look like a YouTube page!</p>"
 369         )
 370     except Exception as e:
 371         print_form(
 372             url=url,
 373             msg="<p class='error'>Sorry, there was an unknown error.</p>"
 374         )
 375         return
 376
 377     write_video(filename, video_data)
 378
 379 def pp_size(size):
 380     suffixes = ["", "KiB", "MiB", "GiB"]
 381     for i, suffix in enumerate(suffixes):
 382         if size < 1024:
 383             break
 384         size /= 1024
 385     return "%.2f %s" % (size, suffix)
 386
 387 def copy_with_progress(content_length, infile, outfile):
 388     def print_status():
 389         rate = 0
 390         if now != last_ts:
 391             rate = last_bytes_read / (now - last_ts)
 392         sys.stdout.write("\33[2K\r")
 393         sys.stdout.write("%s / %s (%s/sec)" % (
 394             pp_size(bytes_read),
 395             pp_size(content_length),
 396             pp_size(rate),
 397         ))
 398         sys.stdout.flush()
 399
 400     last_ts = 0
 401     last_bytes_read = 0
 402     bytes_read = 0
 403     while True:
 404         now = time.time()
 405         if now - last_ts > 0.5:
 406             print_status()
 407             last_ts = now
 408             last_bytes_read = 0
 409
 410         buf = infile.read(32768)
 411         if not buf:
 412             break
 413         outfile.write(buf)
 414         last_bytes_read += len(buf)
 415         bytes_read += len(buf)
 416
 417     # Newline at the end
 418     print_status()
 419     print()
 420
 421 def main():
 422     try:
 423         url = sys.argv[1]
 424     except:
 425         print("Usage: %s https://youtube.com/watch?v=FOOBAR" % sys.argv[0], file=sys.stderr)
 426         sys.exit(1)
 427
 428     page = YouTubeVideoPageParser()
 429     parse_url(url, page)
 430     video_url, filename = get_video_url(page)
 431     print("Downloading", filename)
 432
 433     outfile = open(filename, "ab")
 434     offset = outfile.tell()
 435     if offset > 0:
 436         print("Resuming download from", pp_size(offset))
 437     total_size = None
 438
 439     while True:
 440         try:
 441             video_data = urlopen(video_url, offset)
 442         except urllib.error.HTTPError as e:
 443             if e.code == 416:
 444                 print("File is complete!")
 445                 break
 446             else:
 447                 raise
 448
 449         content_length = int(video_data.getheader("Content-Length"))
 450         if total_size is None:
 451             total_size = content_length
 452
 453         try:
 454             copy_with_progress(content_length, video_data, outfile)
 455         except IOError as e:
 456             print()
 457
 458         video_data.close()
 459         if outfile.tell() != total_size:
 460             old_offset = offset
 461             offset = outfile.tell()
 462             if old_offset == offset:
 463                 time.sleep(1)
 464             print("Restarting download from", pp_size(offset))
 465         else:
 466             break
 467
 468     outfile.close()
 469
 470
 471 if __name__ == "__main__":
 472     if "SCRIPT_NAME" in os.environ:
 473         cgimain()
 474     else:
 475         try:
 476             main()
 477         except KeyboardInterrupt:
 478             print("\nExiting...")
 479             sys.exit(1)
 480