projects / yt-dlp.git / blame
| Commit | Line | Data |
|---|---|---|
| c5e8d7af | 1 | # coding: utf-8 |
| c5e8d7af | 2 | |
| e0df6211 | 3 | import collections |
| edf3e38e | 4 | import errno |
| e0df6211 | 5 | import io |
| 0ca96d48 | 6 | import itertools |
| c5e8d7af | 7 | import json |
| c4417ddb | 8 | import os.path |
| c5e8d7af PH |
9 | import re |
| 10 | import socket | |
| e0df6211 PH |
11 | import string |
| 12 | import struct | |
| 13 | import traceback | |
| 055e6f36 | 14 | import xml.etree.ElementTree |
| e0df6211 | 15 | import zlib |
| c5e8d7af | 16 | |
| b05654f0 | 17 | from .common import InfoExtractor, SearchInfoExtractor |
| 54d39d8b | 18 | from .subtitles import SubtitlesInfoExtractor |
| c5e8d7af | 19 | from ..utils import ( |
| edf3e38e | 20 | compat_chr, |
| c5e8d7af PH |
21 | compat_http_client, |
| 22 | compat_parse_qs, | |
| 23 | compat_urllib_error, | |
| 24 | compat_urllib_parse, | |
| 25 | compat_urllib_request, | |
| 7c61bd36 | 26 | compat_urlparse, |
| c5e8d7af PH |
27 | compat_str, |
| 28 | ||
| 29 | clean_html, | |
| c38b1e77 | 30 | get_cachedir, |
| c5e8d7af PH |
31 | get_element_by_id, |
| 32 | ExtractorError, | |
| 33 | unescapeHTML, | |
| 34 | unified_strdate, | |
| 04cc9617 | 35 | orderedSet, |
| edf3e38e | 36 | write_json_file, |
| c5e8d7af PH |
37 | ) |
| 38 | ||
| de7f3446 | 39 | class YoutubeBaseInfoExtractor(InfoExtractor): |
| b2e8bc1b JMF |
40 | """Provide base functions for Youtube extractors""" |
| 41 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
| 42 | _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' | |
| 43 | _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' | |
| 44 | _NETRC_MACHINE = 'youtube' | |
| 45 | # If True it will raise an error if no login info is provided | |
| 46 | _LOGIN_REQUIRED = False | |
| 47 | ||
| 48 | def report_lang(self): | |
| 49 | """Report attempt to set language.""" | |
| 50 | self.to_screen(u'Setting language') | |
| 51 | ||
| 52 | def _set_language(self): | |
| 53 | request = compat_urllib_request.Request(self._LANG_URL) | |
| 54 | try: | |
| 55 | self.report_lang() | |
| 56 | compat_urllib_request.urlopen(request).read() | |
| 57 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
| 58 | self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) | |
| 59 | return False | |
| 60 | return True | |
| 61 | ||
| 62 | def _login(self): | |
| 63 | (username, password) = self._get_login_info() | |
| 64 | # No authentication to be performed | |
| 65 | if username is None: | |
| 66 | if self._LOGIN_REQUIRED: | |
| 67 | raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) | |
| 68 | return False | |
| 69 | ||
| 70 | request = compat_urllib_request.Request(self._LOGIN_URL) | |
| 71 | try: | |
| 72 | login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
| 73 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
| 74 | self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) | |
| 75 | return False | |
| 76 | ||
| 795f28f8 PH |
77 | galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"', |
| 78 | login_page, u'Login GALX parameter') | |
| c5e8d7af | 79 | |
| b2e8bc1b JMF |
80 | # Log in |
| 81 | login_form_strs = { | |
| 82 | u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
| 83 | u'Email': username, | |
| 84 | u'GALX': galx, | |
| 85 | u'Passwd': password, | |
| 86 | u'PersistentCookie': u'yes', | |
| 87 | u'_utf8': u'霱', | |
| 88 | u'bgresponse': u'js_disabled', | |
| 89 | u'checkConnection': u'', | |
| 90 | u'checkedDomains': u'youtube', | |
| 91 | u'dnConn': u'', | |
| b2e8bc1b JMF |
92 | u'pstMsg': u'0', |
| 93 | u'rmShown': u'1', | |
| 94 | u'secTok': u'', | |
| 95 | u'signIn': u'Sign in', | |
| 96 | u'timeStmp': u'', | |
| 97 | u'service': u'youtube', | |
| 98 | u'uilel': u'3', | |
| 99 | u'hl': u'en_US', | |
| 100 | } | |
| 101 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
| 102 | # chokes on unicode | |
| 103 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) | |
| 104 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
| 105 | request = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
| 106 | try: | |
| 107 | self.report_login() | |
| 108 | login_results = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
| 109 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
| 110 | self._downloader.report_warning(u'unable to log in: bad username or password') | |
| 111 | return False | |
| 112 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
| 113 | self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) | |
| 114 | return False | |
| 115 | return True | |
| 116 | ||
| 117 | def _confirm_age(self): | |
| 118 | age_form = { | |
| 119 | 'next_url': '/', | |
| 120 | 'action_confirm': 'Confirm', | |
| 121 | } | |
| 122 | request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) | |
| 123 | try: | |
| 124 | self.report_age_confirmation() | |
| 125 | compat_urllib_request.urlopen(request).read().decode('utf-8') | |
| 126 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
| 127 | raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) | |
| 128 | return True | |
| 129 | ||
| 130 | def _real_initialize(self): | |
| 131 | if self._downloader is None: | |
| 132 | return | |
| 133 | if not self._set_language(): | |
| 134 | return | |
| 135 | if not self._login(): | |
| 136 | return | |
| 137 | self._confirm_age() | |
| c5e8d7af | 138 | |
| 8377574c | 139 | |
| de7f3446 | 140 | class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): |
| 0f818663 | 141 | IE_DESC = u'YouTube.com' |
| c5e8d7af PH |
142 | _VALID_URL = r"""^ |
| 143 | ( | |
| 144 | (?:https?://)? # http(s):// (optional) | |
| f4b05232 | 145 | (?:(?:(?:(?:\w+\.)?youtube(?:-nocookie)?\.com/| |
| e69ae5b9 JMF |
146 | tube\.majestyc\.net/| |
| 147 | youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains | |
| c5e8d7af PH |
148 | (?:.*?\#/)? # handle anchor (#/) redirect urls |
| 149 | (?: # the various things that can precede the ID: | |
| 150 | (?:(?:v|embed|e)/) # v/ or embed/ or e/ | |
| 151 | |(?: # or the v= param in all its forms | |
| d741e55a | 152 | (?:(?:watch|movie)(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) |
| c5e8d7af PH |
153 | (?:\?|\#!?) # the params delimiter ? or # or #! |
| 154 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
| 155 | v= | |
| 156 | ) | |
| f4b05232 JMF |
157 | )) |
| 158 | |youtu\.be/ # just youtu.be/xxxx | |
| 159 | ) | |
| c5e8d7af | 160 | )? # all until now is optional -> you can pass the naked ID |
| 8963d9c2 | 161 | ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID |
| c5e8d7af PH |
162 | (?(1).+)? # if we found the ID, everything can follow |
| 163 | $""" | |
| c5e8d7af | 164 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' |
| c5e8d7af | 165 | # Listed in order of quality |
| bdc6b3fc | 166 | _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '36', '17', '13', |
| 96fb5605 | 167 | # Apple HTTP Live Streaming |
| bdc6b3fc | 168 | '96', '95', '94', '93', '92', '132', '151', |
| 939fbd26 JMF |
169 | # 3D |
| 170 | '85', '84', '102', '83', '101', '82', '100', | |
| 171 | # Dash video | |
| 172 | '138', '137', '248', '136', '247', '135', '246', | |
| 173 | '245', '244', '134', '243', '133', '242', '160', | |
| 174 | # Dash audio | |
| 175 | '141', '172', '140', '171', '139', | |
| 1d043b93 | 176 | ] |
| bdc6b3fc | 177 | _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '36', '17', '13', |
| 96fb5605 | 178 | # Apple HTTP Live Streaming |
| bdc6b3fc AZ |
179 | '96', '95', '94', '93', '92', '132', '151', |
| 180 | # 3D | |
| 86fe61c8 | 181 | '85', '102', '84', '101', '83', '100', '82', |
| 939fbd26 JMF |
182 | # Dash video |
| 183 | '138', '248', '137', '247', '136', '246', '245', | |
| 184 | '244', '135', '243', '134', '242', '133', '160', | |
| 185 | # Dash audio | |
| 186 | '172', '141', '171', '140', '139', | |
| 1d043b93 | 187 | ] |
| bdc6b3fc AZ |
188 | _video_formats_map = { |
| 189 | 'flv': ['35', '34', '6', '5'], | |
| 190 | '3gp': ['36', '17', '13'], | |
| 191 | 'mp4': ['38', '37', '22', '18'], | |
| 192 | 'webm': ['46', '45', '44', '43'], | |
| 193 | } | |
| c5e8d7af PH |
194 | _video_extensions = { |
| 195 | '13': '3gp', | |
| bdc6b3fc | 196 | '17': '3gp', |
| c5e8d7af PH |
197 | '18': 'mp4', |
| 198 | '22': 'mp4', | |
| bdc6b3fc | 199 | '36': '3gp', |
| c5e8d7af | 200 | '37': 'mp4', |
| d69cf69a | 201 | '38': 'mp4', |
| c5e8d7af PH |
202 | '43': 'webm', |
| 203 | '44': 'webm', | |
| 204 | '45': 'webm', | |
| 205 | '46': 'webm', | |
| 1d043b93 | 206 | |
| 86fe61c8 AZ |
207 | # 3d videos |
| 208 | '82': 'mp4', | |
| 209 | '83': 'mp4', | |
| 210 | '84': 'mp4', | |
| 211 | '85': 'mp4', | |
| 212 | '100': 'webm', | |
| 213 | '101': 'webm', | |
| 214 | '102': 'webm', | |
| 836a086c | 215 | |
| 96fb5605 | 216 | # Apple HTTP Live Streaming |
| 1d043b93 JMF |
217 | '92': 'mp4', |
| 218 | '93': 'mp4', | |
| 219 | '94': 'mp4', | |
| 220 | '95': 'mp4', | |
| 221 | '96': 'mp4', | |
| 222 | '132': 'mp4', | |
| 223 | '151': 'mp4', | |
| 836a086c AZ |
224 | |
| 225 | # Dash mp4 | |
| 226 | '133': 'mp4', | |
| 227 | '134': 'mp4', | |
| 228 | '135': 'mp4', | |
| 229 | '136': 'mp4', | |
| 230 | '137': 'mp4', | |
| 231 | '138': 'mp4', | |
| 836a086c AZ |
232 | '160': 'mp4', |
| 233 | ||
| f6f1fc92 RB |
234 | # Dash mp4 audio |
| 235 | '139': 'm4a', | |
| 16f36a6f RB |
236 | '140': 'm4a', |
| 237 | '141': 'm4a', | |
| 836a086c AZ |
238 | |
| 239 | # Dash webm | |
| 240 | '171': 'webm', | |
| 241 | '172': 'webm', | |
| 242 | '242': 'webm', | |
| 243 | '243': 'webm', | |
| 244 | '244': 'webm', | |
| 245 | '245': 'webm', | |
| 246 | '246': 'webm', | |
| 247 | '247': 'webm', | |
| 248 | '248': 'webm', | |
| c5e8d7af PH |
249 | } |
| 250 | _video_dimensions = { | |
| 251 | '5': '240x400', | |
| 252 | '6': '???', | |
| 253 | '13': '???', | |
| 254 | '17': '144x176', | |
| 255 | '18': '360x640', | |
| 256 | '22': '720x1280', | |
| 257 | '34': '360x640', | |
| 258 | '35': '480x854', | |
| bdc6b3fc | 259 | '36': '240x320', |
| c5e8d7af PH |
260 | '37': '1080x1920', |
| 261 | '38': '3072x4096', | |
| 262 | '43': '360x640', | |
| 263 | '44': '480x854', | |
| 264 | '45': '720x1280', | |
| 265 | '46': '1080x1920', | |
| 86fe61c8 AZ |
266 | '82': '360p', |
| 267 | '83': '480p', | |
| 268 | '84': '720p', | |
| 269 | '85': '1080p', | |
| 1d043b93 JMF |
270 | '92': '240p', |
| 271 | '93': '360p', | |
| 272 | '94': '480p', | |
| 273 | '95': '720p', | |
| 274 | '96': '1080p', | |
| 86fe61c8 AZ |
275 | '100': '360p', |
| 276 | '101': '480p', | |
| 836a086c | 277 | '102': '720p', |
| 1d043b93 JMF |
278 | '132': '240p', |
| 279 | '151': '72p', | |
| 836a086c AZ |
280 | '133': '240p', |
| 281 | '134': '360p', | |
| 282 | '135': '480p', | |
| 283 | '136': '720p', | |
| 284 | '137': '1080p', | |
| 285 | '138': '>1080p', | |
| 286 | '139': '48k', | |
| 287 | '140': '128k', | |
| 288 | '141': '256k', | |
| 289 | '160': '192p', | |
| 290 | '171': '128k', | |
| 291 | '172': '256k', | |
| 292 | '242': '240p', | |
| 293 | '243': '360p', | |
| 294 | '244': '480p', | |
| 295 | '245': '480p', | |
| 296 | '246': '480p', | |
| 297 | '247': '720p', | |
| 298 | '248': '1080p', | |
| c5e8d7af | 299 | } |
| 836a086c AZ |
300 | _special_itags = { |
| 301 | '82': '3D', | |
| 302 | '83': '3D', | |
| 303 | '84': '3D', | |
| 304 | '85': '3D', | |
| 305 | '100': '3D', | |
| 306 | '101': '3D', | |
| 307 | '102': '3D', | |
| 308 | '133': 'DASH Video', | |
| 309 | '134': 'DASH Video', | |
| 310 | '135': 'DASH Video', | |
| 311 | '136': 'DASH Video', | |
| 312 | '137': 'DASH Video', | |
| 313 | '138': 'DASH Video', | |
| 314 | '139': 'DASH Audio', | |
| 315 | '140': 'DASH Audio', | |
| 316 | '141': 'DASH Audio', | |
| 317 | '160': 'DASH Video', | |
| 318 | '171': 'DASH Audio', | |
| 319 | '172': 'DASH Audio', | |
| 320 | '242': 'DASH Video', | |
| 321 | '243': 'DASH Video', | |
| 322 | '244': 'DASH Video', | |
| 323 | '245': 'DASH Video', | |
| 324 | '246': 'DASH Video', | |
| 325 | '247': 'DASH Video', | |
| 326 | '248': 'DASH Video', | |
| c5e8d7af | 327 | } |
| 836a086c | 328 | |
| c5e8d7af | 329 | IE_NAME = u'youtube' |
| 2eb88d95 PH |
330 | _TESTS = [ |
| 331 | { | |
| 0e853ca4 PH |
332 | u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", |
| 333 | u"file": u"BaW_jenozKc.mp4", | |
| 334 | u"info_dict": { | |
| 335 | u"title": u"youtube-dl test video \"'/\\ä↭𝕐", | |
| 336 | u"uploader": u"Philipp Hagemeister", | |
| 337 | u"uploader_id": u"phihag", | |
| 338 | u"upload_date": u"20121002", | |
| 339 | u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ." | |
| 2eb88d95 | 340 | } |
| 0e853ca4 PH |
341 | }, |
| 342 | { | |
| 343 | u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U", | |
| ea32fbac | 344 | u"file": u"1ltcDfZMA3U.mp4", |
| 0e853ca4 PH |
345 | u"note": u"Test VEVO video (#897)", |
| 346 | u"info_dict": { | |
| 347 | u"upload_date": u"20070518", | |
| 348 | u"title": u"Maps - It Will Find You", | |
| 349 | u"description": u"Music video by Maps performing It Will Find You.", | |
| 350 | u"uploader": u"MuteUSA", | |
| 351 | u"uploader_id": u"MuteUSA" | |
| 2eb88d95 | 352 | } |
| 0e853ca4 PH |
353 | }, |
| 354 | { | |
| 355 | u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", | |
| 356 | u"file": u"UxxajLWwzqY.mp4", | |
| 357 | u"note": u"Test generic use_cipher_signature video (#897)", | |
| 358 | u"info_dict": { | |
| 359 | u"upload_date": u"20120506", | |
| 360 | u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", | |
| 2dc59299 | 361 | u"description": u"md5:5b292926389560516e384ac437c0ec07", |
| 45ed795c | 362 | u"uploader": u"Icona Pop", |
| 0e853ca4 | 363 | u"uploader_id": u"IconaPop" |
| 2eb88d95 | 364 | } |
| c108eb73 JMF |
365 | }, |
| 366 | { | |
| 367 | u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", | |
| 368 | u"file": u"07FYdnEawAQ.mp4", | |
| 369 | u"note": u"Test VEVO video with age protection (#956)", | |
| 370 | u"info_dict": { | |
| 371 | u"upload_date": u"20130703", | |
| 372 | u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", | |
| 373 | u"description": u"md5:64249768eec3bc4276236606ea996373", | |
| 374 | u"uploader": u"justintimberlakeVEVO", | |
| 375 | u"uploader_id": u"justintimberlakeVEVO" | |
| 376 | } | |
| 377 | }, | |
| 2eb88d95 PH |
378 | ] |
| 379 | ||
| c5e8d7af PH |
380 | |
| 381 | @classmethod | |
| 382 | def suitable(cls, url): | |
| 383 | """Receives a URL and returns True if suitable for this IE.""" | |
| e3ea4790 | 384 | if YoutubePlaylistIE.suitable(url): return False |
| c5e8d7af PH |
385 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None |
| 386 | ||
| e0df6211 PH |
387 | def __init__(self, *args, **kwargs): |
| 388 | super(YoutubeIE, self).__init__(*args, **kwargs) | |
| 83799698 | 389 | self._player_cache = {} |
| e0df6211 | 390 | |
| c5e8d7af PH |
391 | def report_video_webpage_download(self, video_id): |
| 392 | """Report attempt to download video webpage.""" | |
| 393 | self.to_screen(u'%s: Downloading video webpage' % video_id) | |
| 394 | ||
| 395 | def report_video_info_webpage_download(self, video_id): | |
| 396 | """Report attempt to download video info webpage.""" | |
| 397 | self.to_screen(u'%s: Downloading video info webpage' % video_id) | |
| 398 | ||
| c5e8d7af PH |
399 | def report_information_extraction(self, video_id): |
| 400 | """Report attempt to extract video information.""" | |
| 401 | self.to_screen(u'%s: Extracting video information' % video_id) | |
| 402 | ||
| 403 | def report_unavailable_format(self, video_id, format): | |
| 404 | """Report extracted video URL.""" | |
| 405 | self.to_screen(u'%s: Format %s not available' % (video_id, format)) | |
| 406 | ||
| 407 | def report_rtmp_download(self): | |
| 408 | """Indicate the download will use the RTMP protocol.""" | |
| 409 | self.to_screen(u'RTMP download detected') | |
| 410 | ||
| c4417ddb PH |
411 | def _extract_signature_function(self, video_id, player_url, slen): |
| 412 | id_m = re.match(r'.*-(?P<id>[a-zA-Z0-9_-]+)\.(?P<ext>[a-z]+)$', | |
| 83799698 | 413 | player_url) |
| e0df6211 PH |
414 | player_type = id_m.group('ext') |
| 415 | player_id = id_m.group('id') | |
| 416 | ||
| c4417ddb PH |
417 | # Read from filesystem cache |
| 418 | func_id = '%s_%s_%d' % (player_type, player_id, slen) | |
| 419 | assert os.path.basename(func_id) == func_id | |
| c38b1e77 | 420 | cache_dir = get_cachedir(self._downloader.params) |
| c4417ddb | 421 | |
| c3c88a26 | 422 | cache_enabled = cache_dir is not None |
| f8061589 | 423 | if cache_enabled: |
| c4417ddb PH |
424 | cache_fn = os.path.join(os.path.expanduser(cache_dir), |
| 425 | u'youtube-sigfuncs', | |
| 426 | func_id + '.json') | |
| 427 | try: | |
| edf3e38e | 428 | with io.open(cache_fn, 'r', encoding='utf-8') as cachef: |
| c4417ddb PH |
429 | cache_spec = json.load(cachef) |
| 430 | return lambda s: u''.join(s[i] for i in cache_spec) | |
| edf3e38e | 431 | except IOError: |
| c4417ddb | 432 | pass # No cache available |
| 83799698 | 433 | |
| e0df6211 PH |
434 | if player_type == 'js': |
| 435 | code = self._download_webpage( | |
| 436 | player_url, video_id, | |
| 83799698 | 437 | note=u'Downloading %s player %s' % (player_type, player_id), |
| e0df6211 | 438 | errnote=u'Download of %s failed' % player_url) |
| 83799698 | 439 | res = self._parse_sig_js(code) |
| c4417ddb | 440 | elif player_type == 'swf': |
| e0df6211 PH |
441 | urlh = self._request_webpage( |
| 442 | player_url, video_id, | |
| 83799698 | 443 | note=u'Downloading %s player %s' % (player_type, player_id), |
| e0df6211 PH |
444 | errnote=u'Download of %s failed' % player_url) |
| 445 | code = urlh.read() | |
| 83799698 | 446 | res = self._parse_sig_swf(code) |
| e0df6211 PH |
447 | else: |
| 448 | assert False, 'Invalid player type %r' % player_type | |
| 449 | ||
| f8061589 | 450 | if cache_enabled: |
| edf3e38e | 451 | try: |
| c705320f PH |
452 | test_string = u''.join(map(compat_chr, range(slen))) |
| 453 | cache_res = res(test_string) | |
| edf3e38e PH |
454 | cache_spec = [ord(c) for c in cache_res] |
| 455 | try: | |
| 456 | os.makedirs(os.path.dirname(cache_fn)) | |
| 457 | except OSError as ose: | |
| 458 | if ose.errno != errno.EEXIST: | |
| 459 | raise | |
| 460 | write_json_file(cache_spec, cache_fn) | |
| 0ca96d48 | 461 | except Exception: |
| edf3e38e PH |
462 | tb = traceback.format_exc() |
| 463 | self._downloader.report_warning( | |
| 464 | u'Writing cache to %r failed: %s' % (cache_fn, tb)) | |
| 83799698 PH |
465 | |
| 466 | return res | |
| 467 | ||
| edf3e38e PH |
468 | def _print_sig_code(self, func, slen): |
| 469 | def gen_sig_code(idxs): | |
| 470 | def _genslice(start, end, step): | |
| 471 | starts = u'' if start == 0 else str(start) | |
| e35e4ddc PH |
472 | ends = (u':%d' % (end+step)) if end + step >= 0 else u':' |
| 473 | steps = u'' if step == 1 else (u':%d' % step) | |
| edf3e38e PH |
474 | return u's[%s%s%s]' % (starts, ends, steps) |
| 475 | ||
| 476 | step = None | |
| 0ca96d48 PH |
477 | start = '(Never used)' # Quelch pyflakes warnings - start will be |
| 478 | # set as soon as step is set | |
| edf3e38e PH |
479 | for i, prev in zip(idxs[1:], idxs[:-1]): |
| 480 | if step is not None: | |
| 481 | if i - prev == step: | |
| 482 | continue | |
| 483 | yield _genslice(start, prev, step) | |
| 484 | step = None | |
| 485 | continue | |
| 486 | if i - prev in [-1, 1]: | |
| 487 | step = i - prev | |
| 488 | start = prev | |
| 489 | continue | |
| 490 | else: | |
| 491 | yield u's[%d]' % prev | |
| 492 | if step is None: | |
| 493 | yield u's[%d]' % i | |
| 494 | else: | |
| 495 | yield _genslice(start, i, step) | |
| 496 | ||
| c705320f PH |
497 | test_string = u''.join(map(compat_chr, range(slen))) |
| 498 | cache_res = func(test_string) | |
| edf3e38e PH |
499 | cache_spec = [ord(c) for c in cache_res] |
| 500 | expr_code = u' + '.join(gen_sig_code(cache_spec)) | |
| 501 | code = u'if len(s) == %d:\n return %s\n' % (slen, expr_code) | |
| f8061589 | 502 | self.to_screen(u'Extracted signature function:\n' + code) |
| edf3e38e | 503 | |
| e0df6211 PH |
504 | def _parse_sig_js(self, jscode): |
| 505 | funcname = self._search_regex( | |
| 506 | r'signature=([a-zA-Z]+)', jscode, | |
| 507 | u'Initial JS player signature function name') | |
| 508 | ||
| 509 | functions = {} | |
| 510 | ||
| 511 | def argidx(varname): | |
| 512 | return string.lowercase.index(varname) | |
| 513 | ||
| 514 | def interpret_statement(stmt, local_vars, allow_recursion=20): | |
| 515 | if allow_recursion < 0: | |
| 0ca96d48 | 516 | raise ExtractorError(u'Recursion limit reached') |
| e0df6211 PH |
517 | |
| 518 | if stmt.startswith(u'var '): | |
| 519 | stmt = stmt[len(u'var '):] | |
| 520 | ass_m = re.match(r'^(?P<out>[a-z]+)(?:\[(?P<index>[^\]]+)\])?' + | |
| 521 | r'=(?P<expr>.*)$', stmt) | |
| 522 | if ass_m: | |
| 523 | if ass_m.groupdict().get('index'): | |
| 524 | def assign(val): | |
| 525 | lvar = local_vars[ass_m.group('out')] | |
| 526 | idx = interpret_expression(ass_m.group('index'), | |
| 527 | local_vars, allow_recursion) | |
| 528 | assert isinstance(idx, int) | |
| 529 | lvar[idx] = val | |
| 530 | return val | |
| 531 | expr = ass_m.group('expr') | |
| 532 | else: | |
| 533 | def assign(val): | |
| 534 | local_vars[ass_m.group('out')] = val | |
| 535 | return val | |
| 536 | expr = ass_m.group('expr') | |
| 537 | elif stmt.startswith(u'return '): | |
| 538 | assign = lambda v: v | |
| 539 | expr = stmt[len(u'return '):] | |
| 540 | else: | |
| 541 | raise ExtractorError( | |
| 542 | u'Cannot determine left side of statement in %r' % stmt) | |
| 543 | ||
| 544 | v = interpret_expression(expr, local_vars, allow_recursion) | |
| 545 | return assign(v) | |
| 546 | ||
| 547 | def interpret_expression(expr, local_vars, allow_recursion): | |
| 548 | if expr.isdigit(): | |
| 549 | return int(expr) | |
| 550 | ||
| 551 | if expr.isalpha(): | |
| 552 | return local_vars[expr] | |
| 553 | ||
| 554 | m = re.match(r'^(?P<in>[a-z]+)\.(?P<member>.*)$', expr) | |
| 555 | if m: | |
| 556 | member = m.group('member') | |
| 557 | val = local_vars[m.group('in')] | |
| 558 | if member == 'split("")': | |
| 559 | return list(val) | |
| 560 | if member == 'join("")': | |
| 561 | return u''.join(val) | |
| 562 | if member == 'length': | |
| 563 | return len(val) | |
| 564 | if member == 'reverse()': | |
| 565 | return val[::-1] | |
| 566 | slice_m = re.match(r'slice\((?P<idx>.*)\)', member) | |
| 567 | if slice_m: | |
| 568 | idx = interpret_expression( | |
| 569 | slice_m.group('idx'), local_vars, allow_recursion-1) | |
| 570 | return val[idx:] | |
| 571 | ||
| 572 | m = re.match( | |
| 573 | r'^(?P<in>[a-z]+)\[(?P<idx>.+)\]$', expr) | |
| 574 | if m: | |
| 575 | val = local_vars[m.group('in')] | |
| 576 | idx = interpret_expression(m.group('idx'), local_vars, | |
| 577 | allow_recursion-1) | |
| 578 | return val[idx] | |
| 579 | ||
| 580 | m = re.match(r'^(?P<a>.+?)(?P<op>[%])(?P<b>.+?)$', expr) | |
| 581 | if m: | |
| 582 | a = interpret_expression(m.group('a'), | |
| 583 | local_vars, allow_recursion) | |
| 584 | b = interpret_expression(m.group('b'), | |
| 585 | local_vars, allow_recursion) | |
| 586 | return a % b | |
| 587 | ||
| 588 | m = re.match( | |
| 589 | r'^(?P<func>[a-zA-Z]+)\((?P<args>[a-z0-9,]+)\)$', expr) | |
| 590 | if m: | |
| 591 | fname = m.group('func') | |
| 592 | if fname not in functions: | |
| 593 | functions[fname] = extract_function(fname) | |
| 594 | argvals = [int(v) if v.isdigit() else local_vars[v] | |
| 595 | for v in m.group('args').split(',')] | |
| 596 | return functions[fname](argvals) | |
| 597 | raise ExtractorError(u'Unsupported JS expression %r' % expr) | |
| 598 | ||
| 599 | def extract_function(funcname): | |
| 600 | func_m = re.search( | |
| 601 | r'function ' + re.escape(funcname) + | |
| 602 | r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}', | |
| 603 | jscode) | |
| 604 | argnames = func_m.group('args').split(',') | |
| 605 | ||
| 606 | def resf(args): | |
| 607 | local_vars = dict(zip(argnames, args)) | |
| 608 | for stmt in func_m.group('code').split(';'): | |
| 609 | res = interpret_statement(stmt, local_vars) | |
| 610 | return res | |
| 611 | return resf | |
| 612 | ||
| 613 | initial_function = extract_function(funcname) | |
| 614 | return lambda s: initial_function([s]) | |
| 615 | ||
| 616 | def _parse_sig_swf(self, file_contents): | |
| 617 | if file_contents[1:3] != b'WS': | |
| 618 | raise ExtractorError( | |
| 619 | u'Not an SWF file; header is %r' % file_contents[:3]) | |
| 620 | if file_contents[:1] == b'C': | |
| 621 | content = zlib.decompress(file_contents[8:]) | |
| 622 | else: | |
| 623 | raise NotImplementedError(u'Unsupported compression format %r' % | |
| 624 | file_contents[:1]) | |
| 625 | ||
| 626 | def extract_tags(content): | |
| 627 | pos = 0 | |
| 628 | while pos < len(content): | |
| 629 | header16 = struct.unpack('<H', content[pos:pos+2])[0] | |
| 630 | pos += 2 | |
| 631 | tag_code = header16 >> 6 | |
| 632 | tag_len = header16 & 0x3f | |
| 633 | if tag_len == 0x3f: | |
| 634 | tag_len = struct.unpack('<I', content[pos:pos+4])[0] | |
| 635 | pos += 4 | |
| 636 | assert pos+tag_len <= len(content) | |
| 637 | yield (tag_code, content[pos:pos+tag_len]) | |
| 638 | pos += tag_len | |
| 639 | ||
| 640 | code_tag = next(tag | |
| 641 | for tag_code, tag in extract_tags(content) | |
| 642 | if tag_code == 82) | |
| 643 | p = code_tag.index(b'\0', 4) + 1 | |
| ba552f54 | 644 | code_reader = io.BytesIO(code_tag[p:]) |
| e0df6211 PH |
645 | |
| 646 | # Parse ABC (AVM2 ByteCode) | |
| ba552f54 PH |
647 | def read_int(reader=None): |
| 648 | if reader is None: | |
| 649 | reader = code_reader | |
| e0df6211 PH |
650 | res = 0 |
| 651 | shift = 0 | |
| 652 | for _ in range(5): | |
| ba552f54 PH |
653 | buf = reader.read(1) |
| 654 | assert len(buf) == 1 | |
| 655 | b = struct.unpack('<B', buf)[0] | |
| e0df6211 PH |
656 | res = res | ((b & 0x7f) << shift) |
| 657 | if b & 0x80 == 0: | |
| 658 | break | |
| 659 | shift += 7 | |
| ba552f54 PH |
660 | return res |
| 661 | ||
| 662 | def u30(reader=None): | |
| 663 | res = read_int(reader) | |
| 664 | assert res & 0xf0000000 == 0 | |
| e0df6211 PH |
665 | return res |
| 666 | u32 = read_int | |
| 667 | ||
| ba552f54 PH |
668 | def s32(reader=None): |
| 669 | v = read_int(reader) | |
| e0df6211 PH |
670 | if v & 0x80000000 != 0: |
| 671 | v = - ((v ^ 0xffffffff) + 1) | |
| ba552f54 PH |
672 | return v |
| 673 | ||
| 0ca96d48 | 674 | def read_string(reader=None): |
| ba552f54 PH |
675 | if reader is None: |
| 676 | reader = code_reader | |
| 677 | slen = u30(reader) | |
| 678 | resb = reader.read(slen) | |
| 679 | assert len(resb) == slen | |
| 680 | return resb.decode('utf-8') | |
| 681 | ||
| 682 | def read_bytes(count, reader=None): | |
| 683 | if reader is None: | |
| 684 | reader = code_reader | |
| 685 | resb = reader.read(count) | |
| 686 | assert len(resb) == count | |
| 687 | return resb | |
| 688 | ||
| 689 | def read_byte(reader=None): | |
| 690 | resb = read_bytes(1, reader=reader) | |
| 691 | res = struct.unpack('<B', resb)[0] | |
| 692 | return res | |
| e0df6211 PH |
693 | |
| 694 | # minor_version + major_version | |
| 0ca96d48 | 695 | read_bytes(2 + 2) |
| e0df6211 PH |
696 | |
| 697 | # Constant pool | |
| ba552f54 | 698 | int_count = u30() |
| e0df6211 | 699 | for _c in range(1, int_count): |
| 0ca96d48 | 700 | s32() |
| ba552f54 | 701 | uint_count = u30() |
| e0df6211 | 702 | for _c in range(1, uint_count): |
| 0ca96d48 | 703 | u32() |
| ba552f54 | 704 | double_count = u30() |
| 0ca96d48 | 705 | read_bytes((double_count-1) * 8) |
| ba552f54 | 706 | string_count = u30() |
| e0df6211 PH |
707 | constant_strings = [u''] |
| 708 | for _c in range(1, string_count): | |
| 0ca96d48 | 709 | s = read_string() |
| e0df6211 | 710 | constant_strings.append(s) |
| ba552f54 | 711 | namespace_count = u30() |
| e0df6211 | 712 | for _c in range(1, namespace_count): |
| 0ca96d48 PH |
713 | read_bytes(1) # kind |
| 714 | u30() # name | |
| ba552f54 | 715 | ns_set_count = u30() |
| e0df6211 | 716 | for _c in range(1, ns_set_count): |
| ba552f54 | 717 | count = u30() |
| e0df6211 | 718 | for _c2 in range(count): |
| 0ca96d48 | 719 | u30() |
| ba552f54 | 720 | multiname_count = u30() |
| e0df6211 PH |
721 | MULTINAME_SIZES = { |
| 722 | 0x07: 2, # QName | |
| 723 | 0x0d: 2, # QNameA | |
| 724 | 0x0f: 1, # RTQName | |
| 725 | 0x10: 1, # RTQNameA | |
| 726 | 0x11: 0, # RTQNameL | |
| 727 | 0x12: 0, # RTQNameLA | |
| 728 | 0x09: 2, # Multiname | |
| 729 | 0x0e: 2, # MultinameA | |
| 730 | 0x1b: 1, # MultinameL | |
| 731 | 0x1c: 1, # MultinameLA | |
| 732 | } | |
| 733 | multinames = [u''] | |
| 734 | for _c in range(1, multiname_count): | |
| ba552f54 | 735 | kind = u30() |
| e0df6211 PH |
736 | assert kind in MULTINAME_SIZES, u'Invalid multiname kind %r' % kind |
| 737 | if kind == 0x07: | |
| 0ca96d48 | 738 | u30() # namespace_idx |
| ba552f54 | 739 | name_idx = u30() |
| e0df6211 PH |
740 | multinames.append(constant_strings[name_idx]) |
| 741 | else: | |
| 742 | multinames.append('[MULTINAME kind: %d]' % kind) | |
| 743 | for _c2 in range(MULTINAME_SIZES[kind]): | |
| 0ca96d48 | 744 | u30() |
| e0df6211 PH |
745 | |
| 746 | # Methods | |
| ba552f54 | 747 | method_count = u30() |
| e0df6211 PH |
748 | MethodInfo = collections.namedtuple( |
| 749 | 'MethodInfo', | |
| 750 | ['NEED_ARGUMENTS', 'NEED_REST']) | |
| 751 | method_infos = [] | |
| 752 | for method_id in range(method_count): | |
| ba552f54 | 753 | param_count = u30() |
| 0ca96d48 | 754 | u30() # return type |
| e0df6211 | 755 | for _ in range(param_count): |
| 0ca96d48 PH |
756 | u30() # param type |
| 757 | u30() # name index (always 0 for youtube) | |
| ba552f54 | 758 | flags = read_byte() |
| e0df6211 PH |
759 | if flags & 0x08 != 0: |
| 760 | # Options present | |
| ba552f54 | 761 | option_count = u30() |
| e0df6211 | 762 | for c in range(option_count): |
| 0ca96d48 PH |
763 | u30() # val |
| 764 | read_bytes(1) # kind | |
| e0df6211 PH |
765 | if flags & 0x80 != 0: |
| 766 | # Param names present | |
| 767 | for _ in range(param_count): | |
| 0ca96d48 | 768 | u30() # param name |
| e0df6211 PH |
769 | mi = MethodInfo(flags & 0x01 != 0, flags & 0x04 != 0) |
| 770 | method_infos.append(mi) | |
| 771 | ||
| 772 | # Metadata | |
| ba552f54 | 773 | metadata_count = u30() |
| e0df6211 | 774 | for _c in range(metadata_count): |
| 0ca96d48 | 775 | u30() # name |
| ba552f54 | 776 | item_count = u30() |
| e0df6211 | 777 | for _c2 in range(item_count): |
| 0ca96d48 PH |
778 | u30() # key |
| 779 | u30() # value | |
| ba552f54 PH |
780 | |
| 781 | def parse_traits_info(): | |
| 782 | trait_name_idx = u30() | |
| 783 | kind_full = read_byte() | |
| e0df6211 PH |
784 | kind = kind_full & 0x0f |
| 785 | attrs = kind_full >> 4 | |
| 786 | methods = {} | |
| 787 | if kind in [0x00, 0x06]: # Slot or Const | |
| 0ca96d48 PH |
788 | u30() # Slot id |
| 789 | u30() # type_name_idx | |
| ba552f54 | 790 | vindex = u30() |
| e0df6211 | 791 | if vindex != 0: |
| 0ca96d48 | 792 | read_byte() # vkind |
| e0df6211 | 793 | elif kind in [0x01, 0x02, 0x03]: # Method / Getter / Setter |
| 0ca96d48 | 794 | u30() # disp_id |
| ba552f54 | 795 | method_idx = u30() |
| e0df6211 PH |
796 | methods[multinames[trait_name_idx]] = method_idx |
| 797 | elif kind == 0x04: # Class | |
| 0ca96d48 PH |
798 | u30() # slot_id |
| 799 | u30() # classi | |
| e0df6211 | 800 | elif kind == 0x05: # Function |
| 0ca96d48 | 801 | u30() # slot_id |
| ba552f54 | 802 | function_idx = u30() |
| e0df6211 PH |
803 | methods[function_idx] = multinames[trait_name_idx] |
| 804 | else: | |
| 805 | raise ExtractorError(u'Unsupported trait kind %d' % kind) | |
| 806 | ||
| 807 | if attrs & 0x4 != 0: # Metadata present | |
| ba552f54 | 808 | metadata_count = u30() |
| e0df6211 | 809 | for _c3 in range(metadata_count): |
| 0ca96d48 | 810 | u30() # metadata index |
| e0df6211 | 811 | |
| ba552f54 | 812 | return methods |
| e0df6211 PH |
813 | |
| 814 | # Classes | |
| 815 | TARGET_CLASSNAME = u'SignatureDecipher' | |
| 816 | searched_idx = multinames.index(TARGET_CLASSNAME) | |
| 817 | searched_class_id = None | |
| ba552f54 | 818 | class_count = u30() |
| e0df6211 | 819 | for class_id in range(class_count): |
| ba552f54 | 820 | name_idx = u30() |
| e0df6211 PH |
821 | if name_idx == searched_idx: |
| 822 | # We found the class we're looking for! | |
| 823 | searched_class_id = class_id | |
| 0ca96d48 | 824 | u30() # super_name idx |
| ba552f54 | 825 | flags = read_byte() |
| e0df6211 | 826 | if flags & 0x08 != 0: # Protected namespace is present |
| 0ca96d48 | 827 | u30() # protected_ns_idx |
| ba552f54 | 828 | intrf_count = u30() |
| e0df6211 | 829 | for _c2 in range(intrf_count): |
| 0ca96d48 PH |
830 | u30() |
| 831 | u30() # iinit | |
| ba552f54 | 832 | trait_count = u30() |
| e0df6211 | 833 | for _c2 in range(trait_count): |
| 0ca96d48 | 834 | parse_traits_info() |
| e0df6211 PH |
835 | |
| 836 | if searched_class_id is None: | |
| 837 | raise ExtractorError(u'Target class %r not found' % | |
| 838 | TARGET_CLASSNAME) | |
| 839 | ||
| 840 | method_names = {} | |
| 841 | method_idxs = {} | |
| 842 | for class_id in range(class_count): | |
| 0ca96d48 | 843 | u30() # cinit |
| ba552f54 | 844 | trait_count = u30() |
| e0df6211 | 845 | for _c2 in range(trait_count): |
| ba552f54 | 846 | trait_methods = parse_traits_info() |
| e0df6211 PH |
847 | if class_id == searched_class_id: |
| 848 | method_names.update(trait_methods.items()) | |
| 849 | method_idxs.update(dict( | |
| 850 | (idx, name) | |
| 851 | for name, idx in trait_methods.items())) | |
| 852 | ||
| 853 | # Scripts | |
| ba552f54 | 854 | script_count = u30() |
| e0df6211 | 855 | for _c in range(script_count): |
| 0ca96d48 | 856 | u30() # init |
| ba552f54 | 857 | trait_count = u30() |
| e0df6211 | 858 | for _c2 in range(trait_count): |
| 0ca96d48 | 859 | parse_traits_info() |
| e0df6211 PH |
860 | |
| 861 | # Method bodies | |
| ba552f54 | 862 | method_body_count = u30() |
| e0df6211 PH |
863 | Method = collections.namedtuple('Method', ['code', 'local_count']) |
| 864 | methods = {} | |
| 865 | for _c in range(method_body_count): | |
| ba552f54 | 866 | method_idx = u30() |
| 0ca96d48 | 867 | u30() # max_stack |
| ba552f54 | 868 | local_count = u30() |
| 0ca96d48 PH |
869 | u30() # init_scope_depth |
| 870 | u30() # max_scope_depth | |
| ba552f54 PH |
871 | code_length = u30() |
| 872 | code = read_bytes(code_length) | |
| e0df6211 | 873 | if method_idx in method_idxs: |
| ba552f54 | 874 | m = Method(code, local_count) |
| e0df6211 | 875 | methods[method_idxs[method_idx]] = m |
| ba552f54 | 876 | exception_count = u30() |
| e0df6211 | 877 | for _c2 in range(exception_count): |
| 0ca96d48 PH |
878 | u30() # from |
| 879 | u30() # to | |
| 880 | u30() # target | |
| 881 | u30() # exc_type | |
| 882 | u30() # var_name | |
| ba552f54 | 883 | trait_count = u30() |
| e0df6211 | 884 | for _c2 in range(trait_count): |
| 0ca96d48 | 885 | parse_traits_info() |
| e0df6211 | 886 | |
| ba552f54 | 887 | assert p + code_reader.tell() == len(code_tag) |
| e0df6211 PH |
888 | assert len(methods) == len(method_idxs) |
| 889 | ||
| 890 | method_pyfunctions = {} | |
| 891 | ||
| 892 | def extract_function(func_name): | |
| 893 | if func_name in method_pyfunctions: | |
| 894 | return method_pyfunctions[func_name] | |
| 895 | if func_name not in methods: | |
| 896 | raise ExtractorError(u'Cannot find function %r' % func_name) | |
| 897 | m = methods[func_name] | |
| 898 | ||
| 899 | def resfunc(args): | |
| e0df6211 PH |
900 | registers = ['(this)'] + list(args) + [None] * m.local_count |
| 901 | stack = [] | |
| 902 | coder = io.BytesIO(m.code) | |
| 903 | while True: | |
| 904 | opcode = struct.unpack('!B', coder.read(1))[0] | |
| a7177865 | 905 | if opcode == 36: # pushbyte |
| e0df6211 PH |
906 | v = struct.unpack('!B', coder.read(1))[0] |
| 907 | stack.append(v) | |
| 908 | elif opcode == 44: # pushstring | |
| 909 | idx = u30(coder) | |
| 910 | stack.append(constant_strings[idx]) | |
| 911 | elif opcode == 48: # pushscope | |
| 912 | # We don't implement the scope register, so we'll just | |
| 913 | # ignore the popped value | |
| 914 | stack.pop() | |
| 915 | elif opcode == 70: # callproperty | |
| 916 | index = u30(coder) | |
| 917 | mname = multinames[index] | |
| 918 | arg_count = u30(coder) | |
| 919 | args = list(reversed( | |
| 920 | [stack.pop() for _ in range(arg_count)])) | |
| 921 | obj = stack.pop() | |
| 922 | if mname == u'split': | |
| 923 | assert len(args) == 1 | |
| 924 | assert isinstance(args[0], compat_str) | |
| 925 | assert isinstance(obj, compat_str) | |
| 926 | if args[0] == u'': | |
| 927 | res = list(obj) | |
| 928 | else: | |
| 929 | res = obj.split(args[0]) | |
| 930 | stack.append(res) | |
| a7177865 PH |
931 | elif mname == u'slice': |
| 932 | assert len(args) == 1 | |
| 933 | assert isinstance(args[0], int) | |
| 934 | assert isinstance(obj, list) | |
| 935 | res = obj[args[0]:] | |
| 936 | stack.append(res) | |
| 937 | elif mname == u'join': | |
| 938 | assert len(args) == 1 | |
| 939 | assert isinstance(args[0], compat_str) | |
| 940 | assert isinstance(obj, list) | |
| 941 | res = args[0].join(obj) | |
| 942 | stack.append(res) | |
| e0df6211 PH |
943 | elif mname in method_pyfunctions: |
| 944 | stack.append(method_pyfunctions[mname](args)) | |
| 945 | else: | |
| 946 | raise NotImplementedError( | |
| 947 | u'Unsupported property %r on %r' | |
| 948 | % (mname, obj)) | |
| a7177865 PH |
949 | elif opcode == 72: # returnvalue |
| 950 | res = stack.pop() | |
| 951 | return res | |
| 952 | elif opcode == 79: # callpropvoid | |
| 953 | index = u30(coder) | |
| 954 | mname = multinames[index] | |
| 955 | arg_count = u30(coder) | |
| 956 | args = list(reversed( | |
| 957 | [stack.pop() for _ in range(arg_count)])) | |
| 958 | obj = stack.pop() | |
| 959 | if mname == u'reverse': | |
| 960 | assert isinstance(obj, list) | |
| 961 | obj.reverse() | |
| 962 | else: | |
| 963 | raise NotImplementedError( | |
| 964 | u'Unsupported (void) property %r on %r' | |
| 965 | % (mname, obj)) | |
| e0df6211 PH |
966 | elif opcode == 93: # findpropstrict |
| 967 | index = u30(coder) | |
| 968 | mname = multinames[index] | |
| 969 | res = extract_function(mname) | |
| 970 | stack.append(res) | |
| 971 | elif opcode == 97: # setproperty | |
| 972 | index = u30(coder) | |
| 973 | value = stack.pop() | |
| 974 | idx = stack.pop() | |
| 975 | obj = stack.pop() | |
| 976 | assert isinstance(obj, list) | |
| 977 | assert isinstance(idx, int) | |
| 978 | obj[idx] = value | |
| 979 | elif opcode == 98: # getlocal | |
| 980 | index = u30(coder) | |
| 981 | stack.append(registers[index]) | |
| 982 | elif opcode == 99: # setlocal | |
| 983 | index = u30(coder) | |
| 984 | value = stack.pop() | |
| 985 | registers[index] = value | |
| 986 | elif opcode == 102: # getproperty | |
| 987 | index = u30(coder) | |
| 988 | pname = multinames[index] | |
| 989 | if pname == u'length': | |
| 990 | obj = stack.pop() | |
| 991 | assert isinstance(obj, list) | |
| 992 | stack.append(len(obj)) | |
| 993 | else: # Assume attribute access | |
| 994 | idx = stack.pop() | |
| 995 | assert isinstance(idx, int) | |
| 996 | obj = stack.pop() | |
| 997 | assert isinstance(obj, list) | |
| 998 | stack.append(obj[idx]) | |
| 999 | elif opcode == 128: # coerce | |
| 0ca96d48 | 1000 | u30(coder) |
| e0df6211 PH |
1001 | elif opcode == 133: # coerce_s |
| 1002 | assert isinstance(stack[-1], (type(None), compat_str)) | |
| 1003 | elif opcode == 164: # modulo | |
| 1004 | value2 = stack.pop() | |
| 1005 | value1 = stack.pop() | |
| 1006 | res = value1 % value2 | |
| 1007 | stack.append(res) | |
| a7177865 PH |
1008 | elif opcode == 208: # getlocal_0 |
| 1009 | stack.append(registers[0]) | |
| 1010 | elif opcode == 209: # getlocal_1 | |
| 1011 | stack.append(registers[1]) | |
| 1012 | elif opcode == 210: # getlocal_2 | |
| 1013 | stack.append(registers[2]) | |
| 1014 | elif opcode == 211: # getlocal_3 | |
| 1015 | stack.append(registers[3]) | |
| e0df6211 PH |
1016 | elif opcode == 214: # setlocal_2 |
| 1017 | registers[2] = stack.pop() | |
| 1018 | elif opcode == 215: # setlocal_3 | |
| 1019 | registers[3] = stack.pop() | |
| 1020 | else: | |
| 1021 | raise NotImplementedError( | |
| 1022 | u'Unsupported opcode %d' % opcode) | |
| 1023 | ||
| 1024 | method_pyfunctions[func_name] = resfunc | |
| 1025 | return resfunc | |
| 1026 | ||
| 1027 | initial_function = extract_function(u'decipher') | |
| 1028 | return lambda s: initial_function([s]) | |
| 1029 | ||
| 83799698 | 1030 | def _decrypt_signature(self, s, video_id, player_url, age_gate=False): |
| 257a2501 | 1031 | """Turn the encrypted s field into a working signature""" |
| 6b37f0be | 1032 | |
| 83799698 | 1033 | if player_url is not None: |
| e0df6211 | 1034 | try: |
| 7f8ae73a PH |
1035 | player_id = (player_url, len(s)) |
| 1036 | if player_id not in self._player_cache: | |
| 83799698 | 1037 | func = self._extract_signature_function( |
| c4417ddb | 1038 | video_id, player_url, len(s) |
| e0df6211 | 1039 | ) |
| 7f8ae73a PH |
1040 | self._player_cache[player_id] = func |
| 1041 | func = self._player_cache[player_id] | |
| edf3e38e PH |
1042 | if self._downloader.params.get('youtube_print_sig_code'): |
| 1043 | self._print_sig_code(func, len(s)) | |
| 1044 | return func(s) | |
| 0ca96d48 | 1045 | except Exception: |
| e0df6211 | 1046 | tb = traceback.format_exc() |
| 83799698 PH |
1047 | self._downloader.report_warning( |
| 1048 | u'Automatic signature extraction failed: ' + tb) | |
| e0df6211 | 1049 | |
| d2d8f895 PH |
1050 | self._downloader.report_warning( |
| 1051 | u'Warning: Falling back to static signature algorithm') | |
| 920de7a2 | 1052 | |
| 2f2ffea9 PH |
1053 | return self._static_decrypt_signature( |
| 1054 | s, video_id, player_url, age_gate) | |
| e0df6211 | 1055 | |
| 2f2ffea9 | 1056 | def _static_decrypt_signature(self, s, video_id, player_url, age_gate): |
| e0df6211 PH |
1057 | if age_gate: |
| 1058 | # The videos with age protection use another player, so the | |
| 1059 | # algorithms can be different. | |
| 1060 | if len(s) == 86: | |
| 1061 | return s[2:63] + s[82] + s[64:82] + s[63] | |
| 1062 | ||
| bc4b9008 | 1063 | if len(s) == 93: |
| 1064 | return s[86:29:-1] + s[88] + s[28:5:-1] | |
| 1065 | elif len(s) == 92: | |
| 444b1165 | 1066 | return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83] |
| 38d025b3 JMF |
1067 | elif len(s) == 91: |
| 1068 | return s[84:27:-1] + s[86] + s[26:5:-1] | |
| 444b1165 JMF |
1069 | elif len(s) == 90: |
| 1070 | return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81] | |
| 8a9d86a2 | 1071 | elif len(s) == 89: |
| 1072 | return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1] | |
| 444b1165 | 1073 | elif len(s) == 88: |
| 3e223834 | 1074 | return s[7:28] + s[87] + s[29:45] + s[55] + s[46:55] + s[2] + s[56:87] + s[28] |
| be547e1d | 1075 | elif len(s) == 87: |
| 3a725669 | 1076 | return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:] |
| be547e1d | 1077 | elif len(s) == 86: |
| f2c327fd | 1078 | return s[80:72:-1] + s[16] + s[71:39:-1] + s[72] + s[38:16:-1] + s[82] + s[15::-1] |
| be547e1d | 1079 | elif len(s) == 85: |
| 6ae8ee3f | 1080 | return s[3:11] + s[0] + s[12:55] + s[84] + s[56:84] |
| be547e1d | 1081 | elif len(s) == 84: |
| 6f56389b | 1082 | return s[78:70:-1] + s[14] + s[69:37:-1] + s[70] + s[36:14:-1] + s[80] + s[:14][::-1] |
| be547e1d | 1083 | elif len(s) == 83: |
| 920de7a2 | 1084 | return s[80:63:-1] + s[0] + s[62:0:-1] + s[63] |
| be547e1d | 1085 | elif len(s) == 82: |
| c21315f2 | 1086 | return s[80:37:-1] + s[7] + s[36:7:-1] + s[0] + s[6:0:-1] + s[37] |
| be547e1d | 1087 | elif len(s) == 81: |
| aedd6bb9 | 1088 | return s[56] + s[79:56:-1] + s[41] + s[55:41:-1] + s[80] + s[40:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9] |
| 066090dd JMF |
1089 | elif len(s) == 80: |
| 1090 | return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80] | |
| 5c468ca8 JMF |
1091 | elif len(s) == 79: |
| 1092 | return s[54] + s[77:54:-1] + s[39] + s[53:39:-1] + s[78] + s[38:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9] | |
| be547e1d PH |
1093 | |
| 1094 | else: | |
| 1095 | raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) | |
| c5e8d7af | 1096 | |
| de7f3446 | 1097 | def _get_available_subtitles(self, video_id): |
| de7f3446 | 1098 | try: |
| 7fad1c63 JMF |
1099 | sub_list = self._download_webpage( |
| 1100 | 'http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, | |
| 1101 | video_id, note=False) | |
| 1102 | except ExtractorError as err: | |
| de7f3446 JMF |
1103 | self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) |
| 1104 | return {} | |
| 1105 | lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
| 1106 | ||
| 1107 | sub_lang_list = {} | |
| 1108 | for l in lang_list: | |
| 1109 | lang = l[1] | |
| 1110 | params = compat_urllib_parse.urlencode({ | |
| 1111 | 'lang': lang, | |
| 1112 | 'v': video_id, | |
| 1113 | 'fmt': self._downloader.params.get('subtitlesformat'), | |
| 8eddf3e9 | 1114 | 'name': l[0].encode('utf-8'), |
| de7f3446 JMF |
1115 | }) |
| 1116 | url = u'http://www.youtube.com/api/timedtext?' + params | |
| 1117 | sub_lang_list[lang] = url | |
| 1118 | if not sub_lang_list: | |
| 1119 | self._downloader.report_warning(u'video doesn\'t have subtitles') | |
| 1120 | return {} | |
| 1121 | return sub_lang_list | |
| 1122 | ||
| 055e6f36 | 1123 | def _get_available_automatic_caption(self, video_id, webpage): |
| de7f3446 JMF |
1124 | """We need the webpage for getting the captions url, pass it as an |
| 1125 | argument to speed up the process.""" | |
| de7f3446 JMF |
1126 | sub_format = self._downloader.params.get('subtitlesformat') |
| 1127 | self.to_screen(u'%s: Looking for automatic captions' % video_id) | |
| 1128 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
| 055e6f36 | 1129 | err_msg = u'Couldn\'t find automatic captions for %s' % video_id |
| de7f3446 JMF |
1130 | if mobj is None: |
| 1131 | self._downloader.report_warning(err_msg) | |
| 1132 | return {} | |
| 1133 | player_config = json.loads(mobj.group(1)) | |
| 1134 | try: | |
| 1135 | args = player_config[u'args'] | |
| 1136 | caption_url = args[u'ttsurl'] | |
| 1137 | timestamp = args[u'timestamp'] | |
| 055e6f36 JMF |
1138 | # We get the available subtitles |
| 1139 | list_params = compat_urllib_parse.urlencode({ | |
| 1140 | 'type': 'list', | |
| 1141 | 'tlangs': 1, | |
| 1142 | 'asrs': 1, | |
| de7f3446 | 1143 | }) |
| 055e6f36 JMF |
1144 | list_url = caption_url + '&' + list_params |
| 1145 | list_page = self._download_webpage(list_url, video_id) | |
| 1146 | caption_list = xml.etree.ElementTree.fromstring(list_page.encode('utf-8')) | |
| e3dc22ca | 1147 | original_lang_node = caption_list.find('track') |
| f6a54188 | 1148 | if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' : |
| e3dc22ca JMF |
1149 | self._downloader.report_warning(u'Video doesn\'t have automatic captions') |
| 1150 | return {} | |
| 1151 | original_lang = original_lang_node.attrib['lang_code'] | |
| 055e6f36 JMF |
1152 | |
| 1153 | sub_lang_list = {} | |
| 1154 | for lang_node in caption_list.findall('target'): | |
| 1155 | sub_lang = lang_node.attrib['lang_code'] | |
| 1156 | params = compat_urllib_parse.urlencode({ | |
| 1157 | 'lang': original_lang, | |
| 1158 | 'tlang': sub_lang, | |
| 1159 | 'fmt': sub_format, | |
| 1160 | 'ts': timestamp, | |
| 1161 | 'kind': 'asr', | |
| 1162 | }) | |
| 1163 | sub_lang_list[sub_lang] = caption_url + '&' + params | |
| 1164 | return sub_lang_list | |
| de7f3446 JMF |
1165 | # An extractor error can be raise by the download process if there are |
| 1166 | # no automatic captions but there are subtitles | |
| 1167 | except (KeyError, ExtractorError): | |
| 1168 | self._downloader.report_warning(err_msg) | |
| 1169 | return {} | |
| 1170 | ||
| c5e8d7af PH |
1171 | def _print_formats(self, formats): |
| 1172 | print('Available formats:') | |
| 1173 | for x in formats: | |
| 03cc7c20 JMF |
1174 | print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'), |
| 1175 | self._video_dimensions.get(x, '???'), | |
| 836a086c | 1176 | ' ('+self._special_itags[x]+')' if x in self._special_itags else '')) |
| c5e8d7af PH |
1177 | |
| 1178 | def _extract_id(self, url): | |
| 1179 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
| 1180 | if mobj is None: | |
| 1181 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 1182 | video_id = mobj.group(2) | |
| 1183 | return video_id | |
| 1184 | ||
| 1d043b93 JMF |
1185 | def _get_video_url_list(self, url_map): |
| 1186 | """ | |
| 1187 | Transform a dictionary in the format {itag:url} to a list of (itag, url) | |
| 1188 | with the requested formats. | |
| 1189 | """ | |
| 1190 | req_format = self._downloader.params.get('format', None) | |
| 1191 | format_limit = self._downloader.params.get('format_limit', None) | |
| 1192 | available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats | |
| 1193 | if format_limit is not None and format_limit in available_formats: | |
| 1194 | format_list = available_formats[available_formats.index(format_limit):] | |
| 1195 | else: | |
| 1196 | format_list = available_formats | |
| 1197 | existing_formats = [x for x in format_list if x in url_map] | |
| 1198 | if len(existing_formats) == 0: | |
| 1199 | raise ExtractorError(u'no known formats available for video') | |
| 1200 | if self._downloader.params.get('listformats', None): | |
| 1201 | self._print_formats(existing_formats) | |
| 1202 | return | |
| 1203 | if req_format is None or req_format == 'best': | |
| 1204 | video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality | |
| 1205 | elif req_format == 'worst': | |
| 1206 | video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality | |
| 1207 | elif req_format in ('-1', 'all'): | |
| 1208 | video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats | |
| 1209 | else: | |
| 1210 | # Specific formats. We pick the first in a slash-delimeted sequence. | |
| bdc6b3fc AZ |
1211 | # Format can be specified as itag or 'mp4' or 'flv' etc. We pick the highest quality |
| 1212 | # available in the specified format. For example, | |
| 1213 | # if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. | |
| 1214 | # if '1/mp4/3/4' is requested and '1' and '5' (is a mp4) are available, we pick '1'. | |
| 1215 | # if '1/mp4/3/4' is requested and '4' and '5' (is a mp4) are available, we pick '5'. | |
| 1d043b93 JMF |
1216 | req_formats = req_format.split('/') |
| 1217 | video_url_list = None | |
| 1218 | for rf in req_formats: | |
| 1219 | if rf in url_map: | |
| 1220 | video_url_list = [(rf, url_map[rf])] | |
| 1221 | break | |
| bdc6b3fc AZ |
1222 | if rf in self._video_formats_map: |
| 1223 | for srf in self._video_formats_map[rf]: | |
| 1224 | if srf in url_map: | |
| 1225 | video_url_list = [(srf, url_map[srf])] | |
| 1226 | break | |
| 1227 | else: | |
| 1228 | continue | |
| 1229 | break | |
| 1d043b93 JMF |
1230 | if video_url_list is None: |
| 1231 | raise ExtractorError(u'requested format not available') | |
| 1232 | return video_url_list | |
| 1233 | ||
| 1234 | def _extract_from_m3u8(self, manifest_url, video_id): | |
| 1235 | url_map = {} | |
| 1236 | def _get_urls(_manifest): | |
| 1237 | lines = _manifest.split('\n') | |
| 1238 | urls = filter(lambda l: l and not l.startswith('#'), | |
| 1239 | lines) | |
| 1240 | return urls | |
| 1241 | manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') | |
| 1242 | formats_urls = _get_urls(manifest) | |
| 1243 | for format_url in formats_urls: | |
| 890f62e8 | 1244 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') |
| 1d043b93 JMF |
1245 | url_map[itag] = format_url |
| 1246 | return url_map | |
| 1247 | ||
| 1fb07d10 JG |
1248 | def _extract_annotations(self, video_id): |
| 1249 | url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id | |
| 1250 | return self._download_webpage(url, video_id, note=u'Searching for annotations.', errnote=u'Unable to download video annotations.') | |
| 1251 | ||
| c5e8d7af PH |
1252 | def _real_extract(self, url): |
| 1253 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter | |
| 1254 | mobj = re.search(self._NEXT_URL_RE, url) | |
| 1255 | if mobj: | |
| 1256 | url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
| 1257 | video_id = self._extract_id(url) | |
| 1258 | ||
| 1259 | # Get video webpage | |
| 1260 | self.report_video_webpage_download(video_id) | |
| 1261 | url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id | |
| 1262 | request = compat_urllib_request.Request(url) | |
| 1263 | try: | |
| 1264 | video_webpage_bytes = compat_urllib_request.urlopen(request).read() | |
| 1265 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
| 1266 | raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err)) | |
| 1267 | ||
| 1268 | video_webpage = video_webpage_bytes.decode('utf-8', 'ignore') | |
| 1269 | ||
| 1270 | # Attempt to extract SWF player URL | |
| e0df6211 | 1271 | mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) |
| c5e8d7af PH |
1272 | if mobj is not None: |
| 1273 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
| 1274 | else: | |
| 1275 | player_url = None | |
| 1276 | ||
| 1277 | # Get video info | |
| 1278 | self.report_video_info_webpage_download(video_id) | |
| c108eb73 JMF |
1279 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
| 1280 | self.report_age_confirmation() | |
| 1281 | age_gate = True | |
| 1282 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
| 1283 | # this can be viewed without login into Youtube | |
| 1284 | data = compat_urllib_parse.urlencode({'video_id': video_id, | |
| 1285 | 'el': 'embedded', | |
| 1286 | 'gl': 'US', | |
| 1287 | 'hl': 'en', | |
| 1288 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
| 1289 | 'asv': 3, | |
| 1290 | 'sts':'1588', | |
| 1291 | }) | |
| 1292 | video_info_url = 'https://www.youtube.com/get_video_info?' + data | |
| c5e8d7af PH |
1293 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
| 1294 | note=False, | |
| 1295 | errnote='unable to download video info webpage') | |
| 1296 | video_info = compat_parse_qs(video_info_webpage) | |
| c108eb73 JMF |
1297 | else: |
| 1298 | age_gate = False | |
| 1299 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
| 1300 | video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
| 1301 | % (video_id, el_type)) | |
| 1302 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
| 1303 | note=False, | |
| 1304 | errnote='unable to download video info webpage') | |
| 1305 | video_info = compat_parse_qs(video_info_webpage) | |
| 1306 | if 'token' in video_info: | |
| 1307 | break | |
| c5e8d7af PH |
1308 | if 'token' not in video_info: |
| 1309 | if 'reason' in video_info: | |
| 9a82b238 | 1310 | raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True) |
| c5e8d7af PH |
1311 | else: |
| 1312 | raise ExtractorError(u'"token" parameter not in video info for unknown reason') | |
| 1313 | ||
| 1314 | # Check for "rental" videos | |
| 1315 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
| 1316 | raise ExtractorError(u'"rental" videos not supported') | |
| 1317 | ||
| 1318 | # Start extracting information | |
| 1319 | self.report_information_extraction(video_id) | |
| 1320 | ||
| 1321 | # uploader | |
| 1322 | if 'author' not in video_info: | |
| 1323 | raise ExtractorError(u'Unable to extract uploader name') | |
| 1324 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
| 1325 | ||
| 1326 | # uploader_id | |
| 1327 | video_uploader_id = None | |
| 1328 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
| 1329 | if mobj is not None: | |
| 1330 | video_uploader_id = mobj.group(1) | |
| 1331 | else: | |
| 1332 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
| 1333 | ||
| 1334 | # title | |
| a8c6b241 PH |
1335 | if 'title' in video_info: |
| 1336 | video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) | |
| 1337 | else: | |
| 1338 | self._downloader.report_warning(u'Unable to extract video title') | |
| 1339 | video_title = u'_' | |
| c5e8d7af PH |
1340 | |
| 1341 | # thumbnail image | |
| 7763b04e JMF |
1342 | # We try first to get a high quality image: |
| 1343 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
| 1344 | video_webpage, re.DOTALL) | |
| 1345 | if m_thumb is not None: | |
| 1346 | video_thumbnail = m_thumb.group(1) | |
| 1347 | elif 'thumbnail_url' not in video_info: | |
| c5e8d7af | 1348 | self._downloader.report_warning(u'unable to extract video thumbnail') |
| f490e77e | 1349 | video_thumbnail = None |
| c5e8d7af PH |
1350 | else: # don't panic if we can't find it |
| 1351 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
| 1352 | ||
| 1353 | # upload date | |
| 1354 | upload_date = None | |
| 1355 | mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL) | |
| 1356 | if mobj is not None: | |
| 1357 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
| 1358 | upload_date = unified_strdate(upload_date) | |
| 1359 | ||
| 1360 | # description | |
| 1361 | video_description = get_element_by_id("eow-description", video_webpage) | |
| 1362 | if video_description: | |
| 1363 | video_description = clean_html(video_description) | |
| 1364 | else: | |
| 1365 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
| 1366 | if fd_mobj: | |
| 1367 | video_description = unescapeHTML(fd_mobj.group(1)) | |
| 1368 | else: | |
| 1369 | video_description = u'' | |
| 1370 | ||
| 1371 | # subtitles | |
| d82134c3 | 1372 | video_subtitles = self.extract_subtitles(video_id, video_webpage) |
| c5e8d7af | 1373 | |
| c5e8d7af | 1374 | if self._downloader.params.get('listsubtitles', False): |
| d665f8d3 | 1375 | self._list_available_subtitles(video_id, video_webpage) |
| c5e8d7af PH |
1376 | return |
| 1377 | ||
| 1378 | if 'length_seconds' not in video_info: | |
| 1379 | self._downloader.report_warning(u'unable to extract video duration') | |
| 1380 | video_duration = '' | |
| 1381 | else: | |
| 1382 | video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]) | |
| 1383 | ||
| 1fb07d10 JG |
1384 | # annotations |
| 1385 | video_annotations = None | |
| 1386 | if self._downloader.params.get('writeannotations', False): | |
| 1387 | video_annotations = self._extract_annotations(video_id) | |
| 1388 | ||
| c5e8d7af | 1389 | # Decide which formats to download |
| c5e8d7af PH |
1390 | |
| 1391 | try: | |
| 1392 | mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) | |
| 50be92c1 PH |
1393 | if not mobj: |
| 1394 | raise ValueError('Could not find vevo ID') | |
| c5e8d7af PH |
1395 | info = json.loads(mobj.group(1)) |
| 1396 | args = info['args'] | |
| 7ce7e394 JMF |
1397 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
| 1398 | # this signatures are encrypted | |
| 44d46655 | 1399 | if 'url_encoded_fmt_stream_map' not in args: |
| f10503db | 1400 | raise ValueError(u'No stream_map present') # caught below |
| 00fe14fc JMF |
1401 | re_signature = re.compile(r'[&,]s=') |
| 1402 | m_s = re_signature.search(args['url_encoded_fmt_stream_map']) | |
| 7ce7e394 JMF |
1403 | if m_s is not None: |
| 1404 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
| c5e8d7af | 1405 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
| 00fe14fc | 1406 | m_s = re_signature.search(args.get('adaptive_fmts', u'')) |
| b7a68384 | 1407 | if m_s is not None: |
| 00fe14fc JMF |
1408 | if 'adaptive_fmts' in video_info: |
| 1409 | video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts'] | |
| 37b6d5f6 | 1410 | else: |
| 00fe14fc | 1411 | video_info['adaptive_fmts'] = [args['adaptive_fmts']] |
| c5e8d7af PH |
1412 | except ValueError: |
| 1413 | pass | |
| 1414 | ||
| 1415 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
| 1416 | self.report_rtmp_download() | |
| 1417 | video_url_list = [(None, video_info['conn'][0])] | |
| 00fe14fc JMF |
1418 | elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1: |
| 1419 | encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0] | |
| 1420 | if 'rtmpe%3Dyes' in encoded_url_map: | |
| a7055eb9 | 1421 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) |
| c5e8d7af | 1422 | url_map = {} |
| 00fe14fc | 1423 | for url_data_str in encoded_url_map.split(','): |
| c5e8d7af PH |
1424 | url_data = compat_parse_qs(url_data_str) |
| 1425 | if 'itag' in url_data and 'url' in url_data: | |
| 1426 | url = url_data['url'][0] | |
| 1427 | if 'sig' in url_data: | |
| 1428 | url += '&signature=' + url_data['sig'][0] | |
| 1429 | elif 's' in url_data: | |
| e0df6211 | 1430 | encrypted_sig = url_data['s'][0] |
| 769fda3c | 1431 | if self._downloader.params.get('verbose'): |
| c108eb73 | 1432 | if age_gate: |
| bdde940e PH |
1433 | if player_url is None: |
| 1434 | player_version = 'unknown' | |
| 1435 | else: | |
| 1436 | player_version = self._search_regex( | |
| 1437 | r'-(.+)\.swf$', player_url, | |
| 1438 | u'flash player', fatal=False) | |
| e0df6211 | 1439 | player_desc = 'flash player %s' % player_version |
| c108eb73 | 1440 | else: |
| 83799698 PH |
1441 | player_version = self._search_regex( |
| 1442 | r'html5player-(.+?)\.js', video_webpage, | |
| c108eb73 | 1443 | 'html5 player', fatal=False) |
| e0df6211 PH |
1444 | player_desc = u'html5 player %s' % player_version |
| 1445 | ||
| 1446 | parts_sizes = u'.'.join(compat_str(len(part)) for part in encrypted_sig.split('.')) | |
| 5a76c651 | 1447 | self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % |
| e0df6211 PH |
1448 | (len(encrypted_sig), parts_sizes, url_data['itag'][0], player_desc)) |
| 1449 | ||
| 83799698 | 1450 | if not age_gate: |
| e0df6211 PH |
1451 | jsplayer_url_json = self._search_regex( |
| 1452 | r'"assets":.+?"js":\s*("[^"]+")', | |
| 1453 | video_webpage, u'JS player URL') | |
| 83799698 | 1454 | player_url = json.loads(jsplayer_url_json) |
| e0df6211 | 1455 | |
| 83799698 PH |
1456 | signature = self._decrypt_signature( |
| 1457 | encrypted_sig, video_id, player_url, age_gate) | |
| c5e8d7af PH |
1458 | url += '&signature=' + signature |
| 1459 | if 'ratebypass' not in url: | |
| 1460 | url += '&ratebypass=yes' | |
| 1461 | url_map[url_data['itag'][0]] = url | |
| 1d043b93 JMF |
1462 | video_url_list = self._get_video_url_list(url_map) |
| 1463 | if not video_url_list: | |
| c5e8d7af | 1464 | return |
| 1d043b93 JMF |
1465 | elif video_info.get('hlsvp'): |
| 1466 | manifest_url = video_info['hlsvp'][0] | |
| 1467 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
| 1468 | video_url_list = self._get_video_url_list(url_map) | |
| 1469 | if not video_url_list: | |
| 1470 | return | |
| 1471 | ||
| c5e8d7af | 1472 | else: |
| 9abb3204 | 1473 | raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') |
| c5e8d7af PH |
1474 | |
| 1475 | results = [] | |
| 600cc1a4 | 1476 | for itag, video_real_url in video_url_list: |
| c5e8d7af | 1477 | # Extension |
| 600cc1a4 | 1478 | video_extension = self._video_extensions.get(itag, 'flv') |
| c5e8d7af | 1479 | |
| 600cc1a4 JMF |
1480 | video_format = '{0} - {1}{2}'.format(itag if itag else video_extension, |
| 1481 | self._video_dimensions.get(itag, '???'), | |
| 1482 | ' ('+self._special_itags[itag]+')' if itag in self._special_itags else '') | |
| c5e8d7af PH |
1483 | |
| 1484 | results.append({ | |
| 1485 | 'id': video_id, | |
| 1486 | 'url': video_real_url, | |
| 1487 | 'uploader': video_uploader, | |
| 1488 | 'uploader_id': video_uploader_id, | |
| 1489 | 'upload_date': upload_date, | |
| 1490 | 'title': video_title, | |
| 1491 | 'ext': video_extension, | |
| 1492 | 'format': video_format, | |
| 600cc1a4 | 1493 | 'format_id': itag, |
| c5e8d7af PH |
1494 | 'thumbnail': video_thumbnail, |
| 1495 | 'description': video_description, | |
| 1496 | 'player_url': player_url, | |
| 1497 | 'subtitles': video_subtitles, | |
| 8dbe9899 | 1498 | 'duration': video_duration, |
| cfadd183 | 1499 | 'age_limit': 18 if age_gate else 0, |
| 1fb07d10 | 1500 | 'annotations': video_annotations |
| c5e8d7af PH |
1501 | }) |
| 1502 | return results | |
| 1503 | ||
| 1504 | class YoutubePlaylistIE(InfoExtractor): | |
| 0f818663 | 1505 | IE_DESC = u'YouTube.com playlists' |
| c5e8d7af PH |
1506 | _VALID_URL = r"""(?: |
| 1507 | (?:https?://)? | |
| 1508 | (?:\w+\.)? | |
| 1509 | youtube\.com/ | |
| 1510 | (?: | |
| 1511 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
| 1512 | \? (?:.*?&)*? (?:p|a|list)= | |
| 1513 | | p/ | |
| 1514 | ) | |
| c626a3d9 | 1515 | ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,}) |
| c5e8d7af PH |
1516 | .* |
| 1517 | | | |
| c626a3d9 | 1518 | ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,}) |
| c5e8d7af PH |
1519 | )""" |
| 1520 | _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none' | |
| 1521 | _MAX_RESULTS = 50 | |
| 1522 | IE_NAME = u'youtube:playlist' | |
| 1523 | ||
| 1524 | @classmethod | |
| 1525 | def suitable(cls, url): | |
| 1526 | """Receives a URL and returns True if suitable for this IE.""" | |
| 1527 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
| 1528 | ||
| 1529 | def _real_extract(self, url): | |
| 1530 | # Extract playlist id | |
| 1531 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
| 1532 | if mobj is None: | |
| 1533 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 47192f92 FV |
1534 | playlist_id = mobj.group(1) or mobj.group(2) |
| 1535 | ||
| 1536 | # Check if it's a video-specific URL | |
| 7c61bd36 | 1537 | query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) |
| 47192f92 FV |
1538 | if 'v' in query_dict: |
| 1539 | video_id = query_dict['v'][0] | |
| 1540 | if self._downloader.params.get('noplaylist'): | |
| 1541 | self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id) | |
| 1542 | return self.url_result('https://www.youtube.com/watch?v=' + video_id, 'Youtube') | |
| 1543 | else: | |
| 1544 | self.to_screen(u'Downloading playlist PL%s - add --no-playlist to just download video %s' % (playlist_id, video_id)) | |
| c5e8d7af PH |
1545 | |
| 1546 | # Download playlist videos from API | |
| c5e8d7af PH |
1547 | videos = [] |
| 1548 | ||
| 755eb032 | 1549 | for page_num in itertools.count(1): |
| 771822eb JMF |
1550 | start_index = self._MAX_RESULTS * (page_num - 1) + 1 |
| 1551 | if start_index >= 1000: | |
| 1552 | self._downloader.report_warning(u'Max number of results reached') | |
| 1553 | break | |
| 1554 | url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index) | |
| c5e8d7af PH |
1555 | page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) |
| 1556 | ||
| 1557 | try: | |
| 1558 | response = json.loads(page) | |
| 1559 | except ValueError as err: | |
| 1560 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
| 1561 | ||
| 1562 | if 'feed' not in response: | |
| 1563 | raise ExtractorError(u'Got a malformed response from YouTube API') | |
| 1564 | playlist_title = response['feed']['title']['$t'] | |
| 1565 | if 'entry' not in response['feed']: | |
| 1566 | # Number of videos is a multiple of self._MAX_RESULTS | |
| 1567 | break | |
| 1568 | ||
| 1569 | for entry in response['feed']['entry']: | |
| 1570 | index = entry['yt$position']['$t'] | |
| c215217e JMF |
1571 | if 'media$group' in entry and 'yt$videoid' in entry['media$group']: |
| 1572 | videos.append(( | |
| 1573 | index, | |
| 1574 | 'https://www.youtube.com/watch?v=' + entry['media$group']['yt$videoid']['$t'] | |
| 1575 | )) | |
| c5e8d7af | 1576 | |
| c5e8d7af PH |
1577 | videos = [v[1] for v in sorted(videos)] |
| 1578 | ||
| 20c3893f | 1579 | url_results = [self.url_result(vurl, 'Youtube') for vurl in videos] |
| c5e8d7af PH |
1580 | return [self.playlist_result(url_results, playlist_id, playlist_title)] |
| 1581 | ||
| 1582 | ||
| 1583 | class YoutubeChannelIE(InfoExtractor): | |
| 0f818663 | 1584 | IE_DESC = u'YouTube.com channels' |
| c5e8d7af PH |
1585 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
| 1586 | _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en' | |
| 1587 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' | |
| 252580c5 | 1588 | _MORE_PAGES_URL = 'http://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' |
| c5e8d7af PH |
1589 | IE_NAME = u'youtube:channel' |
| 1590 | ||
| 1591 | def extract_videos_from_page(self, page): | |
| 1592 | ids_in_page = [] | |
| 1593 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
| 1594 | if mobj.group(1) not in ids_in_page: | |
| 1595 | ids_in_page.append(mobj.group(1)) | |
| 1596 | return ids_in_page | |
| 1597 | ||
| 1598 | def _real_extract(self, url): | |
| 1599 | # Extract channel id | |
| 1600 | mobj = re.match(self._VALID_URL, url) | |
| 1601 | if mobj is None: | |
| 1602 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 1603 | ||
| 1604 | # Download channel page | |
| 1605 | channel_id = mobj.group(1) | |
| 1606 | video_ids = [] | |
| 1607 | pagenum = 1 | |
| 1608 | ||
| 1609 | url = self._TEMPLATE_URL % (channel_id, pagenum) | |
| 1610 | page = self._download_webpage(url, channel_id, | |
| 1611 | u'Downloading page #%s' % pagenum) | |
| 1612 | ||
| 1613 | # Extract video identifiers | |
| 1614 | ids_in_page = self.extract_videos_from_page(page) | |
| 1615 | video_ids.extend(ids_in_page) | |
| 1616 | ||
| 1617 | # Download any subsequent channel pages using the json-based channel_ajax query | |
| 1618 | if self._MORE_PAGES_INDICATOR in page: | |
| 755eb032 | 1619 | for pagenum in itertools.count(1): |
| c5e8d7af PH |
1620 | url = self._MORE_PAGES_URL % (pagenum, channel_id) |
| 1621 | page = self._download_webpage(url, channel_id, | |
| 1622 | u'Downloading page #%s' % pagenum) | |
| 1623 | ||
| 1624 | page = json.loads(page) | |
| 1625 | ||
| 1626 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
| 1627 | video_ids.extend(ids_in_page) | |
| 1628 | ||
| 1629 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
| 1630 | break | |
| 1631 | ||
| 1632 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
| 1633 | ||
| 1634 | urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids] | |
| 20c3893f | 1635 | url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls] |
| c5e8d7af PH |
1636 | return [self.playlist_result(url_entries, channel_id)] |
| 1637 | ||
| 1638 | ||
| 1639 | class YoutubeUserIE(InfoExtractor): | |
| 0f818663 | 1640 | IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' |
| 57da92b7 | 1641 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)' |
| c5e8d7af PH |
1642 | _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' |
| 1643 | _GDATA_PAGE_SIZE = 50 | |
| fd9cf738 | 1644 | _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' |
| c5e8d7af PH |
1645 | IE_NAME = u'youtube:user' |
| 1646 | ||
| e3ea4790 | 1647 | @classmethod |
| f4b05232 | 1648 | def suitable(cls, url): |
| e3ea4790 JMF |
1649 | # Don't return True if the url can be extracted with other youtube |
| 1650 | # extractor, the regex would is too permissive and it would match. | |
| 1651 | other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) | |
| 1652 | if any(ie.suitable(url) for ie in other_ies): return False | |
| f4b05232 JMF |
1653 | else: return super(YoutubeUserIE, cls).suitable(url) |
| 1654 | ||
| c5e8d7af PH |
1655 | def _real_extract(self, url): |
| 1656 | # Extract username | |
| 1657 | mobj = re.match(self._VALID_URL, url) | |
| 1658 | if mobj is None: | |
| 1659 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 1660 | ||
| 1661 | username = mobj.group(1) | |
| 1662 | ||
| 1663 | # Download video ids using YouTube Data API. Result size per | |
| 1664 | # query is limited (currently to 50 videos) so we need to query | |
| 1665 | # page by page until there are no video ids - it means we got | |
| 1666 | # all of them. | |
| 1667 | ||
| 1668 | video_ids = [] | |
| c5e8d7af | 1669 | |
| 755eb032 | 1670 | for pagenum in itertools.count(0): |
| c5e8d7af PH |
1671 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 |
| 1672 | ||
| 1673 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
| 1674 | page = self._download_webpage(gdata_url, username, | |
| 1675 | u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) | |
| 1676 | ||
| fd9cf738 JMF |
1677 | try: |
| 1678 | response = json.loads(page) | |
| 1679 | except ValueError as err: | |
| 1680 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
| 71c82637 JMF |
1681 | if 'entry' not in response['feed']: |
| 1682 | # Number of videos is a multiple of self._MAX_RESULTS | |
| 1683 | break | |
| fd9cf738 | 1684 | |
| c5e8d7af PH |
1685 | # Extract video identifiers |
| 1686 | ids_in_page = [] | |
| fd9cf738 JMF |
1687 | for entry in response['feed']['entry']: |
| 1688 | ids_in_page.append(entry['id']['$t'].split('/')[-1]) | |
| c5e8d7af PH |
1689 | video_ids.extend(ids_in_page) |
| 1690 | ||
| 1691 | # A little optimization - if current page is not | |
| 1692 | # "full", ie. does not contain PAGE_SIZE video ids then | |
| 1693 | # we can assume that this page is the last one - there | |
| 1694 | # are no more ids on further pages - no need to query | |
| 1695 | # again. | |
| 1696 | ||
| 1697 | if len(ids_in_page) < self._GDATA_PAGE_SIZE: | |
| 1698 | break | |
| 1699 | ||
| c5e8d7af | 1700 | urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] |
| 20c3893f | 1701 | url_results = [self.url_result(rurl, 'Youtube') for rurl in urls] |
| c5e8d7af | 1702 | return [self.playlist_result(url_results, playlist_title = username)] |
| b05654f0 PH |
1703 | |
| 1704 | class YoutubeSearchIE(SearchInfoExtractor): | |
| 0f818663 | 1705 | IE_DESC = u'YouTube.com searches' |
| b05654f0 PH |
1706 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' |
| 1707 | _MAX_RESULTS = 1000 | |
| 1708 | IE_NAME = u'youtube:search' | |
| 1709 | _SEARCH_KEY = 'ytsearch' | |
| 1710 | ||
| 1711 | def report_download_page(self, query, pagenum): | |
| 1712 | """Report attempt to download search page with given number.""" | |
| 1713 | self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) | |
| 1714 | ||
| 1715 | def _get_n_results(self, query, n): | |
| 1716 | """Get a specified number of results for a query""" | |
| 1717 | ||
| 1718 | video_ids = [] | |
| 1719 | pagenum = 0 | |
| 1720 | limit = n | |
| 1721 | ||
| 1722 | while (50 * pagenum) < limit: | |
| 1723 | self.report_download_page(query, pagenum+1) | |
| 1724 | result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) | |
| 1725 | request = compat_urllib_request.Request(result_url) | |
| 1726 | try: | |
| 1727 | data = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
| 1728 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
| 1729 | raise ExtractorError(u'Unable to download API page: %s' % compat_str(err)) | |
| 1730 | api_response = json.loads(data)['data'] | |
| 1731 | ||
| 1732 | if not 'items' in api_response: | |
| 1733 | raise ExtractorError(u'[youtube] No video results') | |
| 1734 | ||
| 1735 | new_ids = list(video['id'] for video in api_response['items']) | |
| 1736 | video_ids += new_ids | |
| 1737 | ||
| 1738 | limit = min(n, api_response['totalItems']) | |
| 1739 | pagenum += 1 | |
| 1740 | ||
| 1741 | if len(video_ids) > n: | |
| 1742 | video_ids = video_ids[:n] | |
| 1743 | videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids] | |
| 1744 | return self.playlist_result(videos, query) | |
| 75dff0ee JMF |
1745 | |
| 1746 | ||
| 1747 | class YoutubeShowIE(InfoExtractor): | |
| 0f818663 | 1748 | IE_DESC = u'YouTube.com (multi-season) shows' |
| 75dff0ee JMF |
1749 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
| 1750 | IE_NAME = u'youtube:show' | |
| 1751 | ||
| 1752 | def _real_extract(self, url): | |
| 1753 | mobj = re.match(self._VALID_URL, url) | |
| 1754 | show_name = mobj.group(1) | |
| 1755 | webpage = self._download_webpage(url, show_name, u'Downloading show webpage') | |
| 1756 | # There's one playlist for each season of the show | |
| 1757 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
| 1758 | self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) | |
| 1759 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] | |
| 04cc9617 JMF |
1760 | |
| 1761 | ||
| b2e8bc1b | 1762 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): |
| d7ae0639 JMF |
1763 | """ |
| 1764 | Base class for extractors that fetch info from | |
| 1765 | http://www.youtube.com/feed_ajax | |
| 1766 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
| 1767 | """ | |
| b2e8bc1b | 1768 | _LOGIN_REQUIRED = True |
| 04cc9617 | 1769 | _PAGING_STEP = 30 |
| 43ba5456 JMF |
1770 | # use action_load_personal_feed instead of action_load_system_feed |
| 1771 | _PERSONAL_FEED = False | |
| 04cc9617 | 1772 | |
| d7ae0639 JMF |
1773 | @property |
| 1774 | def _FEED_TEMPLATE(self): | |
| 43ba5456 JMF |
1775 | action = 'action_load_system_feed' |
| 1776 | if self._PERSONAL_FEED: | |
| 1777 | action = 'action_load_personal_feed' | |
| 1778 | return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) | |
| d7ae0639 JMF |
1779 | |
| 1780 | @property | |
| 1781 | def IE_NAME(self): | |
| 1782 | return u'youtube:%s' % self._FEED_NAME | |
| 04cc9617 | 1783 | |
| 81f0259b | 1784 | def _real_initialize(self): |
| b2e8bc1b | 1785 | self._login() |
| 81f0259b | 1786 | |
| 04cc9617 JMF |
1787 | def _real_extract(self, url): |
| 1788 | feed_entries = [] | |
| 1789 | # The step argument is available only in 2.7 or higher | |
| 1790 | for i in itertools.count(0): | |
| 1791 | paging = i*self._PAGING_STEP | |
| d7ae0639 JMF |
1792 | info = self._download_webpage(self._FEED_TEMPLATE % paging, |
| 1793 | u'%s feed' % self._FEED_NAME, | |
| 04cc9617 JMF |
1794 | u'Downloading page %s' % i) |
| 1795 | info = json.loads(info) | |
| 1796 | feed_html = info['feed_html'] | |
| 43ba5456 | 1797 | m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) |
| 04cc9617 JMF |
1798 | ids = orderedSet(m.group(1) for m in m_ids) |
| 1799 | feed_entries.extend(self.url_result(id, 'Youtube') for id in ids) | |
| 1800 | if info['paging'] is None: | |
| 1801 | break | |
| d7ae0639 JMF |
1802 | return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) |
| 1803 | ||
| 1804 | class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): | |
| 1805 | IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' | |
| 1806 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
| 1807 | _FEED_NAME = 'subscriptions' | |
| 1808 | _PLAYLIST_TITLE = u'Youtube Subscriptions' | |
| 1809 | ||
| 1810 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): | |
| 1811 | IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' | |
| 1812 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' | |
| 1813 | _FEED_NAME = 'recommended' | |
| 1814 | _PLAYLIST_TITLE = u'Youtube Recommended videos' | |
| c626a3d9 | 1815 | |
| 43ba5456 JMF |
1816 | class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): |
| 1817 | IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' | |
| 1818 | _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' | |
| 1819 | _FEED_NAME = 'watch_later' | |
| 1820 | _PLAYLIST_TITLE = u'Youtube Watch Later' | |
| 1821 | _PAGING_STEP = 100 | |
| 1822 | _PERSONAL_FEED = True | |
| c626a3d9 JMF |
1823 | |
| 1824 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): | |
| 1825 | IE_NAME = u'youtube:favorites' | |
| 1826 | IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' | |
| c7a7750d | 1827 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?' |
| c626a3d9 JMF |
1828 | _LOGIN_REQUIRED = True |
| 1829 | ||
| 1830 | def _real_extract(self, url): | |
| 1831 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
| 1832 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id') | |
| 1833 | return self.url_result(playlist_id, 'YoutubePlaylist') | |
| 15870e90 PH |
1834 | |
| 1835 | ||
| 1836 | class YoutubeTruncatedURLIE(InfoExtractor): | |
| 1837 | IE_NAME = 'youtube:truncated_url' | |
| 1838 | IE_DESC = False # Do not list | |
| 1839 | _VALID_URL = r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$' | |
| 1840 | ||
| 1841 | def _real_extract(self, url): | |
| 1842 | raise ExtractorError( | |
| 1843 | u'Did you forget to quote the URL? Remember that & is a meta ' | |
| 1844 | u'character in most shells, so you want to put the URL in quotes, ' | |
| 1845 | u'like youtube-dl ' | |
| 1846 | u'\'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\'' | |
| 1847 | u' (or simply youtube-dl BaW_jenozKc ).', | |
| 1848 | expected=True) |