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 | 9 | import re |
| e0df6211 PH |
10 | import string |
| 11 | import struct | |
| 12 | import traceback | |
| 13 | import zlib | |
| c5e8d7af | 14 | |
| b05654f0 | 15 | from .common import InfoExtractor, SearchInfoExtractor |
| 54d39d8b | 16 | from .subtitles import SubtitlesInfoExtractor |
| c5e8d7af | 17 | from ..utils import ( |
| edf3e38e | 18 | compat_chr, |
| c5e8d7af | 19 | compat_parse_qs, |
| c5e8d7af PH |
20 | compat_urllib_parse, |
| 21 | compat_urllib_request, | |
| 7c61bd36 | 22 | compat_urlparse, |
| c5e8d7af PH |
23 | compat_str, |
| 24 | ||
| 25 | clean_html, | |
| c38b1e77 | 26 | get_cachedir, |
| c5e8d7af | 27 | get_element_by_id, |
| 652cdaa2 | 28 | get_element_by_attribute, |
| c5e8d7af PH |
29 | ExtractorError, |
| 30 | unescapeHTML, | |
| 31 | unified_strdate, | |
| 04cc9617 | 32 | orderedSet, |
| edf3e38e | 33 | write_json_file, |
| c5e8d7af PH |
34 | ) |
| 35 | ||
| de7f3446 | 36 | class YoutubeBaseInfoExtractor(InfoExtractor): |
| b2e8bc1b JMF |
37 | """Provide base functions for Youtube extractors""" |
| 38 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
| 39 | _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' | |
| 40 | _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' | |
| 41 | _NETRC_MACHINE = 'youtube' | |
| 42 | # If True it will raise an error if no login info is provided | |
| 43 | _LOGIN_REQUIRED = False | |
| 44 | ||
| b2e8bc1b | 45 | def _set_language(self): |
| 7cc3570e PH |
46 | return bool(self._download_webpage( |
| 47 | self._LANG_URL, None, | |
| 48 | note=u'Setting language', errnote='unable to set language', | |
| 49 | fatal=False)) | |
| b2e8bc1b JMF |
50 | |
| 51 | def _login(self): | |
| 52 | (username, password) = self._get_login_info() | |
| 53 | # No authentication to be performed | |
| 54 | if username is None: | |
| 55 | if self._LOGIN_REQUIRED: | |
| 56 | raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) | |
| 57 | return False | |
| 58 | ||
| 7cc3570e PH |
59 | login_page = self._download_webpage( |
| 60 | self._LOGIN_URL, None, | |
| 61 | note=u'Downloading login page', | |
| 62 | errnote=u'unable to fetch login page', fatal=False) | |
| 63 | if login_page is False: | |
| 64 | return | |
| b2e8bc1b | 65 | |
| 795f28f8 PH |
66 | galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"', |
| 67 | login_page, u'Login GALX parameter') | |
| c5e8d7af | 68 | |
| b2e8bc1b JMF |
69 | # Log in |
| 70 | login_form_strs = { | |
| 71 | u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
| 72 | u'Email': username, | |
| 73 | u'GALX': galx, | |
| 74 | u'Passwd': password, | |
| 75 | u'PersistentCookie': u'yes', | |
| 76 | u'_utf8': u'霱', | |
| 77 | u'bgresponse': u'js_disabled', | |
| 78 | u'checkConnection': u'', | |
| 79 | u'checkedDomains': u'youtube', | |
| 80 | u'dnConn': u'', | |
| b2e8bc1b JMF |
81 | u'pstMsg': u'0', |
| 82 | u'rmShown': u'1', | |
| 83 | u'secTok': u'', | |
| 84 | u'signIn': u'Sign in', | |
| 85 | u'timeStmp': u'', | |
| 86 | u'service': u'youtube', | |
| 87 | u'uilel': u'3', | |
| 88 | u'hl': u'en_US', | |
| 89 | } | |
| 90 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
| 91 | # chokes on unicode | |
| 92 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) | |
| 93 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
| 7cc3570e PH |
94 | |
| 95 | req = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
| 96 | login_results = self._download_webpage( | |
| 97 | req, None, | |
| 98 | note=u'Logging in', errnote=u'unable to log in', fatal=False) | |
| 99 | if login_results is False: | |
| 100 | return False | |
| 101 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
| 102 | self._downloader.report_warning(u'unable to log in: bad username or password') | |
| b2e8bc1b JMF |
103 | return False |
| 104 | return True | |
| 105 | ||
| 106 | def _confirm_age(self): | |
| 107 | age_form = { | |
| 7cc3570e PH |
108 | 'next_url': '/', |
| 109 | 'action_confirm': 'Confirm', | |
| 110 | } | |
| 111 | req = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) | |
| 112 | ||
| 113 | self._download_webpage( | |
| 114 | req, None, | |
| 115 | note=u'Confirming age', errnote=u'Unable to confirm age') | |
| b2e8bc1b JMF |
116 | return True |
| 117 | ||
| 118 | def _real_initialize(self): | |
| 119 | if self._downloader is None: | |
| 120 | return | |
| 121 | if not self._set_language(): | |
| 122 | return | |
| 123 | if not self._login(): | |
| 124 | return | |
| 125 | self._confirm_age() | |
| c5e8d7af | 126 | |
| 8377574c | 127 | |
| de7f3446 | 128 | class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): |
| 0f818663 | 129 | IE_DESC = u'YouTube.com' |
| cb7dfeea | 130 | _VALID_URL = r"""(?x)^ |
| c5e8d7af | 131 | ( |
| 83aa5293 | 132 | (?:https?://|//)? # http(s):// or protocol-independent URL (optional) |
| cb7dfeea | 133 | (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| |
| e69ae5b9 JMF |
134 | tube\.majestyc\.net/| |
| 135 | youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains | |
| c5e8d7af PH |
136 | (?:.*?\#/)? # handle anchor (#/) redirect urls |
| 137 | (?: # the various things that can precede the ID: | |
| 138 | (?:(?:v|embed|e)/) # v/ or embed/ or e/ | |
| 139 | |(?: # or the v= param in all its forms | |
| d741e55a | 140 | (?:(?:watch|movie)(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) |
| c5e8d7af PH |
141 | (?:\?|\#!?) # the params delimiter ? or # or #! |
| 142 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
| 143 | v= | |
| 144 | ) | |
| f4b05232 JMF |
145 | )) |
| 146 | |youtu\.be/ # just youtu.be/xxxx | |
| 147 | ) | |
| c5e8d7af | 148 | )? # all until now is optional -> you can pass the naked ID |
| 8963d9c2 | 149 | ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID |
| c5e8d7af PH |
150 | (?(1).+)? # if we found the ID, everything can follow |
| 151 | $""" | |
| c5e8d7af | 152 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' |
| c5e8d7af | 153 | # Listed in order of quality |
| bdc6b3fc | 154 | _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '36', '17', '13', |
| 96fb5605 | 155 | # Apple HTTP Live Streaming |
| bdc6b3fc | 156 | '96', '95', '94', '93', '92', '132', '151', |
| 939fbd26 JMF |
157 | # 3D |
| 158 | '85', '84', '102', '83', '101', '82', '100', | |
| 159 | # Dash video | |
| 160 | '138', '137', '248', '136', '247', '135', '246', | |
| 161 | '245', '244', '134', '243', '133', '242', '160', | |
| 162 | # Dash audio | |
| 163 | '141', '172', '140', '171', '139', | |
| 1d043b93 | 164 | ] |
| c5e8d7af PH |
165 | _video_extensions = { |
| 166 | '13': '3gp', | |
| bdc6b3fc | 167 | '17': '3gp', |
| c5e8d7af PH |
168 | '18': 'mp4', |
| 169 | '22': 'mp4', | |
| bdc6b3fc | 170 | '36': '3gp', |
| c5e8d7af | 171 | '37': 'mp4', |
| d69cf69a | 172 | '38': 'mp4', |
| c5e8d7af PH |
173 | '43': 'webm', |
| 174 | '44': 'webm', | |
| 175 | '45': 'webm', | |
| 176 | '46': 'webm', | |
| 1d043b93 | 177 | |
| 86fe61c8 AZ |
178 | # 3d videos |
| 179 | '82': 'mp4', | |
| 180 | '83': 'mp4', | |
| 181 | '84': 'mp4', | |
| 182 | '85': 'mp4', | |
| 183 | '100': 'webm', | |
| 184 | '101': 'webm', | |
| 185 | '102': 'webm', | |
| 836a086c | 186 | |
| 96fb5605 | 187 | # Apple HTTP Live Streaming |
| 1d043b93 JMF |
188 | '92': 'mp4', |
| 189 | '93': 'mp4', | |
| 190 | '94': 'mp4', | |
| 191 | '95': 'mp4', | |
| 192 | '96': 'mp4', | |
| 193 | '132': 'mp4', | |
| 194 | '151': 'mp4', | |
| 836a086c AZ |
195 | |
| 196 | # Dash mp4 | |
| 197 | '133': 'mp4', | |
| 198 | '134': 'mp4', | |
| 199 | '135': 'mp4', | |
| 200 | '136': 'mp4', | |
| 201 | '137': 'mp4', | |
| 202 | '138': 'mp4', | |
| 836a086c AZ |
203 | '160': 'mp4', |
| 204 | ||
| f6f1fc92 RB |
205 | # Dash mp4 audio |
| 206 | '139': 'm4a', | |
| 16f36a6f RB |
207 | '140': 'm4a', |
| 208 | '141': 'm4a', | |
| 836a086c AZ |
209 | |
| 210 | # Dash webm | |
| 211 | '171': 'webm', | |
| 212 | '172': 'webm', | |
| 213 | '242': 'webm', | |
| 214 | '243': 'webm', | |
| 215 | '244': 'webm', | |
| 216 | '245': 'webm', | |
| 217 | '246': 'webm', | |
| 218 | '247': 'webm', | |
| 219 | '248': 'webm', | |
| c5e8d7af PH |
220 | } |
| 221 | _video_dimensions = { | |
| dbd1988e | 222 | '5': {'width': 400, 'height': 240}, |
| 223 | '6': {}, | |
| 224 | '13': {}, | |
| 225 | '17': {'width': 176, 'height': 144}, | |
| 226 | '18': {'width': 640, 'height': 360}, | |
| 227 | '22': {'width': 1280, 'height': 720}, | |
| 228 | '34': {'width': 640, 'height': 360}, | |
| 229 | '35': {'width': 854, 'height': 480}, | |
| 230 | '36': {'width': 320, 'height': 240}, | |
| 231 | '37': {'width': 1920, 'height': 1080}, | |
| 232 | '38': {'width': 4096, 'height': 3072}, | |
| 233 | '43': {'width': 640, 'height': 360}, | |
| 234 | '44': {'width': 854, 'height': 480}, | |
| 235 | '45': {'width': 1280, 'height': 720}, | |
| 236 | '46': {'width': 1920, 'height': 1080}, | |
| 237 | '82': {'height': 360, 'display': '360p'}, | |
| 238 | '83': {'height': 480, 'display': '480p'}, | |
| 239 | '84': {'height': 720, 'display': '720p'}, | |
| 240 | '85': {'height': 1080, 'display': '1080p'}, | |
| 241 | '92': {'height': 240, 'display': '240p'}, | |
| 242 | '93': {'height': 360, 'display': '360p'}, | |
| 243 | '94': {'height': 480, 'display': '480p'}, | |
| 244 | '95': {'height': 720, 'display': '720p'}, | |
| 245 | '96': {'height': 1080, 'display': '1080p'}, | |
| 246 | '100': {'height': 360, 'display': '360p'}, | |
| 247 | '101': {'height': 480, 'display': '480p'}, | |
| 248 | '102': {'height': 720, 'display': '720p'}, | |
| 249 | '132': {'height': 240, 'display': '240p'}, | |
| 250 | '151': {'height': 72, 'display': '72p'}, | |
| 251 | '133': {'height': 240, 'display': '240p'}, | |
| 252 | '134': {'height': 360, 'display': '360p'}, | |
| 253 | '135': {'height': 480, 'display': '480p'}, | |
| 254 | '136': {'height': 720, 'display': '720p'}, | |
| 255 | '137': {'height': 1080, 'display': '1080p'}, | |
| 256 | '138': {'height': 1081, 'display': '>1080p'}, | |
| 257 | '139': {'display': '48k'}, | |
| 258 | '140': {'display': '128k'}, | |
| 259 | '141': {'display': '256k'}, | |
| 260 | '160': {'height': 192, 'display': '192p'}, | |
| 261 | '171': {'display': '128k'}, | |
| 262 | '172': {'display': '256k'}, | |
| 263 | '242': {'height': 240, 'display': '240p'}, | |
| 264 | '243': {'height': 360, 'display': '360p'}, | |
| 265 | '244': {'height': 480, 'display': '480p'}, | |
| 266 | '245': {'height': 480, 'display': '480p'}, | |
| 267 | '246': {'height': 480, 'display': '480p'}, | |
| 268 | '247': {'height': 720, 'display': '720p'}, | |
| 269 | '248': {'height': 1080, 'display': '1080p'}, | |
| c5e8d7af | 270 | } |
| 836a086c AZ |
271 | _special_itags = { |
| 272 | '82': '3D', | |
| 273 | '83': '3D', | |
| 274 | '84': '3D', | |
| 275 | '85': '3D', | |
| 276 | '100': '3D', | |
| 277 | '101': '3D', | |
| 278 | '102': '3D', | |
| 279 | '133': 'DASH Video', | |
| 280 | '134': 'DASH Video', | |
| 281 | '135': 'DASH Video', | |
| 282 | '136': 'DASH Video', | |
| 283 | '137': 'DASH Video', | |
| 284 | '138': 'DASH Video', | |
| 285 | '139': 'DASH Audio', | |
| 286 | '140': 'DASH Audio', | |
| 287 | '141': 'DASH Audio', | |
| 288 | '160': 'DASH Video', | |
| 289 | '171': 'DASH Audio', | |
| 290 | '172': 'DASH Audio', | |
| 291 | '242': 'DASH Video', | |
| 292 | '243': 'DASH Video', | |
| 293 | '244': 'DASH Video', | |
| 294 | '245': 'DASH Video', | |
| 295 | '246': 'DASH Video', | |
| 296 | '247': 'DASH Video', | |
| 297 | '248': 'DASH Video', | |
| c5e8d7af | 298 | } |
| 836a086c | 299 | |
| c5e8d7af | 300 | IE_NAME = u'youtube' |
| 2eb88d95 PH |
301 | _TESTS = [ |
| 302 | { | |
| 0e853ca4 PH |
303 | u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", |
| 304 | u"file": u"BaW_jenozKc.mp4", | |
| 305 | u"info_dict": { | |
| 306 | u"title": u"youtube-dl test video \"'/\\ä↭𝕐", | |
| 307 | u"uploader": u"Philipp Hagemeister", | |
| 308 | u"uploader_id": u"phihag", | |
| 309 | u"upload_date": u"20121002", | |
| 27dcce19 | 310 | u"description": u"test chars: \"'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ." |
| 2eb88d95 | 311 | } |
| 0e853ca4 | 312 | }, |
| 0e853ca4 PH |
313 | { |
| 314 | u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", | |
| 315 | u"file": u"UxxajLWwzqY.mp4", | |
| 316 | u"note": u"Test generic use_cipher_signature video (#897)", | |
| 317 | u"info_dict": { | |
| 318 | u"upload_date": u"20120506", | |
| 319 | u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", | |
| 2dc59299 | 320 | u"description": u"md5:5b292926389560516e384ac437c0ec07", |
| 45ed795c | 321 | u"uploader": u"Icona Pop", |
| 0e853ca4 | 322 | u"uploader_id": u"IconaPop" |
| 2eb88d95 | 323 | } |
| c108eb73 JMF |
324 | }, |
| 325 | { | |
| 326 | u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", | |
| 327 | u"file": u"07FYdnEawAQ.mp4", | |
| 328 | u"note": u"Test VEVO video with age protection (#956)", | |
| 329 | u"info_dict": { | |
| 330 | u"upload_date": u"20130703", | |
| 331 | u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", | |
| 332 | u"description": u"md5:64249768eec3bc4276236606ea996373", | |
| 333 | u"uploader": u"justintimberlakeVEVO", | |
| 334 | u"uploader_id": u"justintimberlakeVEVO" | |
| 335 | } | |
| 336 | }, | |
| fccd3771 | 337 | { |
| 83aa5293 | 338 | u"url": u"//www.YouTube.com/watch?v=yZIXLfi8CZQ", |
| fccd3771 PH |
339 | u"file": u"yZIXLfi8CZQ.mp4", |
| 340 | u"note": u"Embed-only video (#1746)", | |
| 341 | u"info_dict": { | |
| 342 | u"upload_date": u"20120608", | |
| 343 | u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012", | |
| 344 | u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7", | |
| 345 | u"uploader": u"SET India", | |
| 346 | u"uploader_id": u"setindia" | |
| 347 | } | |
| 348 | }, | |
| 2eb88d95 PH |
349 | ] |
| 350 | ||
| c5e8d7af PH |
351 | |
| 352 | @classmethod | |
| 353 | def suitable(cls, url): | |
| 354 | """Receives a URL and returns True if suitable for this IE.""" | |
| e3ea4790 | 355 | if YoutubePlaylistIE.suitable(url): return False |
| fccd3771 | 356 | return re.match(cls._VALID_URL, url) is not None |
| c5e8d7af | 357 | |
| e0df6211 PH |
358 | def __init__(self, *args, **kwargs): |
| 359 | super(YoutubeIE, self).__init__(*args, **kwargs) | |
| 83799698 | 360 | self._player_cache = {} |
| e0df6211 | 361 | |
| c5e8d7af PH |
362 | def report_video_info_webpage_download(self, video_id): |
| 363 | """Report attempt to download video info webpage.""" | |
| 364 | self.to_screen(u'%s: Downloading video info webpage' % video_id) | |
| 365 | ||
| c5e8d7af PH |
366 | def report_information_extraction(self, video_id): |
| 367 | """Report attempt to extract video information.""" | |
| 368 | self.to_screen(u'%s: Extracting video information' % video_id) | |
| 369 | ||
| 370 | def report_unavailable_format(self, video_id, format): | |
| 371 | """Report extracted video URL.""" | |
| 372 | self.to_screen(u'%s: Format %s not available' % (video_id, format)) | |
| 373 | ||
| 374 | def report_rtmp_download(self): | |
| 375 | """Indicate the download will use the RTMP protocol.""" | |
| 376 | self.to_screen(u'RTMP download detected') | |
| 377 | ||
| c4417ddb PH |
378 | def _extract_signature_function(self, video_id, player_url, slen): |
| 379 | id_m = re.match(r'.*-(?P<id>[a-zA-Z0-9_-]+)\.(?P<ext>[a-z]+)$', | |
| 83799698 | 380 | player_url) |
| e0df6211 PH |
381 | player_type = id_m.group('ext') |
| 382 | player_id = id_m.group('id') | |
| 383 | ||
| c4417ddb PH |
384 | # Read from filesystem cache |
| 385 | func_id = '%s_%s_%d' % (player_type, player_id, slen) | |
| 386 | assert os.path.basename(func_id) == func_id | |
| c38b1e77 | 387 | cache_dir = get_cachedir(self._downloader.params) |
| c4417ddb | 388 | |
| c3c88a26 | 389 | cache_enabled = cache_dir is not None |
| f8061589 | 390 | if cache_enabled: |
| c4417ddb PH |
391 | cache_fn = os.path.join(os.path.expanduser(cache_dir), |
| 392 | u'youtube-sigfuncs', | |
| 393 | func_id + '.json') | |
| 394 | try: | |
| edf3e38e | 395 | with io.open(cache_fn, 'r', encoding='utf-8') as cachef: |
| c4417ddb PH |
396 | cache_spec = json.load(cachef) |
| 397 | return lambda s: u''.join(s[i] for i in cache_spec) | |
| edf3e38e | 398 | except IOError: |
| c4417ddb | 399 | pass # No cache available |
| 83799698 | 400 | |
| e0df6211 PH |
401 | if player_type == 'js': |
| 402 | code = self._download_webpage( | |
| 403 | player_url, video_id, | |
| 83799698 | 404 | note=u'Downloading %s player %s' % (player_type, player_id), |
| e0df6211 | 405 | errnote=u'Download of %s failed' % player_url) |
| 83799698 | 406 | res = self._parse_sig_js(code) |
| c4417ddb | 407 | elif player_type == 'swf': |
| e0df6211 PH |
408 | urlh = self._request_webpage( |
| 409 | player_url, video_id, | |
| 83799698 | 410 | note=u'Downloading %s player %s' % (player_type, player_id), |
| e0df6211 PH |
411 | errnote=u'Download of %s failed' % player_url) |
| 412 | code = urlh.read() | |
| 83799698 | 413 | res = self._parse_sig_swf(code) |
| e0df6211 PH |
414 | else: |
| 415 | assert False, 'Invalid player type %r' % player_type | |
| 416 | ||
| f8061589 | 417 | if cache_enabled: |
| edf3e38e | 418 | try: |
| c705320f PH |
419 | test_string = u''.join(map(compat_chr, range(slen))) |
| 420 | cache_res = res(test_string) | |
| edf3e38e PH |
421 | cache_spec = [ord(c) for c in cache_res] |
| 422 | try: | |
| 423 | os.makedirs(os.path.dirname(cache_fn)) | |
| 424 | except OSError as ose: | |
| 425 | if ose.errno != errno.EEXIST: | |
| 426 | raise | |
| 427 | write_json_file(cache_spec, cache_fn) | |
| 0ca96d48 | 428 | except Exception: |
| edf3e38e PH |
429 | tb = traceback.format_exc() |
| 430 | self._downloader.report_warning( | |
| 431 | u'Writing cache to %r failed: %s' % (cache_fn, tb)) | |
| 83799698 PH |
432 | |
| 433 | return res | |
| 434 | ||
| edf3e38e PH |
435 | def _print_sig_code(self, func, slen): |
| 436 | def gen_sig_code(idxs): | |
| 437 | def _genslice(start, end, step): | |
| 438 | starts = u'' if start == 0 else str(start) | |
| e35e4ddc PH |
439 | ends = (u':%d' % (end+step)) if end + step >= 0 else u':' |
| 440 | steps = u'' if step == 1 else (u':%d' % step) | |
| edf3e38e PH |
441 | return u's[%s%s%s]' % (starts, ends, steps) |
| 442 | ||
| 443 | step = None | |
| 0ca96d48 PH |
444 | start = '(Never used)' # Quelch pyflakes warnings - start will be |
| 445 | # set as soon as step is set | |
| edf3e38e PH |
446 | for i, prev in zip(idxs[1:], idxs[:-1]): |
| 447 | if step is not None: | |
| 448 | if i - prev == step: | |
| 449 | continue | |
| 450 | yield _genslice(start, prev, step) | |
| 451 | step = None | |
| 452 | continue | |
| 453 | if i - prev in [-1, 1]: | |
| 454 | step = i - prev | |
| 455 | start = prev | |
| 456 | continue | |
| 457 | else: | |
| 458 | yield u's[%d]' % prev | |
| 459 | if step is None: | |
| 460 | yield u's[%d]' % i | |
| 461 | else: | |
| 462 | yield _genslice(start, i, step) | |
| 463 | ||
| c705320f PH |
464 | test_string = u''.join(map(compat_chr, range(slen))) |
| 465 | cache_res = func(test_string) | |
| edf3e38e PH |
466 | cache_spec = [ord(c) for c in cache_res] |
| 467 | expr_code = u' + '.join(gen_sig_code(cache_spec)) | |
| 468 | code = u'if len(s) == %d:\n return %s\n' % (slen, expr_code) | |
| f8061589 | 469 | self.to_screen(u'Extracted signature function:\n' + code) |
| edf3e38e | 470 | |
| e0df6211 PH |
471 | def _parse_sig_js(self, jscode): |
| 472 | funcname = self._search_regex( | |
| 473 | r'signature=([a-zA-Z]+)', jscode, | |
| 474 | u'Initial JS player signature function name') | |
| 475 | ||
| 476 | functions = {} | |
| 477 | ||
| 478 | def argidx(varname): | |
| 479 | return string.lowercase.index(varname) | |
| 480 | ||
| 481 | def interpret_statement(stmt, local_vars, allow_recursion=20): | |
| 482 | if allow_recursion < 0: | |
| 0ca96d48 | 483 | raise ExtractorError(u'Recursion limit reached') |
| e0df6211 PH |
484 | |
| 485 | if stmt.startswith(u'var '): | |
| 486 | stmt = stmt[len(u'var '):] | |
| 487 | ass_m = re.match(r'^(?P<out>[a-z]+)(?:\[(?P<index>[^\]]+)\])?' + | |
| 488 | r'=(?P<expr>.*)$', stmt) | |
| 489 | if ass_m: | |
| 490 | if ass_m.groupdict().get('index'): | |
| 491 | def assign(val): | |
| 492 | lvar = local_vars[ass_m.group('out')] | |
| 493 | idx = interpret_expression(ass_m.group('index'), | |
| 494 | local_vars, allow_recursion) | |
| 495 | assert isinstance(idx, int) | |
| 496 | lvar[idx] = val | |
| 497 | return val | |
| 498 | expr = ass_m.group('expr') | |
| 499 | else: | |
| 500 | def assign(val): | |
| 501 | local_vars[ass_m.group('out')] = val | |
| 502 | return val | |
| 503 | expr = ass_m.group('expr') | |
| 504 | elif stmt.startswith(u'return '): | |
| 505 | assign = lambda v: v | |
| 506 | expr = stmt[len(u'return '):] | |
| 507 | else: | |
| 508 | raise ExtractorError( | |
| 509 | u'Cannot determine left side of statement in %r' % stmt) | |
| 510 | ||
| 511 | v = interpret_expression(expr, local_vars, allow_recursion) | |
| 512 | return assign(v) | |
| 513 | ||
| 514 | def interpret_expression(expr, local_vars, allow_recursion): | |
| 515 | if expr.isdigit(): | |
| 516 | return int(expr) | |
| 517 | ||
| 518 | if expr.isalpha(): | |
| 519 | return local_vars[expr] | |
| 520 | ||
| 521 | m = re.match(r'^(?P<in>[a-z]+)\.(?P<member>.*)$', expr) | |
| 522 | if m: | |
| 523 | member = m.group('member') | |
| 524 | val = local_vars[m.group('in')] | |
| 525 | if member == 'split("")': | |
| 526 | return list(val) | |
| 527 | if member == 'join("")': | |
| 528 | return u''.join(val) | |
| 529 | if member == 'length': | |
| 530 | return len(val) | |
| 531 | if member == 'reverse()': | |
| 532 | return val[::-1] | |
| 533 | slice_m = re.match(r'slice\((?P<idx>.*)\)', member) | |
| 534 | if slice_m: | |
| 535 | idx = interpret_expression( | |
| 536 | slice_m.group('idx'), local_vars, allow_recursion-1) | |
| 537 | return val[idx:] | |
| 538 | ||
| 539 | m = re.match( | |
| 540 | r'^(?P<in>[a-z]+)\[(?P<idx>.+)\]$', expr) | |
| 541 | if m: | |
| 542 | val = local_vars[m.group('in')] | |
| 543 | idx = interpret_expression(m.group('idx'), local_vars, | |
| 544 | allow_recursion-1) | |
| 545 | return val[idx] | |
| 546 | ||
| 547 | m = re.match(r'^(?P<a>.+?)(?P<op>[%])(?P<b>.+?)$', expr) | |
| 548 | if m: | |
| 549 | a = interpret_expression(m.group('a'), | |
| 550 | local_vars, allow_recursion) | |
| 551 | b = interpret_expression(m.group('b'), | |
| 552 | local_vars, allow_recursion) | |
| 553 | return a % b | |
| 554 | ||
| 555 | m = re.match( | |
| 556 | r'^(?P<func>[a-zA-Z]+)\((?P<args>[a-z0-9,]+)\)$', expr) | |
| 557 | if m: | |
| 558 | fname = m.group('func') | |
| 559 | if fname not in functions: | |
| 560 | functions[fname] = extract_function(fname) | |
| 561 | argvals = [int(v) if v.isdigit() else local_vars[v] | |
| 562 | for v in m.group('args').split(',')] | |
| 563 | return functions[fname](argvals) | |
| 564 | raise ExtractorError(u'Unsupported JS expression %r' % expr) | |
| 565 | ||
| 566 | def extract_function(funcname): | |
| 567 | func_m = re.search( | |
| 568 | r'function ' + re.escape(funcname) + | |
| 569 | r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}', | |
| 570 | jscode) | |
| 571 | argnames = func_m.group('args').split(',') | |
| 572 | ||
| 573 | def resf(args): | |
| 574 | local_vars = dict(zip(argnames, args)) | |
| 575 | for stmt in func_m.group('code').split(';'): | |
| 576 | res = interpret_statement(stmt, local_vars) | |
| 577 | return res | |
| 578 | return resf | |
| 579 | ||
| 580 | initial_function = extract_function(funcname) | |
| 581 | return lambda s: initial_function([s]) | |
| 582 | ||
| 583 | def _parse_sig_swf(self, file_contents): | |
| 584 | if file_contents[1:3] != b'WS': | |
| 585 | raise ExtractorError( | |
| 586 | u'Not an SWF file; header is %r' % file_contents[:3]) | |
| 587 | if file_contents[:1] == b'C': | |
| 588 | content = zlib.decompress(file_contents[8:]) | |
| 589 | else: | |
| 590 | raise NotImplementedError(u'Unsupported compression format %r' % | |
| 591 | file_contents[:1]) | |
| 592 | ||
| 593 | def extract_tags(content): | |
| 594 | pos = 0 | |
| 595 | while pos < len(content): | |
| 596 | header16 = struct.unpack('<H', content[pos:pos+2])[0] | |
| 597 | pos += 2 | |
| 598 | tag_code = header16 >> 6 | |
| 599 | tag_len = header16 & 0x3f | |
| 600 | if tag_len == 0x3f: | |
| 601 | tag_len = struct.unpack('<I', content[pos:pos+4])[0] | |
| 602 | pos += 4 | |
| 603 | assert pos+tag_len <= len(content) | |
| 604 | yield (tag_code, content[pos:pos+tag_len]) | |
| 605 | pos += tag_len | |
| 606 | ||
| 607 | code_tag = next(tag | |
| 608 | for tag_code, tag in extract_tags(content) | |
| 609 | if tag_code == 82) | |
| 610 | p = code_tag.index(b'\0', 4) + 1 | |
| ba552f54 | 611 | code_reader = io.BytesIO(code_tag[p:]) |
| e0df6211 PH |
612 | |
| 613 | # Parse ABC (AVM2 ByteCode) | |
| ba552f54 PH |
614 | def read_int(reader=None): |
| 615 | if reader is None: | |
| 616 | reader = code_reader | |
| e0df6211 PH |
617 | res = 0 |
| 618 | shift = 0 | |
| 619 | for _ in range(5): | |
| ba552f54 PH |
620 | buf = reader.read(1) |
| 621 | assert len(buf) == 1 | |
| 622 | b = struct.unpack('<B', buf)[0] | |
| e0df6211 PH |
623 | res = res | ((b & 0x7f) << shift) |
| 624 | if b & 0x80 == 0: | |
| 625 | break | |
| 626 | shift += 7 | |
| ba552f54 PH |
627 | return res |
| 628 | ||
| 629 | def u30(reader=None): | |
| 630 | res = read_int(reader) | |
| 631 | assert res & 0xf0000000 == 0 | |
| e0df6211 PH |
632 | return res |
| 633 | u32 = read_int | |
| 634 | ||
| ba552f54 PH |
635 | def s32(reader=None): |
| 636 | v = read_int(reader) | |
| e0df6211 PH |
637 | if v & 0x80000000 != 0: |
| 638 | v = - ((v ^ 0xffffffff) + 1) | |
| ba552f54 PH |
639 | return v |
| 640 | ||
| 0ca96d48 | 641 | def read_string(reader=None): |
| ba552f54 PH |
642 | if reader is None: |
| 643 | reader = code_reader | |
| 644 | slen = u30(reader) | |
| 645 | resb = reader.read(slen) | |
| 646 | assert len(resb) == slen | |
| 647 | return resb.decode('utf-8') | |
| 648 | ||
| 649 | def read_bytes(count, reader=None): | |
| 650 | if reader is None: | |
| 651 | reader = code_reader | |
| 652 | resb = reader.read(count) | |
| 653 | assert len(resb) == count | |
| 654 | return resb | |
| 655 | ||
| 656 | def read_byte(reader=None): | |
| 657 | resb = read_bytes(1, reader=reader) | |
| 658 | res = struct.unpack('<B', resb)[0] | |
| 659 | return res | |
| e0df6211 PH |
660 | |
| 661 | # minor_version + major_version | |
| 0ca96d48 | 662 | read_bytes(2 + 2) |
| e0df6211 PH |
663 | |
| 664 | # Constant pool | |
| ba552f54 | 665 | int_count = u30() |
| e0df6211 | 666 | for _c in range(1, int_count): |
| 0ca96d48 | 667 | s32() |
| ba552f54 | 668 | uint_count = u30() |
| e0df6211 | 669 | for _c in range(1, uint_count): |
| 0ca96d48 | 670 | u32() |
| ba552f54 | 671 | double_count = u30() |
| 0ca96d48 | 672 | read_bytes((double_count-1) * 8) |
| ba552f54 | 673 | string_count = u30() |
| e0df6211 PH |
674 | constant_strings = [u''] |
| 675 | for _c in range(1, string_count): | |
| 0ca96d48 | 676 | s = read_string() |
| e0df6211 | 677 | constant_strings.append(s) |
| ba552f54 | 678 | namespace_count = u30() |
| e0df6211 | 679 | for _c in range(1, namespace_count): |
| 0ca96d48 PH |
680 | read_bytes(1) # kind |
| 681 | u30() # name | |
| ba552f54 | 682 | ns_set_count = u30() |
| e0df6211 | 683 | for _c in range(1, ns_set_count): |
| ba552f54 | 684 | count = u30() |
| e0df6211 | 685 | for _c2 in range(count): |
| 0ca96d48 | 686 | u30() |
| ba552f54 | 687 | multiname_count = u30() |
| e0df6211 PH |
688 | MULTINAME_SIZES = { |
| 689 | 0x07: 2, # QName | |
| 690 | 0x0d: 2, # QNameA | |
| 691 | 0x0f: 1, # RTQName | |
| 692 | 0x10: 1, # RTQNameA | |
| 693 | 0x11: 0, # RTQNameL | |
| 694 | 0x12: 0, # RTQNameLA | |
| 695 | 0x09: 2, # Multiname | |
| 696 | 0x0e: 2, # MultinameA | |
| 697 | 0x1b: 1, # MultinameL | |
| 698 | 0x1c: 1, # MultinameLA | |
| 699 | } | |
| 700 | multinames = [u''] | |
| 701 | for _c in range(1, multiname_count): | |
| ba552f54 | 702 | kind = u30() |
| e0df6211 PH |
703 | assert kind in MULTINAME_SIZES, u'Invalid multiname kind %r' % kind |
| 704 | if kind == 0x07: | |
| 0ca96d48 | 705 | u30() # namespace_idx |
| ba552f54 | 706 | name_idx = u30() |
| e0df6211 PH |
707 | multinames.append(constant_strings[name_idx]) |
| 708 | else: | |
| 709 | multinames.append('[MULTINAME kind: %d]' % kind) | |
| 710 | for _c2 in range(MULTINAME_SIZES[kind]): | |
| 0ca96d48 | 711 | u30() |
| e0df6211 PH |
712 | |
| 713 | # Methods | |
| ba552f54 | 714 | method_count = u30() |
| e0df6211 PH |
715 | MethodInfo = collections.namedtuple( |
| 716 | 'MethodInfo', | |
| 717 | ['NEED_ARGUMENTS', 'NEED_REST']) | |
| 718 | method_infos = [] | |
| 719 | for method_id in range(method_count): | |
| ba552f54 | 720 | param_count = u30() |
| 0ca96d48 | 721 | u30() # return type |
| e0df6211 | 722 | for _ in range(param_count): |
| 0ca96d48 PH |
723 | u30() # param type |
| 724 | u30() # name index (always 0 for youtube) | |
| ba552f54 | 725 | flags = read_byte() |
| e0df6211 PH |
726 | if flags & 0x08 != 0: |
| 727 | # Options present | |
| ba552f54 | 728 | option_count = u30() |
| e0df6211 | 729 | for c in range(option_count): |
| 0ca96d48 PH |
730 | u30() # val |
| 731 | read_bytes(1) # kind | |
| e0df6211 PH |
732 | if flags & 0x80 != 0: |
| 733 | # Param names present | |
| 734 | for _ in range(param_count): | |
| 0ca96d48 | 735 | u30() # param name |
| e0df6211 PH |
736 | mi = MethodInfo(flags & 0x01 != 0, flags & 0x04 != 0) |
| 737 | method_infos.append(mi) | |
| 738 | ||
| 739 | # Metadata | |
| ba552f54 | 740 | metadata_count = u30() |
| e0df6211 | 741 | for _c in range(metadata_count): |
| 0ca96d48 | 742 | u30() # name |
| ba552f54 | 743 | item_count = u30() |
| e0df6211 | 744 | for _c2 in range(item_count): |
| 0ca96d48 PH |
745 | u30() # key |
| 746 | u30() # value | |
| ba552f54 PH |
747 | |
| 748 | def parse_traits_info(): | |
| 749 | trait_name_idx = u30() | |
| 750 | kind_full = read_byte() | |
| e0df6211 PH |
751 | kind = kind_full & 0x0f |
| 752 | attrs = kind_full >> 4 | |
| 753 | methods = {} | |
| 754 | if kind in [0x00, 0x06]: # Slot or Const | |
| 0ca96d48 PH |
755 | u30() # Slot id |
| 756 | u30() # type_name_idx | |
| ba552f54 | 757 | vindex = u30() |
| e0df6211 | 758 | if vindex != 0: |
| 0ca96d48 | 759 | read_byte() # vkind |
| e0df6211 | 760 | elif kind in [0x01, 0x02, 0x03]: # Method / Getter / Setter |
| 0ca96d48 | 761 | u30() # disp_id |
| ba552f54 | 762 | method_idx = u30() |
| e0df6211 PH |
763 | methods[multinames[trait_name_idx]] = method_idx |
| 764 | elif kind == 0x04: # Class | |
| 0ca96d48 PH |
765 | u30() # slot_id |
| 766 | u30() # classi | |
| e0df6211 | 767 | elif kind == 0x05: # Function |
| 0ca96d48 | 768 | u30() # slot_id |
| ba552f54 | 769 | function_idx = u30() |
| e0df6211 PH |
770 | methods[function_idx] = multinames[trait_name_idx] |
| 771 | else: | |
| 772 | raise ExtractorError(u'Unsupported trait kind %d' % kind) | |
| 773 | ||
| 774 | if attrs & 0x4 != 0: # Metadata present | |
| ba552f54 | 775 | metadata_count = u30() |
| e0df6211 | 776 | for _c3 in range(metadata_count): |
| 0ca96d48 | 777 | u30() # metadata index |
| e0df6211 | 778 | |
| ba552f54 | 779 | return methods |
| e0df6211 PH |
780 | |
| 781 | # Classes | |
| 782 | TARGET_CLASSNAME = u'SignatureDecipher' | |
| 783 | searched_idx = multinames.index(TARGET_CLASSNAME) | |
| 784 | searched_class_id = None | |
| ba552f54 | 785 | class_count = u30() |
| e0df6211 | 786 | for class_id in range(class_count): |
| ba552f54 | 787 | name_idx = u30() |
| e0df6211 PH |
788 | if name_idx == searched_idx: |
| 789 | # We found the class we're looking for! | |
| 790 | searched_class_id = class_id | |
| 0ca96d48 | 791 | u30() # super_name idx |
| ba552f54 | 792 | flags = read_byte() |
| e0df6211 | 793 | if flags & 0x08 != 0: # Protected namespace is present |
| 0ca96d48 | 794 | u30() # protected_ns_idx |
| ba552f54 | 795 | intrf_count = u30() |
| e0df6211 | 796 | for _c2 in range(intrf_count): |
| 0ca96d48 PH |
797 | u30() |
| 798 | u30() # iinit | |
| ba552f54 | 799 | trait_count = u30() |
| e0df6211 | 800 | for _c2 in range(trait_count): |
| 0ca96d48 | 801 | parse_traits_info() |
| e0df6211 PH |
802 | |
| 803 | if searched_class_id is None: | |
| 804 | raise ExtractorError(u'Target class %r not found' % | |
| 805 | TARGET_CLASSNAME) | |
| 806 | ||
| 807 | method_names = {} | |
| 808 | method_idxs = {} | |
| 809 | for class_id in range(class_count): | |
| 0ca96d48 | 810 | u30() # cinit |
| ba552f54 | 811 | trait_count = u30() |
| e0df6211 | 812 | for _c2 in range(trait_count): |
| ba552f54 | 813 | trait_methods = parse_traits_info() |
| e0df6211 PH |
814 | if class_id == searched_class_id: |
| 815 | method_names.update(trait_methods.items()) | |
| 816 | method_idxs.update(dict( | |
| 817 | (idx, name) | |
| 818 | for name, idx in trait_methods.items())) | |
| 819 | ||
| 820 | # Scripts | |
| ba552f54 | 821 | script_count = u30() |
| e0df6211 | 822 | for _c in range(script_count): |
| 0ca96d48 | 823 | u30() # init |
| ba552f54 | 824 | trait_count = u30() |
| e0df6211 | 825 | for _c2 in range(trait_count): |
| 0ca96d48 | 826 | parse_traits_info() |
| e0df6211 PH |
827 | |
| 828 | # Method bodies | |
| ba552f54 | 829 | method_body_count = u30() |
| e0df6211 PH |
830 | Method = collections.namedtuple('Method', ['code', 'local_count']) |
| 831 | methods = {} | |
| 832 | for _c in range(method_body_count): | |
| ba552f54 | 833 | method_idx = u30() |
| 0ca96d48 | 834 | u30() # max_stack |
| ba552f54 | 835 | local_count = u30() |
| 0ca96d48 PH |
836 | u30() # init_scope_depth |
| 837 | u30() # max_scope_depth | |
| ba552f54 PH |
838 | code_length = u30() |
| 839 | code = read_bytes(code_length) | |
| e0df6211 | 840 | if method_idx in method_idxs: |
| ba552f54 | 841 | m = Method(code, local_count) |
| e0df6211 | 842 | methods[method_idxs[method_idx]] = m |
| ba552f54 | 843 | exception_count = u30() |
| e0df6211 | 844 | for _c2 in range(exception_count): |
| 0ca96d48 PH |
845 | u30() # from |
| 846 | u30() # to | |
| 847 | u30() # target | |
| 848 | u30() # exc_type | |
| 849 | u30() # var_name | |
| ba552f54 | 850 | trait_count = u30() |
| e0df6211 | 851 | for _c2 in range(trait_count): |
| 0ca96d48 | 852 | parse_traits_info() |
| e0df6211 | 853 | |
| ba552f54 | 854 | assert p + code_reader.tell() == len(code_tag) |
| e0df6211 PH |
855 | assert len(methods) == len(method_idxs) |
| 856 | ||
| 857 | method_pyfunctions = {} | |
| 858 | ||
| 859 | def extract_function(func_name): | |
| 860 | if func_name in method_pyfunctions: | |
| 861 | return method_pyfunctions[func_name] | |
| 862 | if func_name not in methods: | |
| 863 | raise ExtractorError(u'Cannot find function %r' % func_name) | |
| 864 | m = methods[func_name] | |
| 865 | ||
| 866 | def resfunc(args): | |
| e0df6211 PH |
867 | registers = ['(this)'] + list(args) + [None] * m.local_count |
| 868 | stack = [] | |
| 869 | coder = io.BytesIO(m.code) | |
| 870 | while True: | |
| 871 | opcode = struct.unpack('!B', coder.read(1))[0] | |
| a7177865 | 872 | if opcode == 36: # pushbyte |
| e0df6211 PH |
873 | v = struct.unpack('!B', coder.read(1))[0] |
| 874 | stack.append(v) | |
| 875 | elif opcode == 44: # pushstring | |
| 876 | idx = u30(coder) | |
| 877 | stack.append(constant_strings[idx]) | |
| 878 | elif opcode == 48: # pushscope | |
| 879 | # We don't implement the scope register, so we'll just | |
| 880 | # ignore the popped value | |
| 881 | stack.pop() | |
| 882 | elif opcode == 70: # callproperty | |
| 883 | index = u30(coder) | |
| 884 | mname = multinames[index] | |
| 885 | arg_count = u30(coder) | |
| 886 | args = list(reversed( | |
| 887 | [stack.pop() for _ in range(arg_count)])) | |
| 888 | obj = stack.pop() | |
| 889 | if mname == u'split': | |
| 890 | assert len(args) == 1 | |
| 891 | assert isinstance(args[0], compat_str) | |
| 892 | assert isinstance(obj, compat_str) | |
| 893 | if args[0] == u'': | |
| 894 | res = list(obj) | |
| 895 | else: | |
| 896 | res = obj.split(args[0]) | |
| 897 | stack.append(res) | |
| a7177865 PH |
898 | elif mname == u'slice': |
| 899 | assert len(args) == 1 | |
| 900 | assert isinstance(args[0], int) | |
| 901 | assert isinstance(obj, list) | |
| 902 | res = obj[args[0]:] | |
| 903 | stack.append(res) | |
| 904 | elif mname == u'join': | |
| 905 | assert len(args) == 1 | |
| 906 | assert isinstance(args[0], compat_str) | |
| 907 | assert isinstance(obj, list) | |
| 908 | res = args[0].join(obj) | |
| 909 | stack.append(res) | |
| e0df6211 PH |
910 | elif mname in method_pyfunctions: |
| 911 | stack.append(method_pyfunctions[mname](args)) | |
| 912 | else: | |
| 913 | raise NotImplementedError( | |
| 914 | u'Unsupported property %r on %r' | |
| 915 | % (mname, obj)) | |
| a7177865 PH |
916 | elif opcode == 72: # returnvalue |
| 917 | res = stack.pop() | |
| 918 | return res | |
| 919 | elif opcode == 79: # callpropvoid | |
| 920 | index = u30(coder) | |
| 921 | mname = multinames[index] | |
| 922 | arg_count = u30(coder) | |
| 923 | args = list(reversed( | |
| 924 | [stack.pop() for _ in range(arg_count)])) | |
| 925 | obj = stack.pop() | |
| 926 | if mname == u'reverse': | |
| 927 | assert isinstance(obj, list) | |
| 928 | obj.reverse() | |
| 929 | else: | |
| 930 | raise NotImplementedError( | |
| 931 | u'Unsupported (void) property %r on %r' | |
| 932 | % (mname, obj)) | |
| e0df6211 PH |
933 | elif opcode == 93: # findpropstrict |
| 934 | index = u30(coder) | |
| 935 | mname = multinames[index] | |
| 936 | res = extract_function(mname) | |
| 937 | stack.append(res) | |
| 938 | elif opcode == 97: # setproperty | |
| 939 | index = u30(coder) | |
| 940 | value = stack.pop() | |
| 941 | idx = stack.pop() | |
| 942 | obj = stack.pop() | |
| 943 | assert isinstance(obj, list) | |
| 944 | assert isinstance(idx, int) | |
| 945 | obj[idx] = value | |
| 946 | elif opcode == 98: # getlocal | |
| 947 | index = u30(coder) | |
| 948 | stack.append(registers[index]) | |
| 949 | elif opcode == 99: # setlocal | |
| 950 | index = u30(coder) | |
| 951 | value = stack.pop() | |
| 952 | registers[index] = value | |
| 953 | elif opcode == 102: # getproperty | |
| 954 | index = u30(coder) | |
| 955 | pname = multinames[index] | |
| 956 | if pname == u'length': | |
| 957 | obj = stack.pop() | |
| 958 | assert isinstance(obj, list) | |
| 959 | stack.append(len(obj)) | |
| 960 | else: # Assume attribute access | |
| 961 | idx = stack.pop() | |
| 962 | assert isinstance(idx, int) | |
| 963 | obj = stack.pop() | |
| 964 | assert isinstance(obj, list) | |
| 965 | stack.append(obj[idx]) | |
| 966 | elif opcode == 128: # coerce | |
| 0ca96d48 | 967 | u30(coder) |
| e0df6211 PH |
968 | elif opcode == 133: # coerce_s |
| 969 | assert isinstance(stack[-1], (type(None), compat_str)) | |
| 970 | elif opcode == 164: # modulo | |
| 971 | value2 = stack.pop() | |
| 972 | value1 = stack.pop() | |
| 973 | res = value1 % value2 | |
| 974 | stack.append(res) | |
| a7177865 PH |
975 | elif opcode == 208: # getlocal_0 |
| 976 | stack.append(registers[0]) | |
| 977 | elif opcode == 209: # getlocal_1 | |
| 978 | stack.append(registers[1]) | |
| 979 | elif opcode == 210: # getlocal_2 | |
| 980 | stack.append(registers[2]) | |
| 981 | elif opcode == 211: # getlocal_3 | |
| 982 | stack.append(registers[3]) | |
| e0df6211 PH |
983 | elif opcode == 214: # setlocal_2 |
| 984 | registers[2] = stack.pop() | |
| 985 | elif opcode == 215: # setlocal_3 | |
| 986 | registers[3] = stack.pop() | |
| 987 | else: | |
| 988 | raise NotImplementedError( | |
| 989 | u'Unsupported opcode %d' % opcode) | |
| 990 | ||
| 991 | method_pyfunctions[func_name] = resfunc | |
| 992 | return resfunc | |
| 993 | ||
| 994 | initial_function = extract_function(u'decipher') | |
| 995 | return lambda s: initial_function([s]) | |
| 996 | ||
| 83799698 | 997 | def _decrypt_signature(self, s, video_id, player_url, age_gate=False): |
| 257a2501 | 998 | """Turn the encrypted s field into a working signature""" |
| 6b37f0be | 999 | |
| 83799698 | 1000 | if player_url is not None: |
| 9f9be844 PH |
1001 | if player_url.startswith(u'//'): |
| 1002 | player_url = u'https:' + player_url | |
| e0df6211 | 1003 | try: |
| 7f8ae73a PH |
1004 | player_id = (player_url, len(s)) |
| 1005 | if player_id not in self._player_cache: | |
| 83799698 | 1006 | func = self._extract_signature_function( |
| c4417ddb | 1007 | video_id, player_url, len(s) |
| e0df6211 | 1008 | ) |
| 7f8ae73a PH |
1009 | self._player_cache[player_id] = func |
| 1010 | func = self._player_cache[player_id] | |
| edf3e38e PH |
1011 | if self._downloader.params.get('youtube_print_sig_code'): |
| 1012 | self._print_sig_code(func, len(s)) | |
| 1013 | return func(s) | |
| 0ca96d48 | 1014 | except Exception: |
| e0df6211 | 1015 | tb = traceback.format_exc() |
| 83799698 PH |
1016 | self._downloader.report_warning( |
| 1017 | u'Automatic signature extraction failed: ' + tb) | |
| e0df6211 | 1018 | |
| d2d8f895 PH |
1019 | self._downloader.report_warning( |
| 1020 | u'Warning: Falling back to static signature algorithm') | |
| 920de7a2 | 1021 | |
| 2f2ffea9 PH |
1022 | return self._static_decrypt_signature( |
| 1023 | s, video_id, player_url, age_gate) | |
| e0df6211 | 1024 | |
| 2f2ffea9 | 1025 | def _static_decrypt_signature(self, s, video_id, player_url, age_gate): |
| e0df6211 PH |
1026 | if age_gate: |
| 1027 | # The videos with age protection use another player, so the | |
| 1028 | # algorithms can be different. | |
| 1029 | if len(s) == 86: | |
| 1030 | return s[2:63] + s[82] + s[64:82] + s[63] | |
| 1031 | ||
| bc4b9008 | 1032 | if len(s) == 93: |
| 1033 | return s[86:29:-1] + s[88] + s[28:5:-1] | |
| 1034 | elif len(s) == 92: | |
| 444b1165 | 1035 | return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83] |
| 38d025b3 JMF |
1036 | elif len(s) == 91: |
| 1037 | return s[84:27:-1] + s[86] + s[26:5:-1] | |
| 444b1165 JMF |
1038 | elif len(s) == 90: |
| 1039 | return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81] | |
| 8a9d86a2 | 1040 | elif len(s) == 89: |
| 1041 | return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1] | |
| 444b1165 | 1042 | elif len(s) == 88: |
| 3e223834 | 1043 | return s[7:28] + s[87] + s[29:45] + s[55] + s[46:55] + s[2] + s[56:87] + s[28] |
| be547e1d | 1044 | elif len(s) == 87: |
| 3a725669 | 1045 | return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:] |
| be547e1d | 1046 | elif len(s) == 86: |
| f2c327fd | 1047 | return s[80:72:-1] + s[16] + s[71:39:-1] + s[72] + s[38:16:-1] + s[82] + s[15::-1] |
| be547e1d | 1048 | elif len(s) == 85: |
| 6ae8ee3f | 1049 | return s[3:11] + s[0] + s[12:55] + s[84] + s[56:84] |
| be547e1d | 1050 | elif len(s) == 84: |
| 6f56389b | 1051 | return s[78:70:-1] + s[14] + s[69:37:-1] + s[70] + s[36:14:-1] + s[80] + s[:14][::-1] |
| be547e1d | 1052 | elif len(s) == 83: |
| 920de7a2 | 1053 | return s[80:63:-1] + s[0] + s[62:0:-1] + s[63] |
| be547e1d | 1054 | elif len(s) == 82: |
| c21315f2 | 1055 | return s[80:37:-1] + s[7] + s[36:7:-1] + s[0] + s[6:0:-1] + s[37] |
| be547e1d | 1056 | elif len(s) == 81: |
| aedd6bb9 | 1057 | 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 |
1058 | elif len(s) == 80: |
| 1059 | return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80] | |
| 5c468ca8 JMF |
1060 | elif len(s) == 79: |
| 1061 | 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 |
1062 | |
| 1063 | else: | |
| 1064 | raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) | |
| c5e8d7af | 1065 | |
| 1f343eaa | 1066 | def _get_available_subtitles(self, video_id, webpage): |
| de7f3446 | 1067 | try: |
| 7fad1c63 JMF |
1068 | sub_list = self._download_webpage( |
| 1069 | 'http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, | |
| 1070 | video_id, note=False) | |
| 1071 | except ExtractorError as err: | |
| de7f3446 JMF |
1072 | self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) |
| 1073 | return {} | |
| 1074 | lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
| 1075 | ||
| 1076 | sub_lang_list = {} | |
| 1077 | for l in lang_list: | |
| 1078 | lang = l[1] | |
| 1079 | params = compat_urllib_parse.urlencode({ | |
| 1080 | 'lang': lang, | |
| 1081 | 'v': video_id, | |
| ca715127 | 1082 | 'fmt': self._downloader.params.get('subtitlesformat', 'srt'), |
| 8eddf3e9 | 1083 | 'name': l[0].encode('utf-8'), |
| de7f3446 JMF |
1084 | }) |
| 1085 | url = u'http://www.youtube.com/api/timedtext?' + params | |
| 1086 | sub_lang_list[lang] = url | |
| 1087 | if not sub_lang_list: | |
| 1088 | self._downloader.report_warning(u'video doesn\'t have subtitles') | |
| 1089 | return {} | |
| 1090 | return sub_lang_list | |
| 1091 | ||
| 055e6f36 | 1092 | def _get_available_automatic_caption(self, video_id, webpage): |
| de7f3446 JMF |
1093 | """We need the webpage for getting the captions url, pass it as an |
| 1094 | argument to speed up the process.""" | |
| ca715127 | 1095 | sub_format = self._downloader.params.get('subtitlesformat', 'srt') |
| de7f3446 JMF |
1096 | self.to_screen(u'%s: Looking for automatic captions' % video_id) |
| 1097 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
| 055e6f36 | 1098 | err_msg = u'Couldn\'t find automatic captions for %s' % video_id |
| de7f3446 JMF |
1099 | if mobj is None: |
| 1100 | self._downloader.report_warning(err_msg) | |
| 1101 | return {} | |
| 1102 | player_config = json.loads(mobj.group(1)) | |
| 1103 | try: | |
| 1104 | args = player_config[u'args'] | |
| 1105 | caption_url = args[u'ttsurl'] | |
| 1106 | timestamp = args[u'timestamp'] | |
| 055e6f36 JMF |
1107 | # We get the available subtitles |
| 1108 | list_params = compat_urllib_parse.urlencode({ | |
| 1109 | 'type': 'list', | |
| 1110 | 'tlangs': 1, | |
| 1111 | 'asrs': 1, | |
| de7f3446 | 1112 | }) |
| 055e6f36 | 1113 | list_url = caption_url + '&' + list_params |
| e26f8712 | 1114 | caption_list = self._download_xml(list_url, video_id) |
| e3dc22ca | 1115 | original_lang_node = caption_list.find('track') |
| f6a54188 | 1116 | if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' : |
| e3dc22ca JMF |
1117 | self._downloader.report_warning(u'Video doesn\'t have automatic captions') |
| 1118 | return {} | |
| 1119 | original_lang = original_lang_node.attrib['lang_code'] | |
| 055e6f36 JMF |
1120 | |
| 1121 | sub_lang_list = {} | |
| 1122 | for lang_node in caption_list.findall('target'): | |
| 1123 | sub_lang = lang_node.attrib['lang_code'] | |
| 1124 | params = compat_urllib_parse.urlencode({ | |
| 1125 | 'lang': original_lang, | |
| 1126 | 'tlang': sub_lang, | |
| 1127 | 'fmt': sub_format, | |
| 1128 | 'ts': timestamp, | |
| 1129 | 'kind': 'asr', | |
| 1130 | }) | |
| 1131 | sub_lang_list[sub_lang] = caption_url + '&' + params | |
| 1132 | return sub_lang_list | |
| de7f3446 JMF |
1133 | # An extractor error can be raise by the download process if there are |
| 1134 | # no automatic captions but there are subtitles | |
| 1135 | except (KeyError, ExtractorError): | |
| 1136 | self._downloader.report_warning(err_msg) | |
| 1137 | return {} | |
| 1138 | ||
| c5e8d7af PH |
1139 | def _extract_id(self, url): |
| 1140 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
| 1141 | if mobj is None: | |
| 1142 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 1143 | video_id = mobj.group(2) | |
| 1144 | return video_id | |
| 1145 | ||
| 1d043b93 JMF |
1146 | def _get_video_url_list(self, url_map): |
| 1147 | """ | |
| 1148 | Transform a dictionary in the format {itag:url} to a list of (itag, url) | |
| 1149 | with the requested formats. | |
| 1150 | """ | |
| 4ea3be0a | 1151 | existing_formats = [x for x in self._available_formats if x in url_map] |
| 1d043b93 JMF |
1152 | if len(existing_formats) == 0: |
| 1153 | raise ExtractorError(u'no known formats available for video') | |
| 4ea3be0a | 1154 | video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats |
| 1155 | video_url_list.reverse() # order worst to best | |
| 1d043b93 JMF |
1156 | return video_url_list |
| 1157 | ||
| 1158 | def _extract_from_m3u8(self, manifest_url, video_id): | |
| 1159 | url_map = {} | |
| 1160 | def _get_urls(_manifest): | |
| 1161 | lines = _manifest.split('\n') | |
| 1162 | urls = filter(lambda l: l and not l.startswith('#'), | |
| 1163 | lines) | |
| 1164 | return urls | |
| 1165 | manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') | |
| 1166 | formats_urls = _get_urls(manifest) | |
| 1167 | for format_url in formats_urls: | |
| 890f62e8 | 1168 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') |
| 1d043b93 JMF |
1169 | url_map[itag] = format_url |
| 1170 | return url_map | |
| 1171 | ||
| 1fb07d10 JG |
1172 | def _extract_annotations(self, video_id): |
| 1173 | url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id | |
| 1174 | return self._download_webpage(url, video_id, note=u'Searching for annotations.', errnote=u'Unable to download video annotations.') | |
| 1175 | ||
| c5e8d7af PH |
1176 | def _real_extract(self, url): |
| 1177 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter | |
| 1178 | mobj = re.search(self._NEXT_URL_RE, url) | |
| 1179 | if mobj: | |
| 1180 | url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
| 1181 | video_id = self._extract_id(url) | |
| 1182 | ||
| 1183 | # Get video webpage | |
| c5e8d7af | 1184 | url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id |
| 336c3a69 | 1185 | video_webpage = self._download_webpage(url, video_id) |
| c5e8d7af PH |
1186 | |
| 1187 | # Attempt to extract SWF player URL | |
| e0df6211 | 1188 | mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) |
| c5e8d7af PH |
1189 | if mobj is not None: |
| 1190 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
| 1191 | else: | |
| 1192 | player_url = None | |
| 1193 | ||
| 1194 | # Get video info | |
| 1195 | self.report_video_info_webpage_download(video_id) | |
| c108eb73 JMF |
1196 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
| 1197 | self.report_age_confirmation() | |
| 1198 | age_gate = True | |
| 1199 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
| 1200 | # this can be viewed without login into Youtube | |
| 1201 | data = compat_urllib_parse.urlencode({'video_id': video_id, | |
| fccd3771 | 1202 | 'el': 'player_embedded', |
| c108eb73 JMF |
1203 | 'gl': 'US', |
| 1204 | 'hl': 'en', | |
| 1205 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
| 1206 | 'asv': 3, | |
| 1207 | 'sts':'1588', | |
| 1208 | }) | |
| 1209 | video_info_url = 'https://www.youtube.com/get_video_info?' + data | |
| c5e8d7af PH |
1210 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
| 1211 | note=False, | |
| 1212 | errnote='unable to download video info webpage') | |
| 1213 | video_info = compat_parse_qs(video_info_webpage) | |
| c108eb73 JMF |
1214 | else: |
| 1215 | age_gate = False | |
| 1216 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
| 1217 | video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
| 1218 | % (video_id, el_type)) | |
| 1219 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
| 1220 | note=False, | |
| 1221 | errnote='unable to download video info webpage') | |
| 1222 | video_info = compat_parse_qs(video_info_webpage) | |
| 1223 | if 'token' in video_info: | |
| 1224 | break | |
| c5e8d7af PH |
1225 | if 'token' not in video_info: |
| 1226 | if 'reason' in video_info: | |
| 9a82b238 | 1227 | raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True) |
| c5e8d7af PH |
1228 | else: |
| 1229 | raise ExtractorError(u'"token" parameter not in video info for unknown reason') | |
| 1230 | ||
| 1d699755 PH |
1231 | if 'view_count' in video_info: |
| 1232 | view_count = int(video_info['view_count'][0]) | |
| 1233 | else: | |
| 1234 | view_count = None | |
| 1235 | ||
| c5e8d7af PH |
1236 | # Check for "rental" videos |
| 1237 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
| 1238 | raise ExtractorError(u'"rental" videos not supported') | |
| 1239 | ||
| 1240 | # Start extracting information | |
| 1241 | self.report_information_extraction(video_id) | |
| 1242 | ||
| 1243 | # uploader | |
| 1244 | if 'author' not in video_info: | |
| 1245 | raise ExtractorError(u'Unable to extract uploader name') | |
| 1246 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
| 1247 | ||
| 1248 | # uploader_id | |
| 1249 | video_uploader_id = None | |
| 1250 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
| 1251 | if mobj is not None: | |
| 1252 | video_uploader_id = mobj.group(1) | |
| 1253 | else: | |
| 1254 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
| 1255 | ||
| 1256 | # title | |
| a8c6b241 PH |
1257 | if 'title' in video_info: |
| 1258 | video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) | |
| 1259 | else: | |
| 1260 | self._downloader.report_warning(u'Unable to extract video title') | |
| 1261 | video_title = u'_' | |
| c5e8d7af PH |
1262 | |
| 1263 | # thumbnail image | |
| 7763b04e JMF |
1264 | # We try first to get a high quality image: |
| 1265 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
| 1266 | video_webpage, re.DOTALL) | |
| 1267 | if m_thumb is not None: | |
| 1268 | video_thumbnail = m_thumb.group(1) | |
| 1269 | elif 'thumbnail_url' not in video_info: | |
| c5e8d7af | 1270 | self._downloader.report_warning(u'unable to extract video thumbnail') |
| f490e77e | 1271 | video_thumbnail = None |
| c5e8d7af PH |
1272 | else: # don't panic if we can't find it |
| 1273 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
| 1274 | ||
| 1275 | # upload date | |
| 1276 | upload_date = None | |
| 1277 | mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL) | |
| 1278 | if mobj is not None: | |
| 1279 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
| 1280 | upload_date = unified_strdate(upload_date) | |
| 1281 | ||
| 1282 | # description | |
| 1283 | video_description = get_element_by_id("eow-description", video_webpage) | |
| 1284 | if video_description: | |
| 27dcce19 PH |
1285 | video_description = re.sub(r'''(?x) |
| 1286 | <a\s+ | |
| 1287 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
| 1288 | title="([^"]+)"\s+ | |
| 1289 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
| 1290 | class="yt-uix-redirect-link"\s*> | |
| 1291 | [^<]+ | |
| 1292 | </a> | |
| 1293 | ''', r'\1', video_description) | |
| c5e8d7af PH |
1294 | video_description = clean_html(video_description) |
| 1295 | else: | |
| 1296 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
| 1297 | if fd_mobj: | |
| 1298 | video_description = unescapeHTML(fd_mobj.group(1)) | |
| 1299 | else: | |
| 1300 | video_description = u'' | |
| 1301 | ||
| 336c3a69 | 1302 | def _extract_count(klass): |
| 46374a56 PH |
1303 | count = self._search_regex( |
| 1304 | r'class="%s">([\d,]+)</span>' % re.escape(klass), | |
| 1305 | video_webpage, klass, default=None) | |
| 336c3a69 JMF |
1306 | if count is not None: |
| 1307 | return int(count.replace(',', '')) | |
| 1308 | return None | |
| 1309 | like_count = _extract_count(u'likes-count') | |
| 1310 | dislike_count = _extract_count(u'dislikes-count') | |
| 1311 | ||
| c5e8d7af | 1312 | # subtitles |
| d82134c3 | 1313 | video_subtitles = self.extract_subtitles(video_id, video_webpage) |
| c5e8d7af | 1314 | |
| c5e8d7af | 1315 | if self._downloader.params.get('listsubtitles', False): |
| d665f8d3 | 1316 | self._list_available_subtitles(video_id, video_webpage) |
| c5e8d7af PH |
1317 | return |
| 1318 | ||
| 1319 | if 'length_seconds' not in video_info: | |
| 1320 | self._downloader.report_warning(u'unable to extract video duration') | |
| b466b702 | 1321 | video_duration = None |
| c5e8d7af | 1322 | else: |
| b466b702 | 1323 | video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) |
| c5e8d7af | 1324 | |
| 1fb07d10 JG |
1325 | # annotations |
| 1326 | video_annotations = None | |
| 1327 | if self._downloader.params.get('writeannotations', False): | |
| 1328 | video_annotations = self._extract_annotations(video_id) | |
| 1329 | ||
| c5e8d7af | 1330 | # Decide which formats to download |
| c5e8d7af PH |
1331 | |
| 1332 | try: | |
| 1333 | mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) | |
| 50be92c1 PH |
1334 | if not mobj: |
| 1335 | raise ValueError('Could not find vevo ID') | |
| c5e8d7af PH |
1336 | info = json.loads(mobj.group(1)) |
| 1337 | args = info['args'] | |
| 7ce7e394 JMF |
1338 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
| 1339 | # this signatures are encrypted | |
| 44d46655 | 1340 | if 'url_encoded_fmt_stream_map' not in args: |
| f10503db | 1341 | raise ValueError(u'No stream_map present') # caught below |
| 00fe14fc JMF |
1342 | re_signature = re.compile(r'[&,]s=') |
| 1343 | m_s = re_signature.search(args['url_encoded_fmt_stream_map']) | |
| 7ce7e394 JMF |
1344 | if m_s is not None: |
| 1345 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
| c5e8d7af | 1346 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
| 00fe14fc | 1347 | m_s = re_signature.search(args.get('adaptive_fmts', u'')) |
| b7a68384 | 1348 | if m_s is not None: |
| 00fe14fc JMF |
1349 | if 'adaptive_fmts' in video_info: |
| 1350 | video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts'] | |
| 37b6d5f6 | 1351 | else: |
| 00fe14fc | 1352 | video_info['adaptive_fmts'] = [args['adaptive_fmts']] |
| c5e8d7af PH |
1353 | except ValueError: |
| 1354 | pass | |
| 1355 | ||
| 1356 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
| 1357 | self.report_rtmp_download() | |
| 1358 | video_url_list = [(None, video_info['conn'][0])] | |
| 00fe14fc JMF |
1359 | elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1: |
| 1360 | encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0] | |
| 1361 | if 'rtmpe%3Dyes' in encoded_url_map: | |
| a7055eb9 | 1362 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) |
| c5e8d7af | 1363 | url_map = {} |
| 00fe14fc | 1364 | for url_data_str in encoded_url_map.split(','): |
| c5e8d7af PH |
1365 | url_data = compat_parse_qs(url_data_str) |
| 1366 | if 'itag' in url_data and 'url' in url_data: | |
| 1367 | url = url_data['url'][0] | |
| 1368 | if 'sig' in url_data: | |
| 1369 | url += '&signature=' + url_data['sig'][0] | |
| 1370 | elif 's' in url_data: | |
| e0df6211 | 1371 | encrypted_sig = url_data['s'][0] |
| 769fda3c | 1372 | if self._downloader.params.get('verbose'): |
| c108eb73 | 1373 | if age_gate: |
| bdde940e PH |
1374 | if player_url is None: |
| 1375 | player_version = 'unknown' | |
| 1376 | else: | |
| 1377 | player_version = self._search_regex( | |
| 1378 | r'-(.+)\.swf$', player_url, | |
| 1379 | u'flash player', fatal=False) | |
| e0df6211 | 1380 | player_desc = 'flash player %s' % player_version |
| c108eb73 | 1381 | else: |
| 83799698 PH |
1382 | player_version = self._search_regex( |
| 1383 | r'html5player-(.+?)\.js', video_webpage, | |
| c108eb73 | 1384 | 'html5 player', fatal=False) |
| e0df6211 PH |
1385 | player_desc = u'html5 player %s' % player_version |
| 1386 | ||
| 1387 | parts_sizes = u'.'.join(compat_str(len(part)) for part in encrypted_sig.split('.')) | |
| 5a76c651 | 1388 | self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % |
| e0df6211 PH |
1389 | (len(encrypted_sig), parts_sizes, url_data['itag'][0], player_desc)) |
| 1390 | ||
| 83799698 | 1391 | if not age_gate: |
| e0df6211 PH |
1392 | jsplayer_url_json = self._search_regex( |
| 1393 | r'"assets":.+?"js":\s*("[^"]+")', | |
| 1394 | video_webpage, u'JS player URL') | |
| 83799698 | 1395 | player_url = json.loads(jsplayer_url_json) |
| e0df6211 | 1396 | |
| 83799698 PH |
1397 | signature = self._decrypt_signature( |
| 1398 | encrypted_sig, video_id, player_url, age_gate) | |
| c5e8d7af PH |
1399 | url += '&signature=' + signature |
| 1400 | if 'ratebypass' not in url: | |
| 1401 | url += '&ratebypass=yes' | |
| 1402 | url_map[url_data['itag'][0]] = url | |
| 1d043b93 | 1403 | video_url_list = self._get_video_url_list(url_map) |
| 1d043b93 JMF |
1404 | elif video_info.get('hlsvp'): |
| 1405 | manifest_url = video_info['hlsvp'][0] | |
| 1406 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
| 1407 | video_url_list = self._get_video_url_list(url_map) | |
| c5e8d7af | 1408 | else: |
| 9abb3204 | 1409 | raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') |
| c5e8d7af | 1410 | |
| 4ea3be0a | 1411 | formats = [] |
| 600cc1a4 | 1412 | for itag, video_real_url in video_url_list: |
| c5e8d7af | 1413 | # Extension |
| 600cc1a4 | 1414 | video_extension = self._video_extensions.get(itag, 'flv') |
| dbd1988e | 1415 | resolution = self._video_dimensions.get(itag, {}).get('display') |
| 1416 | width = self._video_dimensions.get(itag, {}).get('width') | |
| 1417 | height = self._video_dimensions.get(itag, {}).get('height') | |
| 1418 | note = self._special_itags.get(itag) | |
| c5e8d7af | 1419 | |
| 600cc1a4 | 1420 | video_format = '{0} - {1}{2}'.format(itag if itag else video_extension, |
| dbd1988e | 1421 | '%dx%d' % (width, height) if width is not None and height is not None else (resolution if resolution is not None else '???'), |
| 600cc1a4 | 1422 | ' ('+self._special_itags[itag]+')' if itag in self._special_itags else '') |
| c5e8d7af | 1423 | |
| 4ea3be0a | 1424 | formats.append({ |
| 1425 | 'url': video_real_url, | |
| 1426 | 'ext': video_extension, | |
| 1427 | 'format': video_format, | |
| 1428 | 'format_id': itag, | |
| 1429 | 'player_url': player_url, | |
| 1430 | '_resolution': resolution, | |
| dbd1988e | 1431 | 'width': width, |
| 1432 | 'height': height, | |
| 4ea3be0a | 1433 | 'format_note': note, |
| c5e8d7af | 1434 | }) |
| d80044c2 | 1435 | |
| 4bcc7bd1 | 1436 | self._sort_formats(formats) |
| 4ea3be0a | 1437 | |
| 1438 | return { | |
| 1439 | 'id': video_id, | |
| 1440 | 'uploader': video_uploader, | |
| 1441 | 'uploader_id': video_uploader_id, | |
| 1442 | 'upload_date': upload_date, | |
| 1443 | 'title': video_title, | |
| 1444 | 'thumbnail': video_thumbnail, | |
| 1445 | 'description': video_description, | |
| 1446 | 'subtitles': video_subtitles, | |
| 1447 | 'duration': video_duration, | |
| 1448 | 'age_limit': 18 if age_gate else 0, | |
| 1449 | 'annotations': video_annotations, | |
| 1450 | 'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id, | |
| 1451 | 'view_count': view_count, | |
| 1452 | 'like_count': like_count, | |
| 1453 | 'dislike_count': dislike_count, | |
| 1454 | 'formats': formats, | |
| 1455 | } | |
| c5e8d7af | 1456 | |
| 880e1c52 | 1457 | class YoutubePlaylistIE(YoutubeBaseInfoExtractor): |
| 0f818663 | 1458 | IE_DESC = u'YouTube.com playlists' |
| c5e8d7af PH |
1459 | _VALID_URL = r"""(?: |
| 1460 | (?:https?://)? | |
| 1461 | (?:\w+\.)? | |
| 1462 | youtube\.com/ | |
| 1463 | (?: | |
| 1464 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
| 1465 | \? (?:.*?&)*? (?:p|a|list)= | |
| 1466 | | p/ | |
| 1467 | ) | |
| 715c8e7b | 1468 | ((?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}) |
| c5e8d7af PH |
1469 | .* |
| 1470 | | | |
| 715c8e7b | 1471 | ((?:PL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,}) |
| c5e8d7af | 1472 | )""" |
| dcbb4580 JMF |
1473 | _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s' |
| 1474 | _MORE_PAGES_INDICATOR = r'data-link-type="next"' | |
| 6e47b51e | 1475 | _VIDEO_RE = r'href="/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' |
| c5e8d7af PH |
1476 | IE_NAME = u'youtube:playlist' |
| 1477 | ||
| 1478 | @classmethod | |
| 1479 | def suitable(cls, url): | |
| 1480 | """Receives a URL and returns True if suitable for this IE.""" | |
| 1481 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
| 1482 | ||
| 880e1c52 JMF |
1483 | def _real_initialize(self): |
| 1484 | self._login() | |
| 1485 | ||
| 652cdaa2 JMF |
1486 | def _ids_to_results(self, ids): |
| 1487 | return [self.url_result(vid_id, 'Youtube', video_id=vid_id) | |
| 1488 | for vid_id in ids] | |
| 1489 | ||
| 1490 | def _extract_mix(self, playlist_id): | |
| 1491 | # The mixes are generated from a a single video | |
| 1492 | # the id of the playlist is just 'RD' + video_id | |
| 7d4afc55 | 1493 | url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id) |
| 652cdaa2 | 1494 | webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix') |
| 76d1700b JMF |
1495 | title_span = (get_element_by_attribute('class', 'title long-title', webpage) or |
| 1496 | get_element_by_attribute('class', 'title ', webpage)) | |
| 1497 | title = clean_html(title_span) | |
| 652cdaa2 JMF |
1498 | video_re = r'data-index="\d+".*?href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s' % re.escape(playlist_id) |
| 1499 | ids = orderedSet(re.findall(video_re, webpage)) | |
| 1500 | url_results = self._ids_to_results(ids) | |
| 1501 | ||
| 1502 | return self.playlist_result(url_results, playlist_id, title) | |
| 1503 | ||
| c5e8d7af PH |
1504 | def _real_extract(self, url): |
| 1505 | # Extract playlist id | |
| 1506 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
| 1507 | if mobj is None: | |
| 1508 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 47192f92 FV |
1509 | playlist_id = mobj.group(1) or mobj.group(2) |
| 1510 | ||
| 1511 | # Check if it's a video-specific URL | |
| 7c61bd36 | 1512 | query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) |
| 47192f92 FV |
1513 | if 'v' in query_dict: |
| 1514 | video_id = query_dict['v'][0] | |
| 1515 | if self._downloader.params.get('noplaylist'): | |
| 1516 | self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id) | |
| 7012b23c | 1517 | return self.url_result(video_id, 'Youtube', video_id=video_id) |
| 47192f92 FV |
1518 | else: |
| 1519 | self.to_screen(u'Downloading playlist PL%s - add --no-playlist to just download video %s' % (playlist_id, video_id)) | |
| c5e8d7af | 1520 | |
| 7d4afc55 | 1521 | if playlist_id.startswith('RD'): |
| 652cdaa2 JMF |
1522 | # Mixes require a custom extraction process |
| 1523 | return self._extract_mix(playlist_id) | |
| 0a688bc0 JMF |
1524 | if playlist_id.startswith('TL'): |
| 1525 | raise ExtractorError(u'For downloading YouTube.com top lists, use ' | |
| 1526 | u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) | |
| 652cdaa2 | 1527 | |
| dcbb4580 JMF |
1528 | # Extract the video ids from the playlist pages |
| 1529 | ids = [] | |
| c5e8d7af | 1530 | |
| 755eb032 | 1531 | for page_num in itertools.count(1): |
| dcbb4580 | 1532 | url = self._TEMPLATE_URL % (playlist_id, page_num) |
| c5e8d7af | 1533 | page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) |
| 6e47b51e JMF |
1534 | matches = re.finditer(self._VIDEO_RE, page) |
| 1535 | # We remove the duplicates and the link with index 0 | |
| 1536 | # (it's not the first video of the playlist) | |
| 1537 | new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') | |
| dcbb4580 | 1538 | ids.extend(new_ids) |
| c5e8d7af | 1539 | |
| dcbb4580 | 1540 | if re.search(self._MORE_PAGES_INDICATOR, page) is None: |
| c5e8d7af PH |
1541 | break |
| 1542 | ||
| dcbb4580 | 1543 | playlist_title = self._og_search_title(page) |
| c5e8d7af | 1544 | |
| 652cdaa2 | 1545 | url_results = self._ids_to_results(ids) |
| dcbb4580 | 1546 | return self.playlist_result(url_results, playlist_id, playlist_title) |
| c5e8d7af PH |
1547 | |
| 1548 | ||
| 0a688bc0 JMF |
1549 | class YoutubeTopListIE(YoutubePlaylistIE): |
| 1550 | IE_NAME = u'youtube:toplist' | |
| 1551 | IE_DESC = (u'YouTube.com top lists, "yttoplist:{channel}:{list title}"' | |
| 1552 | u' (Example: "yttoplist:music:Top Tracks")') | |
| 1553 | _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$' | |
| 1554 | ||
| 1555 | def _real_extract(self, url): | |
| 1556 | mobj = re.match(self._VALID_URL, url) | |
| 1557 | channel = mobj.group('chann') | |
| 1558 | title = mobj.group('title') | |
| 1559 | query = compat_urllib_parse.urlencode({'title': title}) | |
| 1560 | playlist_re = 'href="([^"]+?%s[^"]+?)"' % re.escape(query) | |
| 1561 | channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title) | |
| 1562 | link = self._html_search_regex(playlist_re, channel_page, u'list') | |
| 1563 | url = compat_urlparse.urljoin('https://www.youtube.com/', link) | |
| 1564 | ||
| 1565 | video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"' | |
| 1566 | ids = [] | |
| 1567 | # sometimes the webpage doesn't contain the videos | |
| 1568 | # retry until we get them | |
| 1569 | for i in itertools.count(0): | |
| 1570 | msg = u'Downloading Youtube mix' | |
| 1571 | if i > 0: | |
| 1572 | msg += ', retry #%d' % i | |
| 1573 | webpage = self._download_webpage(url, title, msg) | |
| 1574 | ids = orderedSet(re.findall(video_re, webpage)) | |
| 1575 | if ids: | |
| 1576 | break | |
| 1577 | url_results = self._ids_to_results(ids) | |
| 1578 | return self.playlist_result(url_results, playlist_title=title) | |
| 1579 | ||
| 1580 | ||
| c5e8d7af | 1581 | class YoutubeChannelIE(InfoExtractor): |
| 0f818663 | 1582 | IE_DESC = u'YouTube.com channels' |
| c5e8d7af | 1583 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
| c5e8d7af | 1584 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' |
| 252580c5 | 1585 | _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 |
1586 | IE_NAME = u'youtube:channel' |
| 1587 | ||
| 1588 | def extract_videos_from_page(self, page): | |
| 1589 | ids_in_page = [] | |
| 1590 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
| 1591 | if mobj.group(1) not in ids_in_page: | |
| 1592 | ids_in_page.append(mobj.group(1)) | |
| 1593 | return ids_in_page | |
| 1594 | ||
| 1595 | def _real_extract(self, url): | |
| 1596 | # Extract channel id | |
| 1597 | mobj = re.match(self._VALID_URL, url) | |
| 1598 | if mobj is None: | |
| 1599 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 1600 | ||
| 1601 | # Download channel page | |
| 1602 | channel_id = mobj.group(1) | |
| 1603 | video_ids = [] | |
| b9643eed JMF |
1604 | url = 'https://www.youtube.com/channel/%s/videos' % channel_id |
| 1605 | channel_page = self._download_webpage(url, channel_id) | |
| 31812a9e PH |
1606 | autogenerated = re.search(r'''(?x) |
| 1607 | class="[^"]*?(?: | |
| 1608 | channel-header-autogenerated-label| | |
| 1609 | yt-channel-title-autogenerated | |
| 1610 | )[^"]*"''', channel_page) is not None | |
| c5e8d7af | 1611 | |
| b9643eed JMF |
1612 | if autogenerated: |
| 1613 | # The videos are contained in a single page | |
| 1614 | # the ajax pages can't be used, they are empty | |
| 1615 | video_ids = self.extract_videos_from_page(channel_page) | |
| 1616 | else: | |
| 1617 | # Download all channel pages using the json-based channel_ajax query | |
| 1618 | for pagenum in itertools.count(1): | |
| 1619 | url = self._MORE_PAGES_URL % (pagenum, channel_id) | |
| 1620 | page = self._download_webpage(url, channel_id, | |
| 1621 | u'Downloading page #%s' % pagenum) | |
| 1622 | ||
| 1623 | page = json.loads(page) | |
| 1624 | ||
| 1625 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
| 1626 | video_ids.extend(ids_in_page) | |
| 1627 | ||
| 1628 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
| 1629 | break | |
| c5e8d7af PH |
1630 | |
| 1631 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
| 1632 | ||
| 7012b23c PH |
1633 | url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id) |
| 1634 | for video_id in video_ids] | |
| 1635 | return self.playlist_result(url_entries, channel_id) | |
| c5e8d7af PH |
1636 | |
| 1637 | ||
| 1638 | class YoutubeUserIE(InfoExtractor): | |
| 0f818663 | 1639 | IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' |
| 57da92b7 | 1640 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)' |
| c5e8d7af PH |
1641 | _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' |
| 1642 | _GDATA_PAGE_SIZE = 50 | |
| fd9cf738 | 1643 | _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' |
| c5e8d7af PH |
1644 | IE_NAME = u'youtube:user' |
| 1645 | ||
| e3ea4790 | 1646 | @classmethod |
| f4b05232 | 1647 | def suitable(cls, url): |
| e3ea4790 JMF |
1648 | # Don't return True if the url can be extracted with other youtube |
| 1649 | # extractor, the regex would is too permissive and it would match. | |
| 1650 | other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) | |
| 1651 | if any(ie.suitable(url) for ie in other_ies): return False | |
| f4b05232 JMF |
1652 | else: return super(YoutubeUserIE, cls).suitable(url) |
| 1653 | ||
| c5e8d7af PH |
1654 | def _real_extract(self, url): |
| 1655 | # Extract username | |
| 1656 | mobj = re.match(self._VALID_URL, url) | |
| 1657 | if mobj is None: | |
| 1658 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 1659 | ||
| 1660 | username = mobj.group(1) | |
| 1661 | ||
| 1662 | # Download video ids using YouTube Data API. Result size per | |
| 1663 | # query is limited (currently to 50 videos) so we need to query | |
| 1664 | # page by page until there are no video ids - it means we got | |
| 1665 | # all of them. | |
| 1666 | ||
| e302f9ce | 1667 | url_results = [] |
| c5e8d7af | 1668 | |
| 755eb032 | 1669 | for pagenum in itertools.count(0): |
| c5e8d7af PH |
1670 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 |
| 1671 | ||
| 1672 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
| 1673 | page = self._download_webpage(gdata_url, username, | |
| 1674 | u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) | |
| 1675 | ||
| fd9cf738 JMF |
1676 | try: |
| 1677 | response = json.loads(page) | |
| 1678 | except ValueError as err: | |
| 1679 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
| 71c82637 JMF |
1680 | if 'entry' not in response['feed']: |
| 1681 | # Number of videos is a multiple of self._MAX_RESULTS | |
| 1682 | break | |
| fd9cf738 | 1683 | |
| c5e8d7af | 1684 | # Extract video identifiers |
| e302f9ce PH |
1685 | entries = response['feed']['entry'] |
| 1686 | for entry in entries: | |
| 1687 | title = entry['title']['$t'] | |
| 1688 | video_id = entry['id']['$t'].split('/')[-1] | |
| 1689 | url_results.append({ | |
| 1690 | '_type': 'url', | |
| 1691 | 'url': video_id, | |
| 1692 | 'ie_key': 'Youtube', | |
| 1693 | 'id': 'video_id', | |
| 1694 | 'title': title, | |
| 1695 | }) | |
| c5e8d7af PH |
1696 | |
| 1697 | # A little optimization - if current page is not | |
| 1698 | # "full", ie. does not contain PAGE_SIZE video ids then | |
| 1699 | # we can assume that this page is the last one - there | |
| 1700 | # are no more ids on further pages - no need to query | |
| 1701 | # again. | |
| 1702 | ||
| e302f9ce | 1703 | if len(entries) < self._GDATA_PAGE_SIZE: |
| c5e8d7af PH |
1704 | break |
| 1705 | ||
| 7012b23c PH |
1706 | return self.playlist_result(url_results, playlist_title=username) |
| 1707 | ||
| b05654f0 PH |
1708 | |
| 1709 | class YoutubeSearchIE(SearchInfoExtractor): | |
| 0f818663 | 1710 | IE_DESC = u'YouTube.com searches' |
| b05654f0 PH |
1711 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' |
| 1712 | _MAX_RESULTS = 1000 | |
| 1713 | IE_NAME = u'youtube:search' | |
| 1714 | _SEARCH_KEY = 'ytsearch' | |
| 1715 | ||
| b05654f0 PH |
1716 | def _get_n_results(self, query, n): |
| 1717 | """Get a specified number of results for a query""" | |
| 1718 | ||
| 1719 | video_ids = [] | |
| 1720 | pagenum = 0 | |
| 1721 | limit = n | |
| 1722 | ||
| 1723 | while (50 * pagenum) < limit: | |
| b05654f0 | 1724 | result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) |
| 7cc3570e PH |
1725 | data_json = self._download_webpage( |
| 1726 | result_url, video_id=u'query "%s"' % query, | |
| 1727 | note=u'Downloading page %s' % (pagenum + 1), | |
| 1728 | errnote=u'Unable to download API page') | |
| 1729 | data = json.loads(data_json) | |
| 1730 | api_response = data['data'] | |
| 1731 | ||
| 1732 | if 'items' not in api_response: | |
| b05654f0 PH |
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] | |
| 7012b23c PH |
1743 | videos = [self.url_result(video_id, 'Youtube', video_id=video_id) |
| 1744 | for video_id in video_ids] | |
| b05654f0 | 1745 | return self.playlist_result(videos, query) |
| 75dff0ee | 1746 | |
| a3dd9248 | 1747 | class YoutubeSearchDateIE(YoutubeSearchIE): |
| cb7fb546 | 1748 | IE_NAME = YoutubeSearchIE.IE_NAME + ':date' |
| a3dd9248 CM |
1749 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published' |
| 1750 | _SEARCH_KEY = 'ytsearchdate' | |
| 08fb86c4 | 1751 | IE_DESC = u'YouTube.com searches, newest videos first' |
| 75dff0ee JMF |
1752 | |
| 1753 | class YoutubeShowIE(InfoExtractor): | |
| 0f818663 | 1754 | IE_DESC = u'YouTube.com (multi-season) shows' |
| 75dff0ee JMF |
1755 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
| 1756 | IE_NAME = u'youtube:show' | |
| 1757 | ||
| 1758 | def _real_extract(self, url): | |
| 1759 | mobj = re.match(self._VALID_URL, url) | |
| 1760 | show_name = mobj.group(1) | |
| 1761 | webpage = self._download_webpage(url, show_name, u'Downloading show webpage') | |
| 1762 | # There's one playlist for each season of the show | |
| 1763 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
| 1764 | self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) | |
| 1765 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] | |
| 04cc9617 JMF |
1766 | |
| 1767 | ||
| b2e8bc1b | 1768 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): |
| d7ae0639 JMF |
1769 | """ |
| 1770 | Base class for extractors that fetch info from | |
| 1771 | http://www.youtube.com/feed_ajax | |
| 1772 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
| 1773 | """ | |
| b2e8bc1b | 1774 | _LOGIN_REQUIRED = True |
| 43ba5456 JMF |
1775 | # use action_load_personal_feed instead of action_load_system_feed |
| 1776 | _PERSONAL_FEED = False | |
| 04cc9617 | 1777 | |
| d7ae0639 JMF |
1778 | @property |
| 1779 | def _FEED_TEMPLATE(self): | |
| 43ba5456 JMF |
1780 | action = 'action_load_system_feed' |
| 1781 | if self._PERSONAL_FEED: | |
| 1782 | action = 'action_load_personal_feed' | |
| 1783 | return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) | |
| d7ae0639 JMF |
1784 | |
| 1785 | @property | |
| 1786 | def IE_NAME(self): | |
| 1787 | return u'youtube:%s' % self._FEED_NAME | |
| 04cc9617 | 1788 | |
| 81f0259b | 1789 | def _real_initialize(self): |
| b2e8bc1b | 1790 | self._login() |
| 81f0259b | 1791 | |
| 04cc9617 JMF |
1792 | def _real_extract(self, url): |
| 1793 | feed_entries = [] | |
| 0e44d838 JMF |
1794 | paging = 0 |
| 1795 | for i in itertools.count(1): | |
| d7ae0639 JMF |
1796 | info = self._download_webpage(self._FEED_TEMPLATE % paging, |
| 1797 | u'%s feed' % self._FEED_NAME, | |
| 04cc9617 JMF |
1798 | u'Downloading page %s' % i) |
| 1799 | info = json.loads(info) | |
| 1800 | feed_html = info['feed_html'] | |
| 43ba5456 | 1801 | m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) |
| 04cc9617 | 1802 | ids = orderedSet(m.group(1) for m in m_ids) |
| 7012b23c PH |
1803 | feed_entries.extend( |
| 1804 | self.url_result(video_id, 'Youtube', video_id=video_id) | |
| 1805 | for video_id in ids) | |
| 04cc9617 JMF |
1806 | if info['paging'] is None: |
| 1807 | break | |
| 0e44d838 | 1808 | paging = info['paging'] |
| d7ae0639 JMF |
1809 | return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) |
| 1810 | ||
| 1811 | class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): | |
| 1812 | IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' | |
| 1813 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
| 1814 | _FEED_NAME = 'subscriptions' | |
| 1815 | _PLAYLIST_TITLE = u'Youtube Subscriptions' | |
| 1816 | ||
| 1817 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): | |
| 1818 | IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' | |
| 1819 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' | |
| 1820 | _FEED_NAME = 'recommended' | |
| 1821 | _PLAYLIST_TITLE = u'Youtube Recommended videos' | |
| c626a3d9 | 1822 | |
| 43ba5456 JMF |
1823 | class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): |
| 1824 | IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' | |
| 1825 | _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' | |
| 1826 | _FEED_NAME = 'watch_later' | |
| 1827 | _PLAYLIST_TITLE = u'Youtube Watch Later' | |
| 43ba5456 | 1828 | _PERSONAL_FEED = True |
| c626a3d9 | 1829 | |
| f459d170 JMF |
1830 | class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): |
| 1831 | IE_DESC = u'Youtube watch history, "ythistory" keyword (requires authentication)' | |
| 1832 | _VALID_URL = u'https?://www\.youtube\.com/feed/history|:ythistory' | |
| 1833 | _FEED_NAME = 'history' | |
| 1834 | _PERSONAL_FEED = True | |
| 1835 | _PLAYLIST_TITLE = u'Youtube Watch History' | |
| 1836 | ||
| c626a3d9 JMF |
1837 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): |
| 1838 | IE_NAME = u'youtube:favorites' | |
| 1839 | IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' | |
| c7a7750d | 1840 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?' |
| c626a3d9 JMF |
1841 | _LOGIN_REQUIRED = True |
| 1842 | ||
| 1843 | def _real_extract(self, url): | |
| 1844 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
| 1845 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id') | |
| 1846 | return self.url_result(playlist_id, 'YoutubePlaylist') | |
| 15870e90 PH |
1847 | |
| 1848 | ||
| 1849 | class YoutubeTruncatedURLIE(InfoExtractor): | |
| 1850 | IE_NAME = 'youtube:truncated_url' | |
| 1851 | IE_DESC = False # Do not list | |
| 1852 | _VALID_URL = r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$' | |
| 1853 | ||
| 1854 | def _real_extract(self, url): | |
| 1855 | raise ExtractorError( | |
| 1856 | u'Did you forget to quote the URL? Remember that & is a meta ' | |
| 1857 | u'character in most shells, so you want to put the URL in quotes, ' | |
| 1858 | u'like youtube-dl ' | |
| 1859 | u'\'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\'' | |
| 1860 | u' (or simply youtube-dl BaW_jenozKc ).', | |
| 1861 | expected=True) |