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=([^&]+)' |
| 2c62dc26 PH |
153 | _formats = { |
| 154 | '5': {'ext': 'flv', 'width': 400, 'height': 240}, | |
| 155 | '6': {'ext': 'flv', 'width': 450, 'height': 270}, | |
| 156 | '13': {'ext': '3gp'}, | |
| 157 | '17': {'ext': '3gp', 'width': 176, 'height': 144}, | |
| 158 | '18': {'ext': 'mp4', 'width': 640, 'height': 360}, | |
| 159 | '22': {'ext': 'mp4', 'width': 1280, 'height': 720}, | |
| 160 | '34': {'ext': 'flv', 'width': 640, 'height': 360}, | |
| 161 | '35': {'ext': 'flv', 'width': 854, 'height': 480}, | |
| 162 | '36': {'ext': '3gp', 'width': 320, 'height': 240}, | |
| 163 | '37': {'ext': 'mp4', 'width': 1920, 'height': 1080}, | |
| 164 | '38': {'ext': 'mp4', 'width': 4096, 'height': 3072}, | |
| 165 | '43': {'ext': 'webm', 'width': 640, 'height': 360}, | |
| 166 | '44': {'ext': 'webm', 'width': 854, 'height': 480}, | |
| 167 | '45': {'ext': 'webm', 'width': 1280, 'height': 720}, | |
| 168 | '46': {'ext': 'webm', 'width': 1920, 'height': 1080}, | |
| 169 | ||
| 1d043b93 | 170 | |
| 86fe61c8 | 171 | # 3d videos |
| 2c62dc26 PH |
172 | '82': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': '3D', 'preference': -20}, |
| 173 | '83': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': '3D', 'preference': -20}, | |
| 174 | '84': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': '3D', 'preference': -20}, | |
| 175 | '85': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': '3D', 'preference': -20}, | |
| 176 | '100': {'ext': 'webm', 'height': 360, 'resolution': '360p', 'format_note': '3D', 'preference': -20}, | |
| 177 | '101': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': '3D', 'preference': -20}, | |
| 178 | '102': {'ext': 'webm', 'height': 720, 'resolution': '720p', 'format_note': '3D', 'preference': -20}, | |
| 836a086c | 179 | |
| 96fb5605 | 180 | # Apple HTTP Live Streaming |
| 2c62dc26 PH |
181 | '92': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'HLS', 'preference': -10}, |
| 182 | '93': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': 'HLS', 'preference': -10}, | |
| 183 | '94': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': 'HLS', 'preference': -10}, | |
| 184 | '95': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': 'HLS', 'preference': -10}, | |
| 185 | '96': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'HLS', 'preference': -10}, | |
| 186 | '132': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'HLS', 'preference': -10}, | |
| 187 | '151': {'ext': 'mp4', 'height': 72, 'resolution': '72p', 'format_note': 'HLS', 'preference': -10}, | |
| 188 | ||
| 189 | # DASH mp4 video | |
| 190 | '133': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'DASH video', 'preference': -40}, | |
| 191 | '134': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': 'DASH video', 'preference': -40}, | |
| 192 | '135': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': 'DASH video', 'preference': -40}, | |
| 193 | '136': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': 'DASH video', 'preference': -40}, | |
| 194 | '137': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH video', 'preference': -40}, | |
| 195 | '138': {'ext': 'mp4', 'height': 1081, 'resolution': '>1080p', 'format_note': 'DASH video', 'preference': -40}, | |
| 196 | '160': {'ext': 'mp4', 'height': 192, 'resolution': '192p', 'format_note': 'DASH video', 'preference': -40}, | |
| 8fa8a629 | 197 | '264': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH video', 'preference': -40}, |
| 836a086c | 198 | |
| f6f1fc92 | 199 | # Dash mp4 audio |
| 2c62dc26 PH |
200 | '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50}, |
| 201 | '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50}, | |
| 202 | '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50}, | |
| 836a086c AZ |
203 | |
| 204 | # Dash webm | |
| 2c62dc26 PH |
205 | '242': {'ext': 'webm', 'height': 240, 'resolution': '240p', 'format_note': 'DASH webm', 'preference': -40}, |
| 206 | '243': {'ext': 'webm', 'height': 360, 'resolution': '360p', 'format_note': 'DASH webm', 'preference': -40}, | |
| 207 | '244': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40}, | |
| 208 | '245': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40}, | |
| 209 | '246': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40}, | |
| 210 | '247': {'ext': 'webm', 'height': 720, 'resolution': '720p', 'format_note': 'DASH webm', 'preference': -40}, | |
| 211 | '248': {'ext': 'webm', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH webm', 'preference': -40}, | |
| 212 | ||
| 213 | # Dash webm audio | |
| 214 | '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 48, 'preference': -50}, | |
| 215 | '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 256, 'preference': -50}, | |
| ce6b9a2d PH |
216 | |
| 217 | # RTMP (unnamed) | |
| 218 | '_rtmp': {'protocol': 'rtmp'}, | |
| c5e8d7af | 219 | } |
| 836a086c | 220 | |
| c5e8d7af | 221 | IE_NAME = u'youtube' |
| 2eb88d95 PH |
222 | _TESTS = [ |
| 223 | { | |
| 0e853ca4 PH |
224 | u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", |
| 225 | u"file": u"BaW_jenozKc.mp4", | |
| 226 | u"info_dict": { | |
| 227 | u"title": u"youtube-dl test video \"'/\\ä↭𝕐", | |
| 228 | u"uploader": u"Philipp Hagemeister", | |
| 229 | u"uploader_id": u"phihag", | |
| 230 | u"upload_date": u"20121002", | |
| 27dcce19 | 231 | 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 | 232 | } |
| 0e853ca4 | 233 | }, |
| 0e853ca4 PH |
234 | { |
| 235 | u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", | |
| 236 | u"file": u"UxxajLWwzqY.mp4", | |
| 237 | u"note": u"Test generic use_cipher_signature video (#897)", | |
| 238 | u"info_dict": { | |
| 239 | u"upload_date": u"20120506", | |
| 240 | u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", | |
| 2dc59299 | 241 | u"description": u"md5:5b292926389560516e384ac437c0ec07", |
| 45ed795c | 242 | u"uploader": u"Icona Pop", |
| 0e853ca4 | 243 | u"uploader_id": u"IconaPop" |
| 2eb88d95 | 244 | } |
| c108eb73 JMF |
245 | }, |
| 246 | { | |
| 247 | u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", | |
| 248 | u"file": u"07FYdnEawAQ.mp4", | |
| 249 | u"note": u"Test VEVO video with age protection (#956)", | |
| 250 | u"info_dict": { | |
| 251 | u"upload_date": u"20130703", | |
| 252 | u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", | |
| 253 | u"description": u"md5:64249768eec3bc4276236606ea996373", | |
| 254 | u"uploader": u"justintimberlakeVEVO", | |
| 255 | u"uploader_id": u"justintimberlakeVEVO" | |
| 256 | } | |
| 257 | }, | |
| fccd3771 | 258 | { |
| 83aa5293 | 259 | u"url": u"//www.YouTube.com/watch?v=yZIXLfi8CZQ", |
| fccd3771 PH |
260 | u"file": u"yZIXLfi8CZQ.mp4", |
| 261 | u"note": u"Embed-only video (#1746)", | |
| 262 | u"info_dict": { | |
| 263 | u"upload_date": u"20120608", | |
| 264 | u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012", | |
| 265 | u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7", | |
| 266 | u"uploader": u"SET India", | |
| 267 | u"uploader_id": u"setindia" | |
| 268 | } | |
| 269 | }, | |
| 2eb88d95 PH |
270 | ] |
| 271 | ||
| c5e8d7af PH |
272 | |
| 273 | @classmethod | |
| 274 | def suitable(cls, url): | |
| 275 | """Receives a URL and returns True if suitable for this IE.""" | |
| e3ea4790 | 276 | if YoutubePlaylistIE.suitable(url): return False |
| fccd3771 | 277 | return re.match(cls._VALID_URL, url) is not None |
| c5e8d7af | 278 | |
| e0df6211 PH |
279 | def __init__(self, *args, **kwargs): |
| 280 | super(YoutubeIE, self).__init__(*args, **kwargs) | |
| 83799698 | 281 | self._player_cache = {} |
| e0df6211 | 282 | |
| c5e8d7af PH |
283 | def report_video_info_webpage_download(self, video_id): |
| 284 | """Report attempt to download video info webpage.""" | |
| 285 | self.to_screen(u'%s: Downloading video info webpage' % video_id) | |
| 286 | ||
| c5e8d7af PH |
287 | def report_information_extraction(self, video_id): |
| 288 | """Report attempt to extract video information.""" | |
| 289 | self.to_screen(u'%s: Extracting video information' % video_id) | |
| 290 | ||
| 291 | def report_unavailable_format(self, video_id, format): | |
| 292 | """Report extracted video URL.""" | |
| 293 | self.to_screen(u'%s: Format %s not available' % (video_id, format)) | |
| 294 | ||
| 295 | def report_rtmp_download(self): | |
| 296 | """Indicate the download will use the RTMP protocol.""" | |
| 297 | self.to_screen(u'RTMP download detected') | |
| 298 | ||
| c4417ddb PH |
299 | def _extract_signature_function(self, video_id, player_url, slen): |
| 300 | id_m = re.match(r'.*-(?P<id>[a-zA-Z0-9_-]+)\.(?P<ext>[a-z]+)$', | |
| 83799698 | 301 | player_url) |
| e0df6211 PH |
302 | player_type = id_m.group('ext') |
| 303 | player_id = id_m.group('id') | |
| 304 | ||
| c4417ddb PH |
305 | # Read from filesystem cache |
| 306 | func_id = '%s_%s_%d' % (player_type, player_id, slen) | |
| 307 | assert os.path.basename(func_id) == func_id | |
| c38b1e77 | 308 | cache_dir = get_cachedir(self._downloader.params) |
| c4417ddb | 309 | |
| c3c88a26 | 310 | cache_enabled = cache_dir is not None |
| f8061589 | 311 | if cache_enabled: |
| c4417ddb PH |
312 | cache_fn = os.path.join(os.path.expanduser(cache_dir), |
| 313 | u'youtube-sigfuncs', | |
| 314 | func_id + '.json') | |
| 315 | try: | |
| edf3e38e | 316 | with io.open(cache_fn, 'r', encoding='utf-8') as cachef: |
| c4417ddb PH |
317 | cache_spec = json.load(cachef) |
| 318 | return lambda s: u''.join(s[i] for i in cache_spec) | |
| edf3e38e | 319 | except IOError: |
| c4417ddb | 320 | pass # No cache available |
| 83799698 | 321 | |
| e0df6211 PH |
322 | if player_type == 'js': |
| 323 | code = self._download_webpage( | |
| 324 | player_url, video_id, | |
| 83799698 | 325 | note=u'Downloading %s player %s' % (player_type, player_id), |
| e0df6211 | 326 | errnote=u'Download of %s failed' % player_url) |
| 83799698 | 327 | res = self._parse_sig_js(code) |
| c4417ddb | 328 | elif player_type == 'swf': |
| e0df6211 PH |
329 | urlh = self._request_webpage( |
| 330 | player_url, video_id, | |
| 83799698 | 331 | note=u'Downloading %s player %s' % (player_type, player_id), |
| e0df6211 PH |
332 | errnote=u'Download of %s failed' % player_url) |
| 333 | code = urlh.read() | |
| 83799698 | 334 | res = self._parse_sig_swf(code) |
| e0df6211 PH |
335 | else: |
| 336 | assert False, 'Invalid player type %r' % player_type | |
| 337 | ||
| f8061589 | 338 | if cache_enabled: |
| edf3e38e | 339 | try: |
| c705320f PH |
340 | test_string = u''.join(map(compat_chr, range(slen))) |
| 341 | cache_res = res(test_string) | |
| edf3e38e PH |
342 | cache_spec = [ord(c) for c in cache_res] |
| 343 | try: | |
| 344 | os.makedirs(os.path.dirname(cache_fn)) | |
| 345 | except OSError as ose: | |
| 346 | if ose.errno != errno.EEXIST: | |
| 347 | raise | |
| 348 | write_json_file(cache_spec, cache_fn) | |
| 0ca96d48 | 349 | except Exception: |
| edf3e38e PH |
350 | tb = traceback.format_exc() |
| 351 | self._downloader.report_warning( | |
| 352 | u'Writing cache to %r failed: %s' % (cache_fn, tb)) | |
| 83799698 PH |
353 | |
| 354 | return res | |
| 355 | ||
| edf3e38e PH |
356 | def _print_sig_code(self, func, slen): |
| 357 | def gen_sig_code(idxs): | |
| 358 | def _genslice(start, end, step): | |
| 359 | starts = u'' if start == 0 else str(start) | |
| e35e4ddc PH |
360 | ends = (u':%d' % (end+step)) if end + step >= 0 else u':' |
| 361 | steps = u'' if step == 1 else (u':%d' % step) | |
| edf3e38e PH |
362 | return u's[%s%s%s]' % (starts, ends, steps) |
| 363 | ||
| 364 | step = None | |
| 0ca96d48 PH |
365 | start = '(Never used)' # Quelch pyflakes warnings - start will be |
| 366 | # set as soon as step is set | |
| edf3e38e PH |
367 | for i, prev in zip(idxs[1:], idxs[:-1]): |
| 368 | if step is not None: | |
| 369 | if i - prev == step: | |
| 370 | continue | |
| 371 | yield _genslice(start, prev, step) | |
| 372 | step = None | |
| 373 | continue | |
| 374 | if i - prev in [-1, 1]: | |
| 375 | step = i - prev | |
| 376 | start = prev | |
| 377 | continue | |
| 378 | else: | |
| 379 | yield u's[%d]' % prev | |
| 380 | if step is None: | |
| 381 | yield u's[%d]' % i | |
| 382 | else: | |
| 383 | yield _genslice(start, i, step) | |
| 384 | ||
| c705320f PH |
385 | test_string = u''.join(map(compat_chr, range(slen))) |
| 386 | cache_res = func(test_string) | |
| edf3e38e PH |
387 | cache_spec = [ord(c) for c in cache_res] |
| 388 | expr_code = u' + '.join(gen_sig_code(cache_spec)) | |
| 389 | code = u'if len(s) == %d:\n return %s\n' % (slen, expr_code) | |
| f8061589 | 390 | self.to_screen(u'Extracted signature function:\n' + code) |
| edf3e38e | 391 | |
| e0df6211 PH |
392 | def _parse_sig_js(self, jscode): |
| 393 | funcname = self._search_regex( | |
| 394 | r'signature=([a-zA-Z]+)', jscode, | |
| 395 | u'Initial JS player signature function name') | |
| 396 | ||
| 397 | functions = {} | |
| 398 | ||
| 399 | def argidx(varname): | |
| 400 | return string.lowercase.index(varname) | |
| 401 | ||
| 402 | def interpret_statement(stmt, local_vars, allow_recursion=20): | |
| 403 | if allow_recursion < 0: | |
| 0ca96d48 | 404 | raise ExtractorError(u'Recursion limit reached') |
| e0df6211 PH |
405 | |
| 406 | if stmt.startswith(u'var '): | |
| 407 | stmt = stmt[len(u'var '):] | |
| 408 | ass_m = re.match(r'^(?P<out>[a-z]+)(?:\[(?P<index>[^\]]+)\])?' + | |
| 409 | r'=(?P<expr>.*)$', stmt) | |
| 410 | if ass_m: | |
| 411 | if ass_m.groupdict().get('index'): | |
| 412 | def assign(val): | |
| 413 | lvar = local_vars[ass_m.group('out')] | |
| 414 | idx = interpret_expression(ass_m.group('index'), | |
| 415 | local_vars, allow_recursion) | |
| 416 | assert isinstance(idx, int) | |
| 417 | lvar[idx] = val | |
| 418 | return val | |
| 419 | expr = ass_m.group('expr') | |
| 420 | else: | |
| 421 | def assign(val): | |
| 422 | local_vars[ass_m.group('out')] = val | |
| 423 | return val | |
| 424 | expr = ass_m.group('expr') | |
| 425 | elif stmt.startswith(u'return '): | |
| 426 | assign = lambda v: v | |
| 427 | expr = stmt[len(u'return '):] | |
| 428 | else: | |
| 429 | raise ExtractorError( | |
| 430 | u'Cannot determine left side of statement in %r' % stmt) | |
| 431 | ||
| 432 | v = interpret_expression(expr, local_vars, allow_recursion) | |
| 433 | return assign(v) | |
| 434 | ||
| 435 | def interpret_expression(expr, local_vars, allow_recursion): | |
| 436 | if expr.isdigit(): | |
| 437 | return int(expr) | |
| 438 | ||
| 439 | if expr.isalpha(): | |
| 440 | return local_vars[expr] | |
| 441 | ||
| 442 | m = re.match(r'^(?P<in>[a-z]+)\.(?P<member>.*)$', expr) | |
| 443 | if m: | |
| 444 | member = m.group('member') | |
| 445 | val = local_vars[m.group('in')] | |
| 446 | if member == 'split("")': | |
| 447 | return list(val) | |
| 448 | if member == 'join("")': | |
| 449 | return u''.join(val) | |
| 450 | if member == 'length': | |
| 451 | return len(val) | |
| 452 | if member == 'reverse()': | |
| 453 | return val[::-1] | |
| 454 | slice_m = re.match(r'slice\((?P<idx>.*)\)', member) | |
| 455 | if slice_m: | |
| 456 | idx = interpret_expression( | |
| 457 | slice_m.group('idx'), local_vars, allow_recursion-1) | |
| 458 | return val[idx:] | |
| 459 | ||
| 460 | m = re.match( | |
| 461 | r'^(?P<in>[a-z]+)\[(?P<idx>.+)\]$', expr) | |
| 462 | if m: | |
| 463 | val = local_vars[m.group('in')] | |
| 464 | idx = interpret_expression(m.group('idx'), local_vars, | |
| 465 | allow_recursion-1) | |
| 466 | return val[idx] | |
| 467 | ||
| 468 | m = re.match(r'^(?P<a>.+?)(?P<op>[%])(?P<b>.+?)$', expr) | |
| 469 | if m: | |
| 470 | a = interpret_expression(m.group('a'), | |
| 471 | local_vars, allow_recursion) | |
| 472 | b = interpret_expression(m.group('b'), | |
| 473 | local_vars, allow_recursion) | |
| 474 | return a % b | |
| 475 | ||
| 476 | m = re.match( | |
| 477 | r'^(?P<func>[a-zA-Z]+)\((?P<args>[a-z0-9,]+)\)$', expr) | |
| 478 | if m: | |
| 479 | fname = m.group('func') | |
| 480 | if fname not in functions: | |
| 481 | functions[fname] = extract_function(fname) | |
| 482 | argvals = [int(v) if v.isdigit() else local_vars[v] | |
| 483 | for v in m.group('args').split(',')] | |
| 484 | return functions[fname](argvals) | |
| 485 | raise ExtractorError(u'Unsupported JS expression %r' % expr) | |
| 486 | ||
| 487 | def extract_function(funcname): | |
| 488 | func_m = re.search( | |
| 489 | r'function ' + re.escape(funcname) + | |
| 490 | r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}', | |
| 491 | jscode) | |
| 492 | argnames = func_m.group('args').split(',') | |
| 493 | ||
| 494 | def resf(args): | |
| 495 | local_vars = dict(zip(argnames, args)) | |
| 496 | for stmt in func_m.group('code').split(';'): | |
| 497 | res = interpret_statement(stmt, local_vars) | |
| 498 | return res | |
| 499 | return resf | |
| 500 | ||
| 501 | initial_function = extract_function(funcname) | |
| 502 | return lambda s: initial_function([s]) | |
| 503 | ||
| 504 | def _parse_sig_swf(self, file_contents): | |
| 505 | if file_contents[1:3] != b'WS': | |
| 506 | raise ExtractorError( | |
| 507 | u'Not an SWF file; header is %r' % file_contents[:3]) | |
| 508 | if file_contents[:1] == b'C': | |
| 509 | content = zlib.decompress(file_contents[8:]) | |
| 510 | else: | |
| 511 | raise NotImplementedError(u'Unsupported compression format %r' % | |
| 512 | file_contents[:1]) | |
| 513 | ||
| 514 | def extract_tags(content): | |
| 515 | pos = 0 | |
| 516 | while pos < len(content): | |
| 517 | header16 = struct.unpack('<H', content[pos:pos+2])[0] | |
| 518 | pos += 2 | |
| 519 | tag_code = header16 >> 6 | |
| 520 | tag_len = header16 & 0x3f | |
| 521 | if tag_len == 0x3f: | |
| 522 | tag_len = struct.unpack('<I', content[pos:pos+4])[0] | |
| 523 | pos += 4 | |
| 524 | assert pos+tag_len <= len(content) | |
| 525 | yield (tag_code, content[pos:pos+tag_len]) | |
| 526 | pos += tag_len | |
| 527 | ||
| 528 | code_tag = next(tag | |
| 529 | for tag_code, tag in extract_tags(content) | |
| 530 | if tag_code == 82) | |
| 531 | p = code_tag.index(b'\0', 4) + 1 | |
| ba552f54 | 532 | code_reader = io.BytesIO(code_tag[p:]) |
| e0df6211 PH |
533 | |
| 534 | # Parse ABC (AVM2 ByteCode) | |
| ba552f54 PH |
535 | def read_int(reader=None): |
| 536 | if reader is None: | |
| 537 | reader = code_reader | |
| e0df6211 PH |
538 | res = 0 |
| 539 | shift = 0 | |
| 540 | for _ in range(5): | |
| ba552f54 PH |
541 | buf = reader.read(1) |
| 542 | assert len(buf) == 1 | |
| 543 | b = struct.unpack('<B', buf)[0] | |
| e0df6211 PH |
544 | res = res | ((b & 0x7f) << shift) |
| 545 | if b & 0x80 == 0: | |
| 546 | break | |
| 547 | shift += 7 | |
| ba552f54 PH |
548 | return res |
| 549 | ||
| 550 | def u30(reader=None): | |
| 551 | res = read_int(reader) | |
| 552 | assert res & 0xf0000000 == 0 | |
| e0df6211 PH |
553 | return res |
| 554 | u32 = read_int | |
| 555 | ||
| ba552f54 PH |
556 | def s32(reader=None): |
| 557 | v = read_int(reader) | |
| e0df6211 PH |
558 | if v & 0x80000000 != 0: |
| 559 | v = - ((v ^ 0xffffffff) + 1) | |
| ba552f54 PH |
560 | return v |
| 561 | ||
| 0ca96d48 | 562 | def read_string(reader=None): |
| ba552f54 PH |
563 | if reader is None: |
| 564 | reader = code_reader | |
| 565 | slen = u30(reader) | |
| 566 | resb = reader.read(slen) | |
| 567 | assert len(resb) == slen | |
| 568 | return resb.decode('utf-8') | |
| 569 | ||
| 570 | def read_bytes(count, reader=None): | |
| 571 | if reader is None: | |
| 572 | reader = code_reader | |
| 573 | resb = reader.read(count) | |
| 574 | assert len(resb) == count | |
| 575 | return resb | |
| 576 | ||
| 577 | def read_byte(reader=None): | |
| 578 | resb = read_bytes(1, reader=reader) | |
| 579 | res = struct.unpack('<B', resb)[0] | |
| 580 | return res | |
| e0df6211 PH |
581 | |
| 582 | # minor_version + major_version | |
| 0ca96d48 | 583 | read_bytes(2 + 2) |
| e0df6211 PH |
584 | |
| 585 | # Constant pool | |
| ba552f54 | 586 | int_count = u30() |
| e0df6211 | 587 | for _c in range(1, int_count): |
| 0ca96d48 | 588 | s32() |
| ba552f54 | 589 | uint_count = u30() |
| e0df6211 | 590 | for _c in range(1, uint_count): |
| 0ca96d48 | 591 | u32() |
| ba552f54 | 592 | double_count = u30() |
| 0ca96d48 | 593 | read_bytes((double_count-1) * 8) |
| ba552f54 | 594 | string_count = u30() |
| e0df6211 PH |
595 | constant_strings = [u''] |
| 596 | for _c in range(1, string_count): | |
| 0ca96d48 | 597 | s = read_string() |
| e0df6211 | 598 | constant_strings.append(s) |
| ba552f54 | 599 | namespace_count = u30() |
| e0df6211 | 600 | for _c in range(1, namespace_count): |
| 0ca96d48 PH |
601 | read_bytes(1) # kind |
| 602 | u30() # name | |
| ba552f54 | 603 | ns_set_count = u30() |
| e0df6211 | 604 | for _c in range(1, ns_set_count): |
| ba552f54 | 605 | count = u30() |
| e0df6211 | 606 | for _c2 in range(count): |
| 0ca96d48 | 607 | u30() |
| ba552f54 | 608 | multiname_count = u30() |
| e0df6211 PH |
609 | MULTINAME_SIZES = { |
| 610 | 0x07: 2, # QName | |
| 611 | 0x0d: 2, # QNameA | |
| 612 | 0x0f: 1, # RTQName | |
| 613 | 0x10: 1, # RTQNameA | |
| 614 | 0x11: 0, # RTQNameL | |
| 615 | 0x12: 0, # RTQNameLA | |
| 616 | 0x09: 2, # Multiname | |
| 617 | 0x0e: 2, # MultinameA | |
| 618 | 0x1b: 1, # MultinameL | |
| 619 | 0x1c: 1, # MultinameLA | |
| 620 | } | |
| 621 | multinames = [u''] | |
| 622 | for _c in range(1, multiname_count): | |
| ba552f54 | 623 | kind = u30() |
| e0df6211 PH |
624 | assert kind in MULTINAME_SIZES, u'Invalid multiname kind %r' % kind |
| 625 | if kind == 0x07: | |
| 0ca96d48 | 626 | u30() # namespace_idx |
| ba552f54 | 627 | name_idx = u30() |
| e0df6211 PH |
628 | multinames.append(constant_strings[name_idx]) |
| 629 | else: | |
| 630 | multinames.append('[MULTINAME kind: %d]' % kind) | |
| 631 | for _c2 in range(MULTINAME_SIZES[kind]): | |
| 0ca96d48 | 632 | u30() |
| e0df6211 PH |
633 | |
| 634 | # Methods | |
| ba552f54 | 635 | method_count = u30() |
| e0df6211 PH |
636 | MethodInfo = collections.namedtuple( |
| 637 | 'MethodInfo', | |
| 638 | ['NEED_ARGUMENTS', 'NEED_REST']) | |
| 639 | method_infos = [] | |
| 640 | for method_id in range(method_count): | |
| ba552f54 | 641 | param_count = u30() |
| 0ca96d48 | 642 | u30() # return type |
| e0df6211 | 643 | for _ in range(param_count): |
| 0ca96d48 PH |
644 | u30() # param type |
| 645 | u30() # name index (always 0 for youtube) | |
| ba552f54 | 646 | flags = read_byte() |
| e0df6211 PH |
647 | if flags & 0x08 != 0: |
| 648 | # Options present | |
| ba552f54 | 649 | option_count = u30() |
| e0df6211 | 650 | for c in range(option_count): |
| 0ca96d48 PH |
651 | u30() # val |
| 652 | read_bytes(1) # kind | |
| e0df6211 PH |
653 | if flags & 0x80 != 0: |
| 654 | # Param names present | |
| 655 | for _ in range(param_count): | |
| 0ca96d48 | 656 | u30() # param name |
| e0df6211 PH |
657 | mi = MethodInfo(flags & 0x01 != 0, flags & 0x04 != 0) |
| 658 | method_infos.append(mi) | |
| 659 | ||
| 660 | # Metadata | |
| ba552f54 | 661 | metadata_count = u30() |
| e0df6211 | 662 | for _c in range(metadata_count): |
| 0ca96d48 | 663 | u30() # name |
| ba552f54 | 664 | item_count = u30() |
| e0df6211 | 665 | for _c2 in range(item_count): |
| 0ca96d48 PH |
666 | u30() # key |
| 667 | u30() # value | |
| ba552f54 PH |
668 | |
| 669 | def parse_traits_info(): | |
| 670 | trait_name_idx = u30() | |
| 671 | kind_full = read_byte() | |
| e0df6211 PH |
672 | kind = kind_full & 0x0f |
| 673 | attrs = kind_full >> 4 | |
| 674 | methods = {} | |
| 675 | if kind in [0x00, 0x06]: # Slot or Const | |
| 0ca96d48 PH |
676 | u30() # Slot id |
| 677 | u30() # type_name_idx | |
| ba552f54 | 678 | vindex = u30() |
| e0df6211 | 679 | if vindex != 0: |
| 0ca96d48 | 680 | read_byte() # vkind |
| e0df6211 | 681 | elif kind in [0x01, 0x02, 0x03]: # Method / Getter / Setter |
| 0ca96d48 | 682 | u30() # disp_id |
| ba552f54 | 683 | method_idx = u30() |
| e0df6211 PH |
684 | methods[multinames[trait_name_idx]] = method_idx |
| 685 | elif kind == 0x04: # Class | |
| 0ca96d48 PH |
686 | u30() # slot_id |
| 687 | u30() # classi | |
| e0df6211 | 688 | elif kind == 0x05: # Function |
| 0ca96d48 | 689 | u30() # slot_id |
| ba552f54 | 690 | function_idx = u30() |
| e0df6211 PH |
691 | methods[function_idx] = multinames[trait_name_idx] |
| 692 | else: | |
| 693 | raise ExtractorError(u'Unsupported trait kind %d' % kind) | |
| 694 | ||
| 695 | if attrs & 0x4 != 0: # Metadata present | |
| ba552f54 | 696 | metadata_count = u30() |
| e0df6211 | 697 | for _c3 in range(metadata_count): |
| 0ca96d48 | 698 | u30() # metadata index |
| e0df6211 | 699 | |
| ba552f54 | 700 | return methods |
| e0df6211 PH |
701 | |
| 702 | # Classes | |
| 703 | TARGET_CLASSNAME = u'SignatureDecipher' | |
| 704 | searched_idx = multinames.index(TARGET_CLASSNAME) | |
| 705 | searched_class_id = None | |
| ba552f54 | 706 | class_count = u30() |
| e0df6211 | 707 | for class_id in range(class_count): |
| ba552f54 | 708 | name_idx = u30() |
| e0df6211 PH |
709 | if name_idx == searched_idx: |
| 710 | # We found the class we're looking for! | |
| 711 | searched_class_id = class_id | |
| 0ca96d48 | 712 | u30() # super_name idx |
| ba552f54 | 713 | flags = read_byte() |
| e0df6211 | 714 | if flags & 0x08 != 0: # Protected namespace is present |
| 0ca96d48 | 715 | u30() # protected_ns_idx |
| ba552f54 | 716 | intrf_count = u30() |
| e0df6211 | 717 | for _c2 in range(intrf_count): |
| 0ca96d48 PH |
718 | u30() |
| 719 | u30() # iinit | |
| ba552f54 | 720 | trait_count = u30() |
| e0df6211 | 721 | for _c2 in range(trait_count): |
| 0ca96d48 | 722 | parse_traits_info() |
| e0df6211 PH |
723 | |
| 724 | if searched_class_id is None: | |
| 725 | raise ExtractorError(u'Target class %r not found' % | |
| 726 | TARGET_CLASSNAME) | |
| 727 | ||
| 728 | method_names = {} | |
| 729 | method_idxs = {} | |
| 730 | for class_id in range(class_count): | |
| 0ca96d48 | 731 | u30() # cinit |
| ba552f54 | 732 | trait_count = u30() |
| e0df6211 | 733 | for _c2 in range(trait_count): |
| ba552f54 | 734 | trait_methods = parse_traits_info() |
| e0df6211 PH |
735 | if class_id == searched_class_id: |
| 736 | method_names.update(trait_methods.items()) | |
| 737 | method_idxs.update(dict( | |
| 738 | (idx, name) | |
| 739 | for name, idx in trait_methods.items())) | |
| 740 | ||
| 741 | # Scripts | |
| ba552f54 | 742 | script_count = u30() |
| e0df6211 | 743 | for _c in range(script_count): |
| 0ca96d48 | 744 | u30() # init |
| ba552f54 | 745 | trait_count = u30() |
| e0df6211 | 746 | for _c2 in range(trait_count): |
| 0ca96d48 | 747 | parse_traits_info() |
| e0df6211 PH |
748 | |
| 749 | # Method bodies | |
| ba552f54 | 750 | method_body_count = u30() |
| e0df6211 PH |
751 | Method = collections.namedtuple('Method', ['code', 'local_count']) |
| 752 | methods = {} | |
| 753 | for _c in range(method_body_count): | |
| ba552f54 | 754 | method_idx = u30() |
| 0ca96d48 | 755 | u30() # max_stack |
| ba552f54 | 756 | local_count = u30() |
| 0ca96d48 PH |
757 | u30() # init_scope_depth |
| 758 | u30() # max_scope_depth | |
| ba552f54 PH |
759 | code_length = u30() |
| 760 | code = read_bytes(code_length) | |
| e0df6211 | 761 | if method_idx in method_idxs: |
| ba552f54 | 762 | m = Method(code, local_count) |
| e0df6211 | 763 | methods[method_idxs[method_idx]] = m |
| ba552f54 | 764 | exception_count = u30() |
| e0df6211 | 765 | for _c2 in range(exception_count): |
| 0ca96d48 PH |
766 | u30() # from |
| 767 | u30() # to | |
| 768 | u30() # target | |
| 769 | u30() # exc_type | |
| 770 | u30() # var_name | |
| ba552f54 | 771 | trait_count = u30() |
| e0df6211 | 772 | for _c2 in range(trait_count): |
| 0ca96d48 | 773 | parse_traits_info() |
| e0df6211 | 774 | |
| ba552f54 | 775 | assert p + code_reader.tell() == len(code_tag) |
| e0df6211 PH |
776 | assert len(methods) == len(method_idxs) |
| 777 | ||
| 778 | method_pyfunctions = {} | |
| 779 | ||
| 780 | def extract_function(func_name): | |
| 781 | if func_name in method_pyfunctions: | |
| 782 | return method_pyfunctions[func_name] | |
| 783 | if func_name not in methods: | |
| 784 | raise ExtractorError(u'Cannot find function %r' % func_name) | |
| 785 | m = methods[func_name] | |
| 786 | ||
| 787 | def resfunc(args): | |
| e0df6211 PH |
788 | registers = ['(this)'] + list(args) + [None] * m.local_count |
| 789 | stack = [] | |
| 790 | coder = io.BytesIO(m.code) | |
| 791 | while True: | |
| 792 | opcode = struct.unpack('!B', coder.read(1))[0] | |
| a7177865 | 793 | if opcode == 36: # pushbyte |
| e0df6211 PH |
794 | v = struct.unpack('!B', coder.read(1))[0] |
| 795 | stack.append(v) | |
| 796 | elif opcode == 44: # pushstring | |
| 797 | idx = u30(coder) | |
| 798 | stack.append(constant_strings[idx]) | |
| 799 | elif opcode == 48: # pushscope | |
| 800 | # We don't implement the scope register, so we'll just | |
| 801 | # ignore the popped value | |
| 802 | stack.pop() | |
| 803 | elif opcode == 70: # callproperty | |
| 804 | index = u30(coder) | |
| 805 | mname = multinames[index] | |
| 806 | arg_count = u30(coder) | |
| 807 | args = list(reversed( | |
| 808 | [stack.pop() for _ in range(arg_count)])) | |
| 809 | obj = stack.pop() | |
| 810 | if mname == u'split': | |
| 811 | assert len(args) == 1 | |
| 812 | assert isinstance(args[0], compat_str) | |
| 813 | assert isinstance(obj, compat_str) | |
| 814 | if args[0] == u'': | |
| 815 | res = list(obj) | |
| 816 | else: | |
| 817 | res = obj.split(args[0]) | |
| 818 | stack.append(res) | |
| a7177865 PH |
819 | elif mname == u'slice': |
| 820 | assert len(args) == 1 | |
| 821 | assert isinstance(args[0], int) | |
| 822 | assert isinstance(obj, list) | |
| 823 | res = obj[args[0]:] | |
| 824 | stack.append(res) | |
| 825 | elif mname == u'join': | |
| 826 | assert len(args) == 1 | |
| 827 | assert isinstance(args[0], compat_str) | |
| 828 | assert isinstance(obj, list) | |
| 829 | res = args[0].join(obj) | |
| 830 | stack.append(res) | |
| e0df6211 PH |
831 | elif mname in method_pyfunctions: |
| 832 | stack.append(method_pyfunctions[mname](args)) | |
| 833 | else: | |
| 834 | raise NotImplementedError( | |
| 835 | u'Unsupported property %r on %r' | |
| 836 | % (mname, obj)) | |
| a7177865 PH |
837 | elif opcode == 72: # returnvalue |
| 838 | res = stack.pop() | |
| 839 | return res | |
| 840 | elif opcode == 79: # callpropvoid | |
| 841 | index = u30(coder) | |
| 842 | mname = multinames[index] | |
| 843 | arg_count = u30(coder) | |
| 844 | args = list(reversed( | |
| 845 | [stack.pop() for _ in range(arg_count)])) | |
| 846 | obj = stack.pop() | |
| 847 | if mname == u'reverse': | |
| 848 | assert isinstance(obj, list) | |
| 849 | obj.reverse() | |
| 850 | else: | |
| 851 | raise NotImplementedError( | |
| 852 | u'Unsupported (void) property %r on %r' | |
| 853 | % (mname, obj)) | |
| e0df6211 PH |
854 | elif opcode == 93: # findpropstrict |
| 855 | index = u30(coder) | |
| 856 | mname = multinames[index] | |
| 857 | res = extract_function(mname) | |
| 858 | stack.append(res) | |
| 859 | elif opcode == 97: # setproperty | |
| 860 | index = u30(coder) | |
| 861 | value = stack.pop() | |
| 862 | idx = stack.pop() | |
| 863 | obj = stack.pop() | |
| 864 | assert isinstance(obj, list) | |
| 865 | assert isinstance(idx, int) | |
| 866 | obj[idx] = value | |
| 867 | elif opcode == 98: # getlocal | |
| 868 | index = u30(coder) | |
| 869 | stack.append(registers[index]) | |
| 870 | elif opcode == 99: # setlocal | |
| 871 | index = u30(coder) | |
| 872 | value = stack.pop() | |
| 873 | registers[index] = value | |
| 874 | elif opcode == 102: # getproperty | |
| 875 | index = u30(coder) | |
| 876 | pname = multinames[index] | |
| 877 | if pname == u'length': | |
| 878 | obj = stack.pop() | |
| 879 | assert isinstance(obj, list) | |
| 880 | stack.append(len(obj)) | |
| 881 | else: # Assume attribute access | |
| 882 | idx = stack.pop() | |
| 883 | assert isinstance(idx, int) | |
| 884 | obj = stack.pop() | |
| 885 | assert isinstance(obj, list) | |
| 886 | stack.append(obj[idx]) | |
| 887 | elif opcode == 128: # coerce | |
| 0ca96d48 | 888 | u30(coder) |
| e0df6211 PH |
889 | elif opcode == 133: # coerce_s |
| 890 | assert isinstance(stack[-1], (type(None), compat_str)) | |
| 891 | elif opcode == 164: # modulo | |
| 892 | value2 = stack.pop() | |
| 893 | value1 = stack.pop() | |
| 894 | res = value1 % value2 | |
| 895 | stack.append(res) | |
| a7177865 PH |
896 | elif opcode == 208: # getlocal_0 |
| 897 | stack.append(registers[0]) | |
| 898 | elif opcode == 209: # getlocal_1 | |
| 899 | stack.append(registers[1]) | |
| 900 | elif opcode == 210: # getlocal_2 | |
| 901 | stack.append(registers[2]) | |
| 902 | elif opcode == 211: # getlocal_3 | |
| 903 | stack.append(registers[3]) | |
| e0df6211 PH |
904 | elif opcode == 214: # setlocal_2 |
| 905 | registers[2] = stack.pop() | |
| 906 | elif opcode == 215: # setlocal_3 | |
| 907 | registers[3] = stack.pop() | |
| 908 | else: | |
| 909 | raise NotImplementedError( | |
| 910 | u'Unsupported opcode %d' % opcode) | |
| 911 | ||
| 912 | method_pyfunctions[func_name] = resfunc | |
| 913 | return resfunc | |
| 914 | ||
| 915 | initial_function = extract_function(u'decipher') | |
| 916 | return lambda s: initial_function([s]) | |
| 917 | ||
| 83799698 | 918 | def _decrypt_signature(self, s, video_id, player_url, age_gate=False): |
| 257a2501 | 919 | """Turn the encrypted s field into a working signature""" |
| 6b37f0be | 920 | |
| 83799698 | 921 | if player_url is not None: |
| 9f9be844 PH |
922 | if player_url.startswith(u'//'): |
| 923 | player_url = u'https:' + player_url | |
| e0df6211 | 924 | try: |
| 7f8ae73a PH |
925 | player_id = (player_url, len(s)) |
| 926 | if player_id not in self._player_cache: | |
| 83799698 | 927 | func = self._extract_signature_function( |
| c4417ddb | 928 | video_id, player_url, len(s) |
| e0df6211 | 929 | ) |
| 7f8ae73a PH |
930 | self._player_cache[player_id] = func |
| 931 | func = self._player_cache[player_id] | |
| edf3e38e PH |
932 | if self._downloader.params.get('youtube_print_sig_code'): |
| 933 | self._print_sig_code(func, len(s)) | |
| 934 | return func(s) | |
| 0ca96d48 | 935 | except Exception: |
| e0df6211 | 936 | tb = traceback.format_exc() |
| 83799698 PH |
937 | self._downloader.report_warning( |
| 938 | u'Automatic signature extraction failed: ' + tb) | |
| e0df6211 | 939 | |
| d2d8f895 PH |
940 | self._downloader.report_warning( |
| 941 | u'Warning: Falling back to static signature algorithm') | |
| 920de7a2 | 942 | |
| 2f2ffea9 PH |
943 | return self._static_decrypt_signature( |
| 944 | s, video_id, player_url, age_gate) | |
| e0df6211 | 945 | |
| 2f2ffea9 | 946 | def _static_decrypt_signature(self, s, video_id, player_url, age_gate): |
| e0df6211 PH |
947 | if age_gate: |
| 948 | # The videos with age protection use another player, so the | |
| 949 | # algorithms can be different. | |
| 950 | if len(s) == 86: | |
| 951 | return s[2:63] + s[82] + s[64:82] + s[63] | |
| 952 | ||
| bc4b9008 | 953 | if len(s) == 93: |
| 954 | return s[86:29:-1] + s[88] + s[28:5:-1] | |
| 955 | elif len(s) == 92: | |
| 444b1165 | 956 | return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83] |
| 38d025b3 JMF |
957 | elif len(s) == 91: |
| 958 | return s[84:27:-1] + s[86] + s[26:5:-1] | |
| 444b1165 JMF |
959 | elif len(s) == 90: |
| 960 | return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81] | |
| 8a9d86a2 | 961 | elif len(s) == 89: |
| 962 | return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1] | |
| 444b1165 | 963 | elif len(s) == 88: |
| 3e223834 | 964 | return s[7:28] + s[87] + s[29:45] + s[55] + s[46:55] + s[2] + s[56:87] + s[28] |
| be547e1d | 965 | elif len(s) == 87: |
| 3a725669 | 966 | return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:] |
| be547e1d | 967 | elif len(s) == 86: |
| f2c327fd | 968 | return s[80:72:-1] + s[16] + s[71:39:-1] + s[72] + s[38:16:-1] + s[82] + s[15::-1] |
| be547e1d | 969 | elif len(s) == 85: |
| 6ae8ee3f | 970 | return s[3:11] + s[0] + s[12:55] + s[84] + s[56:84] |
| be547e1d | 971 | elif len(s) == 84: |
| 6f56389b | 972 | return s[78:70:-1] + s[14] + s[69:37:-1] + s[70] + s[36:14:-1] + s[80] + s[:14][::-1] |
| be547e1d | 973 | elif len(s) == 83: |
| 920de7a2 | 974 | return s[80:63:-1] + s[0] + s[62:0:-1] + s[63] |
| be547e1d | 975 | elif len(s) == 82: |
| c21315f2 | 976 | return s[80:37:-1] + s[7] + s[36:7:-1] + s[0] + s[6:0:-1] + s[37] |
| be547e1d | 977 | elif len(s) == 81: |
| aedd6bb9 | 978 | 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 |
979 | elif len(s) == 80: |
| 980 | return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80] | |
| 5c468ca8 JMF |
981 | elif len(s) == 79: |
| 982 | 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 |
983 | |
| 984 | else: | |
| 985 | raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) | |
| c5e8d7af | 986 | |
| 1f343eaa | 987 | def _get_available_subtitles(self, video_id, webpage): |
| de7f3446 | 988 | try: |
| 7fad1c63 JMF |
989 | sub_list = self._download_webpage( |
| 990 | 'http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, | |
| 991 | video_id, note=False) | |
| 992 | except ExtractorError as err: | |
| de7f3446 JMF |
993 | self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) |
| 994 | return {} | |
| 995 | lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
| 996 | ||
| 997 | sub_lang_list = {} | |
| 998 | for l in lang_list: | |
| 999 | lang = l[1] | |
| 1000 | params = compat_urllib_parse.urlencode({ | |
| 1001 | 'lang': lang, | |
| 1002 | 'v': video_id, | |
| ca715127 | 1003 | 'fmt': self._downloader.params.get('subtitlesformat', 'srt'), |
| c3197e3e | 1004 | 'name': unescapeHTML(l[0]).encode('utf-8'), |
| de7f3446 JMF |
1005 | }) |
| 1006 | url = u'http://www.youtube.com/api/timedtext?' + params | |
| 1007 | sub_lang_list[lang] = url | |
| 1008 | if not sub_lang_list: | |
| 1009 | self._downloader.report_warning(u'video doesn\'t have subtitles') | |
| 1010 | return {} | |
| 1011 | return sub_lang_list | |
| 1012 | ||
| 055e6f36 | 1013 | def _get_available_automatic_caption(self, video_id, webpage): |
| de7f3446 JMF |
1014 | """We need the webpage for getting the captions url, pass it as an |
| 1015 | argument to speed up the process.""" | |
| ca715127 | 1016 | sub_format = self._downloader.params.get('subtitlesformat', 'srt') |
| de7f3446 JMF |
1017 | self.to_screen(u'%s: Looking for automatic captions' % video_id) |
| 1018 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
| 055e6f36 | 1019 | err_msg = u'Couldn\'t find automatic captions for %s' % video_id |
| de7f3446 JMF |
1020 | if mobj is None: |
| 1021 | self._downloader.report_warning(err_msg) | |
| 1022 | return {} | |
| 1023 | player_config = json.loads(mobj.group(1)) | |
| 1024 | try: | |
| 1025 | args = player_config[u'args'] | |
| 1026 | caption_url = args[u'ttsurl'] | |
| 1027 | timestamp = args[u'timestamp'] | |
| 055e6f36 JMF |
1028 | # We get the available subtitles |
| 1029 | list_params = compat_urllib_parse.urlencode({ | |
| 1030 | 'type': 'list', | |
| 1031 | 'tlangs': 1, | |
| 1032 | 'asrs': 1, | |
| de7f3446 | 1033 | }) |
| 055e6f36 | 1034 | list_url = caption_url + '&' + list_params |
| e26f8712 | 1035 | caption_list = self._download_xml(list_url, video_id) |
| e3dc22ca | 1036 | original_lang_node = caption_list.find('track') |
| f6a54188 | 1037 | if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' : |
| e3dc22ca JMF |
1038 | self._downloader.report_warning(u'Video doesn\'t have automatic captions') |
| 1039 | return {} | |
| 1040 | original_lang = original_lang_node.attrib['lang_code'] | |
| 055e6f36 JMF |
1041 | |
| 1042 | sub_lang_list = {} | |
| 1043 | for lang_node in caption_list.findall('target'): | |
| 1044 | sub_lang = lang_node.attrib['lang_code'] | |
| 1045 | params = compat_urllib_parse.urlencode({ | |
| 1046 | 'lang': original_lang, | |
| 1047 | 'tlang': sub_lang, | |
| 1048 | 'fmt': sub_format, | |
| 1049 | 'ts': timestamp, | |
| 1050 | 'kind': 'asr', | |
| 1051 | }) | |
| 1052 | sub_lang_list[sub_lang] = caption_url + '&' + params | |
| 1053 | return sub_lang_list | |
| de7f3446 JMF |
1054 | # An extractor error can be raise by the download process if there are |
| 1055 | # no automatic captions but there are subtitles | |
| 1056 | except (KeyError, ExtractorError): | |
| 1057 | self._downloader.report_warning(err_msg) | |
| 1058 | return {} | |
| 1059 | ||
| c5e8d7af PH |
1060 | def _extract_id(self, url): |
| 1061 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
| 1062 | if mobj is None: | |
| 1063 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 1064 | video_id = mobj.group(2) | |
| 1065 | return video_id | |
| 1066 | ||
| 1d043b93 JMF |
1067 | def _get_video_url_list(self, url_map): |
| 1068 | """ | |
| 1069 | Transform a dictionary in the format {itag:url} to a list of (itag, url) | |
| 1070 | with the requested formats. | |
| 1071 | """ | |
| 2c62dc26 | 1072 | existing_formats = [x for x in self._formats if x in url_map] |
| 1d043b93 JMF |
1073 | if len(existing_formats) == 0: |
| 1074 | raise ExtractorError(u'no known formats available for video') | |
| 4ea3be0a | 1075 | video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats |
| 1076 | video_url_list.reverse() # order worst to best | |
| 1d043b93 JMF |
1077 | return video_url_list |
| 1078 | ||
| 1079 | def _extract_from_m3u8(self, manifest_url, video_id): | |
| 1080 | url_map = {} | |
| 1081 | def _get_urls(_manifest): | |
| 1082 | lines = _manifest.split('\n') | |
| 1083 | urls = filter(lambda l: l and not l.startswith('#'), | |
| 1084 | lines) | |
| 1085 | return urls | |
| 1086 | manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') | |
| 1087 | formats_urls = _get_urls(manifest) | |
| 1088 | for format_url in formats_urls: | |
| 890f62e8 | 1089 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') |
| 1d043b93 JMF |
1090 | url_map[itag] = format_url |
| 1091 | return url_map | |
| 1092 | ||
| 1fb07d10 JG |
1093 | def _extract_annotations(self, video_id): |
| 1094 | url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id | |
| 1095 | return self._download_webpage(url, video_id, note=u'Searching for annotations.', errnote=u'Unable to download video annotations.') | |
| 1096 | ||
| c5e8d7af PH |
1097 | def _real_extract(self, url): |
| 1098 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter | |
| 1099 | mobj = re.search(self._NEXT_URL_RE, url) | |
| 1100 | if mobj: | |
| 1101 | url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
| 1102 | video_id = self._extract_id(url) | |
| 1103 | ||
| 1104 | # Get video webpage | |
| c5e8d7af | 1105 | url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id |
| 336c3a69 | 1106 | video_webpage = self._download_webpage(url, video_id) |
| c5e8d7af PH |
1107 | |
| 1108 | # Attempt to extract SWF player URL | |
| e0df6211 | 1109 | mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) |
| c5e8d7af PH |
1110 | if mobj is not None: |
| 1111 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
| 1112 | else: | |
| 1113 | player_url = None | |
| 1114 | ||
| 1115 | # Get video info | |
| 1116 | self.report_video_info_webpage_download(video_id) | |
| c108eb73 JMF |
1117 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
| 1118 | self.report_age_confirmation() | |
| 1119 | age_gate = True | |
| 1120 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
| 1121 | # this can be viewed without login into Youtube | |
| 1122 | data = compat_urllib_parse.urlencode({'video_id': video_id, | |
| fccd3771 | 1123 | 'el': 'player_embedded', |
| c108eb73 JMF |
1124 | 'gl': 'US', |
| 1125 | 'hl': 'en', | |
| 1126 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
| 1127 | 'asv': 3, | |
| 1128 | 'sts':'1588', | |
| 1129 | }) | |
| 1130 | video_info_url = 'https://www.youtube.com/get_video_info?' + data | |
| c5e8d7af PH |
1131 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
| 1132 | note=False, | |
| 1133 | errnote='unable to download video info webpage') | |
| 1134 | video_info = compat_parse_qs(video_info_webpage) | |
| c108eb73 JMF |
1135 | else: |
| 1136 | age_gate = False | |
| 1137 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
| 1138 | video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
| 1139 | % (video_id, el_type)) | |
| 1140 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
| 1141 | note=False, | |
| 1142 | errnote='unable to download video info webpage') | |
| 1143 | video_info = compat_parse_qs(video_info_webpage) | |
| 1144 | if 'token' in video_info: | |
| 1145 | break | |
| c5e8d7af PH |
1146 | if 'token' not in video_info: |
| 1147 | if 'reason' in video_info: | |
| 9a82b238 | 1148 | raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True) |
| c5e8d7af PH |
1149 | else: |
| 1150 | raise ExtractorError(u'"token" parameter not in video info for unknown reason') | |
| 1151 | ||
| 1d699755 PH |
1152 | if 'view_count' in video_info: |
| 1153 | view_count = int(video_info['view_count'][0]) | |
| 1154 | else: | |
| 1155 | view_count = None | |
| 1156 | ||
| c5e8d7af PH |
1157 | # Check for "rental" videos |
| 1158 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
| 1159 | raise ExtractorError(u'"rental" videos not supported') | |
| 1160 | ||
| 1161 | # Start extracting information | |
| 1162 | self.report_information_extraction(video_id) | |
| 1163 | ||
| 1164 | # uploader | |
| 1165 | if 'author' not in video_info: | |
| 1166 | raise ExtractorError(u'Unable to extract uploader name') | |
| 1167 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
| 1168 | ||
| 1169 | # uploader_id | |
| 1170 | video_uploader_id = None | |
| 1171 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
| 1172 | if mobj is not None: | |
| 1173 | video_uploader_id = mobj.group(1) | |
| 1174 | else: | |
| 1175 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
| 1176 | ||
| 1177 | # title | |
| a8c6b241 PH |
1178 | if 'title' in video_info: |
| 1179 | video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) | |
| 1180 | else: | |
| 1181 | self._downloader.report_warning(u'Unable to extract video title') | |
| 1182 | video_title = u'_' | |
| c5e8d7af PH |
1183 | |
| 1184 | # thumbnail image | |
| 7763b04e JMF |
1185 | # We try first to get a high quality image: |
| 1186 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
| 1187 | video_webpage, re.DOTALL) | |
| 1188 | if m_thumb is not None: | |
| 1189 | video_thumbnail = m_thumb.group(1) | |
| 1190 | elif 'thumbnail_url' not in video_info: | |
| c5e8d7af | 1191 | self._downloader.report_warning(u'unable to extract video thumbnail') |
| f490e77e | 1192 | video_thumbnail = None |
| c5e8d7af PH |
1193 | else: # don't panic if we can't find it |
| 1194 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
| 1195 | ||
| 1196 | # upload date | |
| 1197 | upload_date = None | |
| 1198 | mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL) | |
| 1199 | if mobj is not None: | |
| 1200 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
| 1201 | upload_date = unified_strdate(upload_date) | |
| 1202 | ||
| 1203 | # description | |
| 1204 | video_description = get_element_by_id("eow-description", video_webpage) | |
| 1205 | if video_description: | |
| 27dcce19 PH |
1206 | video_description = re.sub(r'''(?x) |
| 1207 | <a\s+ | |
| 1208 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
| 1209 | title="([^"]+)"\s+ | |
| 1210 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
| 1211 | class="yt-uix-redirect-link"\s*> | |
| 1212 | [^<]+ | |
| 1213 | </a> | |
| 1214 | ''', r'\1', video_description) | |
| c5e8d7af PH |
1215 | video_description = clean_html(video_description) |
| 1216 | else: | |
| 1217 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
| 1218 | if fd_mobj: | |
| 1219 | video_description = unescapeHTML(fd_mobj.group(1)) | |
| 1220 | else: | |
| 1221 | video_description = u'' | |
| 1222 | ||
| 336c3a69 | 1223 | def _extract_count(klass): |
| 46374a56 PH |
1224 | count = self._search_regex( |
| 1225 | r'class="%s">([\d,]+)</span>' % re.escape(klass), | |
| 1226 | video_webpage, klass, default=None) | |
| 336c3a69 JMF |
1227 | if count is not None: |
| 1228 | return int(count.replace(',', '')) | |
| 1229 | return None | |
| 1230 | like_count = _extract_count(u'likes-count') | |
| 1231 | dislike_count = _extract_count(u'dislikes-count') | |
| 1232 | ||
| c5e8d7af | 1233 | # subtitles |
| d82134c3 | 1234 | video_subtitles = self.extract_subtitles(video_id, video_webpage) |
| c5e8d7af | 1235 | |
| c5e8d7af | 1236 | if self._downloader.params.get('listsubtitles', False): |
| d665f8d3 | 1237 | self._list_available_subtitles(video_id, video_webpage) |
| c5e8d7af PH |
1238 | return |
| 1239 | ||
| 1240 | if 'length_seconds' not in video_info: | |
| 1241 | self._downloader.report_warning(u'unable to extract video duration') | |
| b466b702 | 1242 | video_duration = None |
| c5e8d7af | 1243 | else: |
| b466b702 | 1244 | video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) |
| c5e8d7af | 1245 | |
| 1fb07d10 JG |
1246 | # annotations |
| 1247 | video_annotations = None | |
| 1248 | if self._downloader.params.get('writeannotations', False): | |
| 1249 | video_annotations = self._extract_annotations(video_id) | |
| 1250 | ||
| c5e8d7af | 1251 | # Decide which formats to download |
| c5e8d7af PH |
1252 | |
| 1253 | try: | |
| 1254 | mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) | |
| 50be92c1 PH |
1255 | if not mobj: |
| 1256 | raise ValueError('Could not find vevo ID') | |
| c5e8d7af PH |
1257 | info = json.loads(mobj.group(1)) |
| 1258 | args = info['args'] | |
| 7ce7e394 JMF |
1259 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
| 1260 | # this signatures are encrypted | |
| 44d46655 | 1261 | if 'url_encoded_fmt_stream_map' not in args: |
| f10503db | 1262 | raise ValueError(u'No stream_map present') # caught below |
| 00fe14fc JMF |
1263 | re_signature = re.compile(r'[&,]s=') |
| 1264 | m_s = re_signature.search(args['url_encoded_fmt_stream_map']) | |
| 7ce7e394 JMF |
1265 | if m_s is not None: |
| 1266 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
| c5e8d7af | 1267 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
| 00fe14fc | 1268 | m_s = re_signature.search(args.get('adaptive_fmts', u'')) |
| b7a68384 | 1269 | if m_s is not None: |
| 00fe14fc JMF |
1270 | if 'adaptive_fmts' in video_info: |
| 1271 | video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts'] | |
| 37b6d5f6 | 1272 | else: |
| 00fe14fc | 1273 | video_info['adaptive_fmts'] = [args['adaptive_fmts']] |
| c5e8d7af PH |
1274 | except ValueError: |
| 1275 | pass | |
| 1276 | ||
| 1277 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
| 1278 | self.report_rtmp_download() | |
| ce6b9a2d | 1279 | video_url_list = [('_rtmp', video_info['conn'][0])] |
| 00fe14fc JMF |
1280 | elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1: |
| 1281 | encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0] | |
| 1282 | if 'rtmpe%3Dyes' in encoded_url_map: | |
| a7055eb9 | 1283 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) |
| c5e8d7af | 1284 | url_map = {} |
| 00fe14fc | 1285 | for url_data_str in encoded_url_map.split(','): |
| c5e8d7af PH |
1286 | url_data = compat_parse_qs(url_data_str) |
| 1287 | if 'itag' in url_data and 'url' in url_data: | |
| 1288 | url = url_data['url'][0] | |
| 1289 | if 'sig' in url_data: | |
| 1290 | url += '&signature=' + url_data['sig'][0] | |
| 1291 | elif 's' in url_data: | |
| e0df6211 | 1292 | encrypted_sig = url_data['s'][0] |
| 769fda3c | 1293 | if self._downloader.params.get('verbose'): |
| c108eb73 | 1294 | if age_gate: |
| bdde940e PH |
1295 | if player_url is None: |
| 1296 | player_version = 'unknown' | |
| 1297 | else: | |
| 1298 | player_version = self._search_regex( | |
| 1299 | r'-(.+)\.swf$', player_url, | |
| 1300 | u'flash player', fatal=False) | |
| e0df6211 | 1301 | player_desc = 'flash player %s' % player_version |
| c108eb73 | 1302 | else: |
| 83799698 PH |
1303 | player_version = self._search_regex( |
| 1304 | r'html5player-(.+?)\.js', video_webpage, | |
| c108eb73 | 1305 | 'html5 player', fatal=False) |
| e0df6211 PH |
1306 | player_desc = u'html5 player %s' % player_version |
| 1307 | ||
| 1308 | parts_sizes = u'.'.join(compat_str(len(part)) for part in encrypted_sig.split('.')) | |
| 5a76c651 | 1309 | self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % |
| e0df6211 PH |
1310 | (len(encrypted_sig), parts_sizes, url_data['itag'][0], player_desc)) |
| 1311 | ||
| 83799698 | 1312 | if not age_gate: |
| e0df6211 PH |
1313 | jsplayer_url_json = self._search_regex( |
| 1314 | r'"assets":.+?"js":\s*("[^"]+")', | |
| 1315 | video_webpage, u'JS player URL') | |
| 83799698 | 1316 | player_url = json.loads(jsplayer_url_json) |
| e0df6211 | 1317 | |
| 83799698 PH |
1318 | signature = self._decrypt_signature( |
| 1319 | encrypted_sig, video_id, player_url, age_gate) | |
| c5e8d7af PH |
1320 | url += '&signature=' + signature |
| 1321 | if 'ratebypass' not in url: | |
| 1322 | url += '&ratebypass=yes' | |
| 1323 | url_map[url_data['itag'][0]] = url | |
| 1d043b93 | 1324 | video_url_list = self._get_video_url_list(url_map) |
| 1d043b93 JMF |
1325 | elif video_info.get('hlsvp'): |
| 1326 | manifest_url = video_info['hlsvp'][0] | |
| 1327 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
| 1328 | video_url_list = self._get_video_url_list(url_map) | |
| c5e8d7af | 1329 | else: |
| 9abb3204 | 1330 | raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') |
| c5e8d7af | 1331 | |
| 4ea3be0a | 1332 | formats = [] |
| 600cc1a4 | 1333 | for itag, video_real_url in video_url_list: |
| 2c62dc26 PH |
1334 | dct = { |
| 1335 | 'format_id': itag, | |
| 1336 | 'url': video_real_url, | |
| 1337 | 'player_url': player_url, | |
| 1338 | } | |
| 1339 | dct.update(self._formats[itag]) | |
| 1340 | formats.append(dct) | |
| d80044c2 | 1341 | |
| 4bcc7bd1 | 1342 | self._sort_formats(formats) |
| 4ea3be0a | 1343 | |
| 1344 | return { | |
| 1345 | 'id': video_id, | |
| 1346 | 'uploader': video_uploader, | |
| 1347 | 'uploader_id': video_uploader_id, | |
| 1348 | 'upload_date': upload_date, | |
| 1349 | 'title': video_title, | |
| 1350 | 'thumbnail': video_thumbnail, | |
| 1351 | 'description': video_description, | |
| 1352 | 'subtitles': video_subtitles, | |
| 1353 | 'duration': video_duration, | |
| 1354 | 'age_limit': 18 if age_gate else 0, | |
| 1355 | 'annotations': video_annotations, | |
| 1356 | 'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id, | |
| 1357 | 'view_count': view_count, | |
| 1358 | 'like_count': like_count, | |
| 1359 | 'dislike_count': dislike_count, | |
| 1360 | 'formats': formats, | |
| 1361 | } | |
| c5e8d7af | 1362 | |
| 880e1c52 | 1363 | class YoutubePlaylistIE(YoutubeBaseInfoExtractor): |
| 0f818663 | 1364 | IE_DESC = u'YouTube.com playlists' |
| c5e8d7af PH |
1365 | _VALID_URL = r"""(?: |
| 1366 | (?:https?://)? | |
| 1367 | (?:\w+\.)? | |
| 1368 | youtube\.com/ | |
| 1369 | (?: | |
| 1370 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
| 1371 | \? (?:.*?&)*? (?:p|a|list)= | |
| 1372 | | p/ | |
| 1373 | ) | |
| 715c8e7b | 1374 | ((?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}) |
| c5e8d7af PH |
1375 | .* |
| 1376 | | | |
| 715c8e7b | 1377 | ((?:PL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,}) |
| c5e8d7af | 1378 | )""" |
| dcbb4580 JMF |
1379 | _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s' |
| 1380 | _MORE_PAGES_INDICATOR = r'data-link-type="next"' | |
| 6e47b51e | 1381 | _VIDEO_RE = r'href="/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' |
| c5e8d7af PH |
1382 | IE_NAME = u'youtube:playlist' |
| 1383 | ||
| 1384 | @classmethod | |
| 1385 | def suitable(cls, url): | |
| 1386 | """Receives a URL and returns True if suitable for this IE.""" | |
| 1387 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
| 1388 | ||
| 880e1c52 JMF |
1389 | def _real_initialize(self): |
| 1390 | self._login() | |
| 1391 | ||
| 652cdaa2 JMF |
1392 | def _ids_to_results(self, ids): |
| 1393 | return [self.url_result(vid_id, 'Youtube', video_id=vid_id) | |
| 1394 | for vid_id in ids] | |
| 1395 | ||
| 1396 | def _extract_mix(self, playlist_id): | |
| 1397 | # The mixes are generated from a a single video | |
| 1398 | # the id of the playlist is just 'RD' + video_id | |
| 7d4afc55 | 1399 | url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id) |
| 652cdaa2 | 1400 | webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix') |
| 76d1700b JMF |
1401 | title_span = (get_element_by_attribute('class', 'title long-title', webpage) or |
| 1402 | get_element_by_attribute('class', 'title ', webpage)) | |
| 1403 | title = clean_html(title_span) | |
| 652cdaa2 JMF |
1404 | video_re = r'data-index="\d+".*?href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s' % re.escape(playlist_id) |
| 1405 | ids = orderedSet(re.findall(video_re, webpage)) | |
| 1406 | url_results = self._ids_to_results(ids) | |
| 1407 | ||
| 1408 | return self.playlist_result(url_results, playlist_id, title) | |
| 1409 | ||
| c5e8d7af PH |
1410 | def _real_extract(self, url): |
| 1411 | # Extract playlist id | |
| 1412 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
| 1413 | if mobj is None: | |
| 1414 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 47192f92 FV |
1415 | playlist_id = mobj.group(1) or mobj.group(2) |
| 1416 | ||
| 1417 | # Check if it's a video-specific URL | |
| 7c61bd36 | 1418 | query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) |
| 47192f92 FV |
1419 | if 'v' in query_dict: |
| 1420 | video_id = query_dict['v'][0] | |
| 1421 | if self._downloader.params.get('noplaylist'): | |
| 1422 | self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id) | |
| 7012b23c | 1423 | return self.url_result(video_id, 'Youtube', video_id=video_id) |
| 47192f92 FV |
1424 | else: |
| 1425 | self.to_screen(u'Downloading playlist PL%s - add --no-playlist to just download video %s' % (playlist_id, video_id)) | |
| c5e8d7af | 1426 | |
| 7d4afc55 | 1427 | if playlist_id.startswith('RD'): |
| 652cdaa2 JMF |
1428 | # Mixes require a custom extraction process |
| 1429 | return self._extract_mix(playlist_id) | |
| 0a688bc0 JMF |
1430 | if playlist_id.startswith('TL'): |
| 1431 | raise ExtractorError(u'For downloading YouTube.com top lists, use ' | |
| 1432 | u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) | |
| 652cdaa2 | 1433 | |
| dcbb4580 JMF |
1434 | # Extract the video ids from the playlist pages |
| 1435 | ids = [] | |
| c5e8d7af | 1436 | |
| 755eb032 | 1437 | for page_num in itertools.count(1): |
| dcbb4580 | 1438 | url = self._TEMPLATE_URL % (playlist_id, page_num) |
| c5e8d7af | 1439 | page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) |
| 6e47b51e JMF |
1440 | matches = re.finditer(self._VIDEO_RE, page) |
| 1441 | # We remove the duplicates and the link with index 0 | |
| 1442 | # (it's not the first video of the playlist) | |
| 1443 | new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') | |
| dcbb4580 | 1444 | ids.extend(new_ids) |
| c5e8d7af | 1445 | |
| dcbb4580 | 1446 | if re.search(self._MORE_PAGES_INDICATOR, page) is None: |
| c5e8d7af PH |
1447 | break |
| 1448 | ||
| dcbb4580 | 1449 | playlist_title = self._og_search_title(page) |
| c5e8d7af | 1450 | |
| 652cdaa2 | 1451 | url_results = self._ids_to_results(ids) |
| dcbb4580 | 1452 | return self.playlist_result(url_results, playlist_id, playlist_title) |
| c5e8d7af PH |
1453 | |
| 1454 | ||
| 0a688bc0 JMF |
1455 | class YoutubeTopListIE(YoutubePlaylistIE): |
| 1456 | IE_NAME = u'youtube:toplist' | |
| 1457 | IE_DESC = (u'YouTube.com top lists, "yttoplist:{channel}:{list title}"' | |
| 1458 | u' (Example: "yttoplist:music:Top Tracks")') | |
| 1459 | _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$' | |
| 1460 | ||
| 1461 | def _real_extract(self, url): | |
| 1462 | mobj = re.match(self._VALID_URL, url) | |
| 1463 | channel = mobj.group('chann') | |
| 1464 | title = mobj.group('title') | |
| 1465 | query = compat_urllib_parse.urlencode({'title': title}) | |
| 1466 | playlist_re = 'href="([^"]+?%s[^"]+?)"' % re.escape(query) | |
| 1467 | channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title) | |
| 1468 | link = self._html_search_regex(playlist_re, channel_page, u'list') | |
| 1469 | url = compat_urlparse.urljoin('https://www.youtube.com/', link) | |
| 1470 | ||
| 1471 | video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"' | |
| 1472 | ids = [] | |
| 1473 | # sometimes the webpage doesn't contain the videos | |
| 1474 | # retry until we get them | |
| 1475 | for i in itertools.count(0): | |
| 1476 | msg = u'Downloading Youtube mix' | |
| 1477 | if i > 0: | |
| 1478 | msg += ', retry #%d' % i | |
| 1479 | webpage = self._download_webpage(url, title, msg) | |
| 1480 | ids = orderedSet(re.findall(video_re, webpage)) | |
| 1481 | if ids: | |
| 1482 | break | |
| 1483 | url_results = self._ids_to_results(ids) | |
| 1484 | return self.playlist_result(url_results, playlist_title=title) | |
| 1485 | ||
| 1486 | ||
| c5e8d7af | 1487 | class YoutubeChannelIE(InfoExtractor): |
| 0f818663 | 1488 | IE_DESC = u'YouTube.com channels' |
| c5e8d7af | 1489 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
| c5e8d7af | 1490 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' |
| 252580c5 | 1491 | _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 |
1492 | IE_NAME = u'youtube:channel' |
| 1493 | ||
| 1494 | def extract_videos_from_page(self, page): | |
| 1495 | ids_in_page = [] | |
| 1496 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
| 1497 | if mobj.group(1) not in ids_in_page: | |
| 1498 | ids_in_page.append(mobj.group(1)) | |
| 1499 | return ids_in_page | |
| 1500 | ||
| 1501 | def _real_extract(self, url): | |
| 1502 | # Extract channel id | |
| 1503 | mobj = re.match(self._VALID_URL, url) | |
| 1504 | if mobj is None: | |
| 1505 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 1506 | ||
| 1507 | # Download channel page | |
| 1508 | channel_id = mobj.group(1) | |
| 1509 | video_ids = [] | |
| b9643eed JMF |
1510 | url = 'https://www.youtube.com/channel/%s/videos' % channel_id |
| 1511 | channel_page = self._download_webpage(url, channel_id) | |
| 31812a9e PH |
1512 | autogenerated = re.search(r'''(?x) |
| 1513 | class="[^"]*?(?: | |
| 1514 | channel-header-autogenerated-label| | |
| 1515 | yt-channel-title-autogenerated | |
| 1516 | )[^"]*"''', channel_page) is not None | |
| c5e8d7af | 1517 | |
| b9643eed JMF |
1518 | if autogenerated: |
| 1519 | # The videos are contained in a single page | |
| 1520 | # the ajax pages can't be used, they are empty | |
| 1521 | video_ids = self.extract_videos_from_page(channel_page) | |
| 1522 | else: | |
| 1523 | # Download all channel pages using the json-based channel_ajax query | |
| 1524 | for pagenum in itertools.count(1): | |
| 1525 | url = self._MORE_PAGES_URL % (pagenum, channel_id) | |
| 1526 | page = self._download_webpage(url, channel_id, | |
| 1527 | u'Downloading page #%s' % pagenum) | |
| 1528 | ||
| 1529 | page = json.loads(page) | |
| 1530 | ||
| 1531 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
| 1532 | video_ids.extend(ids_in_page) | |
| 1533 | ||
| 1534 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
| 1535 | break | |
| c5e8d7af PH |
1536 | |
| 1537 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
| 1538 | ||
| 7012b23c PH |
1539 | url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id) |
| 1540 | for video_id in video_ids] | |
| 1541 | return self.playlist_result(url_entries, channel_id) | |
| c5e8d7af PH |
1542 | |
| 1543 | ||
| 1544 | class YoutubeUserIE(InfoExtractor): | |
| 0f818663 | 1545 | IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' |
| 57da92b7 | 1546 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)' |
| c5e8d7af PH |
1547 | _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' |
| 1548 | _GDATA_PAGE_SIZE = 50 | |
| fd9cf738 | 1549 | _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' |
| c5e8d7af PH |
1550 | IE_NAME = u'youtube:user' |
| 1551 | ||
| e3ea4790 | 1552 | @classmethod |
| f4b05232 | 1553 | def suitable(cls, url): |
| e3ea4790 JMF |
1554 | # Don't return True if the url can be extracted with other youtube |
| 1555 | # extractor, the regex would is too permissive and it would match. | |
| 1556 | other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) | |
| 1557 | if any(ie.suitable(url) for ie in other_ies): return False | |
| f4b05232 JMF |
1558 | else: return super(YoutubeUserIE, cls).suitable(url) |
| 1559 | ||
| c5e8d7af PH |
1560 | def _real_extract(self, url): |
| 1561 | # Extract username | |
| 1562 | mobj = re.match(self._VALID_URL, url) | |
| 1563 | if mobj is None: | |
| 1564 | raise ExtractorError(u'Invalid URL: %s' % url) | |
| 1565 | ||
| 1566 | username = mobj.group(1) | |
| 1567 | ||
| 1568 | # Download video ids using YouTube Data API. Result size per | |
| 1569 | # query is limited (currently to 50 videos) so we need to query | |
| 1570 | # page by page until there are no video ids - it means we got | |
| 1571 | # all of them. | |
| 1572 | ||
| e302f9ce | 1573 | url_results = [] |
| c5e8d7af | 1574 | |
| 755eb032 | 1575 | for pagenum in itertools.count(0): |
| c5e8d7af PH |
1576 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 |
| 1577 | ||
| 1578 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
| 1579 | page = self._download_webpage(gdata_url, username, | |
| 1580 | u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) | |
| 1581 | ||
| fd9cf738 JMF |
1582 | try: |
| 1583 | response = json.loads(page) | |
| 1584 | except ValueError as err: | |
| 1585 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
| 71c82637 JMF |
1586 | if 'entry' not in response['feed']: |
| 1587 | # Number of videos is a multiple of self._MAX_RESULTS | |
| 1588 | break | |
| fd9cf738 | 1589 | |
| c5e8d7af | 1590 | # Extract video identifiers |
| e302f9ce PH |
1591 | entries = response['feed']['entry'] |
| 1592 | for entry in entries: | |
| 1593 | title = entry['title']['$t'] | |
| 1594 | video_id = entry['id']['$t'].split('/')[-1] | |
| 1595 | url_results.append({ | |
| 1596 | '_type': 'url', | |
| 1597 | 'url': video_id, | |
| 1598 | 'ie_key': 'Youtube', | |
| 1599 | 'id': 'video_id', | |
| 1600 | 'title': title, | |
| 1601 | }) | |
| c5e8d7af PH |
1602 | |
| 1603 | # A little optimization - if current page is not | |
| 1604 | # "full", ie. does not contain PAGE_SIZE video ids then | |
| 1605 | # we can assume that this page is the last one - there | |
| 1606 | # are no more ids on further pages - no need to query | |
| 1607 | # again. | |
| 1608 | ||
| e302f9ce | 1609 | if len(entries) < self._GDATA_PAGE_SIZE: |
| c5e8d7af PH |
1610 | break |
| 1611 | ||
| 7012b23c PH |
1612 | return self.playlist_result(url_results, playlist_title=username) |
| 1613 | ||
| b05654f0 PH |
1614 | |
| 1615 | class YoutubeSearchIE(SearchInfoExtractor): | |
| 0f818663 | 1616 | IE_DESC = u'YouTube.com searches' |
| b05654f0 PH |
1617 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' |
| 1618 | _MAX_RESULTS = 1000 | |
| 1619 | IE_NAME = u'youtube:search' | |
| 1620 | _SEARCH_KEY = 'ytsearch' | |
| 1621 | ||
| b05654f0 PH |
1622 | def _get_n_results(self, query, n): |
| 1623 | """Get a specified number of results for a query""" | |
| 1624 | ||
| 1625 | video_ids = [] | |
| 1626 | pagenum = 0 | |
| 1627 | limit = n | |
| 1628 | ||
| 1629 | while (50 * pagenum) < limit: | |
| b05654f0 | 1630 | result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) |
| 7cc3570e PH |
1631 | data_json = self._download_webpage( |
| 1632 | result_url, video_id=u'query "%s"' % query, | |
| 1633 | note=u'Downloading page %s' % (pagenum + 1), | |
| 1634 | errnote=u'Unable to download API page') | |
| 1635 | data = json.loads(data_json) | |
| 1636 | api_response = data['data'] | |
| 1637 | ||
| 1638 | if 'items' not in api_response: | |
| b05654f0 PH |
1639 | raise ExtractorError(u'[youtube] No video results') |
| 1640 | ||
| 1641 | new_ids = list(video['id'] for video in api_response['items']) | |
| 1642 | video_ids += new_ids | |
| 1643 | ||
| 1644 | limit = min(n, api_response['totalItems']) | |
| 1645 | pagenum += 1 | |
| 1646 | ||
| 1647 | if len(video_ids) > n: | |
| 1648 | video_ids = video_ids[:n] | |
| 7012b23c PH |
1649 | videos = [self.url_result(video_id, 'Youtube', video_id=video_id) |
| 1650 | for video_id in video_ids] | |
| b05654f0 | 1651 | return self.playlist_result(videos, query) |
| 75dff0ee | 1652 | |
| a3dd9248 | 1653 | class YoutubeSearchDateIE(YoutubeSearchIE): |
| cb7fb546 | 1654 | IE_NAME = YoutubeSearchIE.IE_NAME + ':date' |
| a3dd9248 CM |
1655 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published' |
| 1656 | _SEARCH_KEY = 'ytsearchdate' | |
| 08fb86c4 | 1657 | IE_DESC = u'YouTube.com searches, newest videos first' |
| 75dff0ee JMF |
1658 | |
| 1659 | class YoutubeShowIE(InfoExtractor): | |
| 0f818663 | 1660 | IE_DESC = u'YouTube.com (multi-season) shows' |
| 75dff0ee JMF |
1661 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
| 1662 | IE_NAME = u'youtube:show' | |
| 1663 | ||
| 1664 | def _real_extract(self, url): | |
| 1665 | mobj = re.match(self._VALID_URL, url) | |
| 1666 | show_name = mobj.group(1) | |
| 1667 | webpage = self._download_webpage(url, show_name, u'Downloading show webpage') | |
| 1668 | # There's one playlist for each season of the show | |
| 1669 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
| 1670 | self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) | |
| 1671 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] | |
| 04cc9617 JMF |
1672 | |
| 1673 | ||
| b2e8bc1b | 1674 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): |
| d7ae0639 JMF |
1675 | """ |
| 1676 | Base class for extractors that fetch info from | |
| 1677 | http://www.youtube.com/feed_ajax | |
| 1678 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
| 1679 | """ | |
| b2e8bc1b | 1680 | _LOGIN_REQUIRED = True |
| 43ba5456 JMF |
1681 | # use action_load_personal_feed instead of action_load_system_feed |
| 1682 | _PERSONAL_FEED = False | |
| 04cc9617 | 1683 | |
| d7ae0639 JMF |
1684 | @property |
| 1685 | def _FEED_TEMPLATE(self): | |
| 43ba5456 JMF |
1686 | action = 'action_load_system_feed' |
| 1687 | if self._PERSONAL_FEED: | |
| 1688 | action = 'action_load_personal_feed' | |
| 1689 | return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) | |
| d7ae0639 JMF |
1690 | |
| 1691 | @property | |
| 1692 | def IE_NAME(self): | |
| 1693 | return u'youtube:%s' % self._FEED_NAME | |
| 04cc9617 | 1694 | |
| 81f0259b | 1695 | def _real_initialize(self): |
| b2e8bc1b | 1696 | self._login() |
| 81f0259b | 1697 | |
| 04cc9617 JMF |
1698 | def _real_extract(self, url): |
| 1699 | feed_entries = [] | |
| 0e44d838 JMF |
1700 | paging = 0 |
| 1701 | for i in itertools.count(1): | |
| d7ae0639 JMF |
1702 | info = self._download_webpage(self._FEED_TEMPLATE % paging, |
| 1703 | u'%s feed' % self._FEED_NAME, | |
| 04cc9617 JMF |
1704 | u'Downloading page %s' % i) |
| 1705 | info = json.loads(info) | |
| 1706 | feed_html = info['feed_html'] | |
| 43ba5456 | 1707 | m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) |
| 04cc9617 | 1708 | ids = orderedSet(m.group(1) for m in m_ids) |
| 7012b23c PH |
1709 | feed_entries.extend( |
| 1710 | self.url_result(video_id, 'Youtube', video_id=video_id) | |
| 1711 | for video_id in ids) | |
| 04cc9617 JMF |
1712 | if info['paging'] is None: |
| 1713 | break | |
| 0e44d838 | 1714 | paging = info['paging'] |
| d7ae0639 JMF |
1715 | return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) |
| 1716 | ||
| 1717 | class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): | |
| 1718 | IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' | |
| 1719 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
| 1720 | _FEED_NAME = 'subscriptions' | |
| 1721 | _PLAYLIST_TITLE = u'Youtube Subscriptions' | |
| 1722 | ||
| 1723 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): | |
| 1724 | IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' | |
| 1725 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' | |
| 1726 | _FEED_NAME = 'recommended' | |
| 1727 | _PLAYLIST_TITLE = u'Youtube Recommended videos' | |
| c626a3d9 | 1728 | |
| 43ba5456 JMF |
1729 | class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): |
| 1730 | IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' | |
| 1731 | _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' | |
| 1732 | _FEED_NAME = 'watch_later' | |
| 1733 | _PLAYLIST_TITLE = u'Youtube Watch Later' | |
| 43ba5456 | 1734 | _PERSONAL_FEED = True |
| c626a3d9 | 1735 | |
| f459d170 JMF |
1736 | class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): |
| 1737 | IE_DESC = u'Youtube watch history, "ythistory" keyword (requires authentication)' | |
| 1738 | _VALID_URL = u'https?://www\.youtube\.com/feed/history|:ythistory' | |
| 1739 | _FEED_NAME = 'history' | |
| 1740 | _PERSONAL_FEED = True | |
| 1741 | _PLAYLIST_TITLE = u'Youtube Watch History' | |
| 1742 | ||
| c626a3d9 JMF |
1743 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): |
| 1744 | IE_NAME = u'youtube:favorites' | |
| 1745 | IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' | |
| c7a7750d | 1746 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?' |
| c626a3d9 JMF |
1747 | _LOGIN_REQUIRED = True |
| 1748 | ||
| 1749 | def _real_extract(self, url): | |
| 1750 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
| 1751 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id') | |
| 1752 | return self.url_result(playlist_id, 'YoutubePlaylist') | |
| 15870e90 PH |
1753 | |
| 1754 | ||
| 1755 | class YoutubeTruncatedURLIE(InfoExtractor): | |
| 1756 | IE_NAME = 'youtube:truncated_url' | |
| 1757 | IE_DESC = False # Do not list | |
| 1758 | _VALID_URL = r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$' | |
| 1759 | ||
| 1760 | def _real_extract(self, url): | |
| 1761 | raise ExtractorError( | |
| 1762 | u'Did you forget to quote the URL? Remember that & is a meta ' | |
| 1763 | u'character in most shells, so you want to put the URL in quotes, ' | |
| 1764 | u'like youtube-dl ' | |
| b4622a32 PH |
1765 | u'"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' |
| 1766 | u' or simply youtube-dl BaW_jenozKc .', | |
| 15870e90 | 1767 | expected=True) |