]>
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 | 29 | ExtractorError, |
dd27fd17 | 30 | int_or_none, |
b7ab0590 | 31 | PagedList, |
c91778f8 | 32 | RegexNotFoundError, |
c5e8d7af PH |
33 | unescapeHTML, |
34 | unified_strdate, | |
04cc9617 | 35 | orderedSet, |
edf3e38e | 36 | write_json_file, |
c5e8d7af PH |
37 | ) |
38 | ||
de7f3446 | 39 | class YoutubeBaseInfoExtractor(InfoExtractor): |
b2e8bc1b JMF |
40 | """Provide base functions for Youtube extractors""" |
41 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
42 | _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' | |
38c2e5b8 | 43 | _AGE_URL = 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' |
b2e8bc1b JMF |
44 | _NETRC_MACHINE = 'youtube' |
45 | # If True it will raise an error if no login info is provided | |
46 | _LOGIN_REQUIRED = False | |
47 | ||
b2e8bc1b | 48 | def _set_language(self): |
7cc3570e PH |
49 | return bool(self._download_webpage( |
50 | self._LANG_URL, None, | |
51 | note=u'Setting language', errnote='unable to set language', | |
52 | fatal=False)) | |
b2e8bc1b JMF |
53 | |
54 | def _login(self): | |
55 | (username, password) = self._get_login_info() | |
56 | # No authentication to be performed | |
57 | if username is None: | |
58 | if self._LOGIN_REQUIRED: | |
59 | raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) | |
60 | return False | |
61 | ||
7cc3570e PH |
62 | login_page = self._download_webpage( |
63 | self._LOGIN_URL, None, | |
64 | note=u'Downloading login page', | |
65 | errnote=u'unable to fetch login page', fatal=False) | |
66 | if login_page is False: | |
67 | return | |
b2e8bc1b | 68 | |
795f28f8 PH |
69 | galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"', |
70 | login_page, u'Login GALX parameter') | |
c5e8d7af | 71 | |
b2e8bc1b JMF |
72 | # Log in |
73 | login_form_strs = { | |
74 | u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
75 | u'Email': username, | |
76 | u'GALX': galx, | |
77 | u'Passwd': password, | |
78 | u'PersistentCookie': u'yes', | |
79 | u'_utf8': u'霱', | |
80 | u'bgresponse': u'js_disabled', | |
81 | u'checkConnection': u'', | |
82 | u'checkedDomains': u'youtube', | |
83 | u'dnConn': u'', | |
b2e8bc1b JMF |
84 | u'pstMsg': u'0', |
85 | u'rmShown': u'1', | |
86 | u'secTok': u'', | |
87 | u'signIn': u'Sign in', | |
88 | u'timeStmp': u'', | |
89 | u'service': u'youtube', | |
90 | u'uilel': u'3', | |
91 | u'hl': u'en_US', | |
92 | } | |
93 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
94 | # chokes on unicode | |
95 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) | |
96 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
7cc3570e PH |
97 | |
98 | req = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
99 | login_results = self._download_webpage( | |
100 | req, None, | |
101 | note=u'Logging in', errnote=u'unable to log in', fatal=False) | |
102 | if login_results is False: | |
103 | return False | |
104 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
105 | self._downloader.report_warning(u'unable to log in: bad username or password') | |
b2e8bc1b JMF |
106 | return False |
107 | return True | |
108 | ||
109 | def _confirm_age(self): | |
110 | age_form = { | |
7cc3570e PH |
111 | 'next_url': '/', |
112 | 'action_confirm': 'Confirm', | |
113 | } | |
5700e779 JMF |
114 | req = compat_urllib_request.Request(self._AGE_URL, |
115 | compat_urllib_parse.urlencode(age_form).encode('ascii')) | |
7cc3570e PH |
116 | |
117 | self._download_webpage( | |
118 | req, None, | |
119 | note=u'Confirming age', errnote=u'Unable to confirm age') | |
b2e8bc1b JMF |
120 | return True |
121 | ||
122 | def _real_initialize(self): | |
123 | if self._downloader is None: | |
124 | return | |
125 | if not self._set_language(): | |
126 | return | |
127 | if not self._login(): | |
128 | return | |
129 | self._confirm_age() | |
c5e8d7af | 130 | |
8377574c | 131 | |
de7f3446 | 132 | class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): |
0f818663 | 133 | IE_DESC = u'YouTube.com' |
cb7dfeea | 134 | _VALID_URL = r"""(?x)^ |
c5e8d7af | 135 | ( |
83aa5293 | 136 | (?:https?://|//)? # http(s):// or protocol-independent URL (optional) |
cb7dfeea | 137 | (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| |
484aaeb2 PH |
138 | (?:www\.)?deturl\.com/www\.youtube\.com/| |
139 | (?:www\.)?pwnyoutube\.com| | |
e69ae5b9 JMF |
140 | tube\.majestyc\.net/| |
141 | youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains | |
c5e8d7af PH |
142 | (?:.*?\#/)? # handle anchor (#/) redirect urls |
143 | (?: # the various things that can precede the ID: | |
144 | (?:(?:v|embed|e)/) # v/ or embed/ or e/ | |
145 | |(?: # or the v= param in all its forms | |
d741e55a | 146 | (?:(?:watch|movie)(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) |
c5e8d7af PH |
147 | (?:\?|\#!?) # the params delimiter ? or # or #! |
148 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
149 | v= | |
150 | ) | |
f4b05232 JMF |
151 | )) |
152 | |youtu\.be/ # just youtu.be/xxxx | |
153 | ) | |
c5e8d7af | 154 | )? # all until now is optional -> you can pass the naked ID |
8963d9c2 | 155 | ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID |
c5e8d7af PH |
156 | (?(1).+)? # if we found the ID, everything can follow |
157 | $""" | |
c5e8d7af | 158 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' |
2c62dc26 PH |
159 | _formats = { |
160 | '5': {'ext': 'flv', 'width': 400, 'height': 240}, | |
161 | '6': {'ext': 'flv', 'width': 450, 'height': 270}, | |
162 | '13': {'ext': '3gp'}, | |
163 | '17': {'ext': '3gp', 'width': 176, 'height': 144}, | |
164 | '18': {'ext': 'mp4', 'width': 640, 'height': 360}, | |
165 | '22': {'ext': 'mp4', 'width': 1280, 'height': 720}, | |
166 | '34': {'ext': 'flv', 'width': 640, 'height': 360}, | |
167 | '35': {'ext': 'flv', 'width': 854, 'height': 480}, | |
168 | '36': {'ext': '3gp', 'width': 320, 'height': 240}, | |
169 | '37': {'ext': 'mp4', 'width': 1920, 'height': 1080}, | |
170 | '38': {'ext': 'mp4', 'width': 4096, 'height': 3072}, | |
171 | '43': {'ext': 'webm', 'width': 640, 'height': 360}, | |
172 | '44': {'ext': 'webm', 'width': 854, 'height': 480}, | |
173 | '45': {'ext': 'webm', 'width': 1280, 'height': 720}, | |
174 | '46': {'ext': 'webm', 'width': 1920, 'height': 1080}, | |
175 | ||
1d043b93 | 176 | |
86fe61c8 | 177 | # 3d videos |
2c62dc26 PH |
178 | '82': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': '3D', 'preference': -20}, |
179 | '83': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': '3D', 'preference': -20}, | |
180 | '84': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': '3D', 'preference': -20}, | |
181 | '85': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': '3D', 'preference': -20}, | |
182 | '100': {'ext': 'webm', 'height': 360, 'resolution': '360p', 'format_note': '3D', 'preference': -20}, | |
183 | '101': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': '3D', 'preference': -20}, | |
184 | '102': {'ext': 'webm', 'height': 720, 'resolution': '720p', 'format_note': '3D', 'preference': -20}, | |
836a086c | 185 | |
96fb5605 | 186 | # Apple HTTP Live Streaming |
2c62dc26 PH |
187 | '92': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'HLS', 'preference': -10}, |
188 | '93': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': 'HLS', 'preference': -10}, | |
189 | '94': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': 'HLS', 'preference': -10}, | |
190 | '95': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': 'HLS', 'preference': -10}, | |
191 | '96': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'HLS', 'preference': -10}, | |
192 | '132': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'HLS', 'preference': -10}, | |
193 | '151': {'ext': 'mp4', 'height': 72, 'resolution': '72p', 'format_note': 'HLS', 'preference': -10}, | |
194 | ||
195 | # DASH mp4 video | |
196 | '133': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'DASH video', 'preference': -40}, | |
197 | '134': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': 'DASH video', 'preference': -40}, | |
198 | '135': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': 'DASH video', 'preference': -40}, | |
199 | '136': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': 'DASH video', 'preference': -40}, | |
200 | '137': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH video', 'preference': -40}, | |
201 | '138': {'ext': 'mp4', 'height': 1081, 'resolution': '>1080p', 'format_note': 'DASH video', 'preference': -40}, | |
202 | '160': {'ext': 'mp4', 'height': 192, 'resolution': '192p', 'format_note': 'DASH video', 'preference': -40}, | |
8fa8a629 | 203 | '264': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH video', 'preference': -40}, |
836a086c | 204 | |
f6f1fc92 | 205 | # Dash mp4 audio |
2c62dc26 PH |
206 | '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50}, |
207 | '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50}, | |
208 | '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50}, | |
836a086c AZ |
209 | |
210 | # Dash webm | |
1394ce65 PH |
211 | '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, |
212 | '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, | |
2c5bae42 PH |
213 | '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, |
214 | '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, | |
1394ce65 PH |
215 | '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, |
216 | '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, | |
2c62dc26 PH |
217 | '242': {'ext': 'webm', 'height': 240, 'resolution': '240p', 'format_note': 'DASH webm', 'preference': -40}, |
218 | '243': {'ext': 'webm', 'height': 360, 'resolution': '360p', 'format_note': 'DASH webm', 'preference': -40}, | |
219 | '244': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40}, | |
220 | '245': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40}, | |
221 | '246': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40}, | |
222 | '247': {'ext': 'webm', 'height': 720, 'resolution': '720p', 'format_note': 'DASH webm', 'preference': -40}, | |
223 | '248': {'ext': 'webm', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH webm', 'preference': -40}, | |
224 | ||
225 | # Dash webm audio | |
226 | '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 48, 'preference': -50}, | |
227 | '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 256, 'preference': -50}, | |
ce6b9a2d PH |
228 | |
229 | # RTMP (unnamed) | |
230 | '_rtmp': {'protocol': 'rtmp'}, | |
c5e8d7af | 231 | } |
836a086c | 232 | |
c5e8d7af | 233 | IE_NAME = u'youtube' |
2eb88d95 PH |
234 | _TESTS = [ |
235 | { | |
0e853ca4 PH |
236 | u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", |
237 | u"file": u"BaW_jenozKc.mp4", | |
238 | u"info_dict": { | |
239 | u"title": u"youtube-dl test video \"'/\\ä↭𝕐", | |
240 | u"uploader": u"Philipp Hagemeister", | |
241 | u"uploader_id": u"phihag", | |
242 | u"upload_date": u"20121002", | |
27dcce19 | 243 | 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 | 244 | } |
0e853ca4 | 245 | }, |
0e853ca4 PH |
246 | { |
247 | u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", | |
248 | u"file": u"UxxajLWwzqY.mp4", | |
249 | u"note": u"Test generic use_cipher_signature video (#897)", | |
250 | u"info_dict": { | |
251 | u"upload_date": u"20120506", | |
252 | u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", | |
2dc59299 | 253 | u"description": u"md5:5b292926389560516e384ac437c0ec07", |
45ed795c | 254 | u"uploader": u"Icona Pop", |
0e853ca4 | 255 | u"uploader_id": u"IconaPop" |
2eb88d95 | 256 | } |
c108eb73 JMF |
257 | }, |
258 | { | |
259 | u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", | |
260 | u"file": u"07FYdnEawAQ.mp4", | |
261 | u"note": u"Test VEVO video with age protection (#956)", | |
262 | u"info_dict": { | |
263 | u"upload_date": u"20130703", | |
264 | u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", | |
265 | u"description": u"md5:64249768eec3bc4276236606ea996373", | |
266 | u"uploader": u"justintimberlakeVEVO", | |
267 | u"uploader_id": u"justintimberlakeVEVO" | |
268 | } | |
269 | }, | |
fccd3771 | 270 | { |
83aa5293 | 271 | u"url": u"//www.YouTube.com/watch?v=yZIXLfi8CZQ", |
fccd3771 PH |
272 | u"file": u"yZIXLfi8CZQ.mp4", |
273 | u"note": u"Embed-only video (#1746)", | |
274 | u"info_dict": { | |
275 | u"upload_date": u"20120608", | |
276 | u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012", | |
277 | u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7", | |
278 | u"uploader": u"SET India", | |
279 | u"uploader_id": u"setindia" | |
280 | } | |
281 | }, | |
dd27fd17 PH |
282 | { |
283 | u"url": u"http://www.youtube.com/watch?v=a9LDPn-MO4I", | |
284 | u"file": u"a9LDPn-MO4I.m4a", | |
285 | u"note": u"256k DASH audio (format 141) via DASH manifest", | |
dd27fd17 PH |
286 | u"info_dict": { |
287 | u"upload_date": "20121002", | |
288 | u"uploader_id": "8KVIDEO", | |
289 | u"description": "No description available.", | |
290 | u"uploader": "8KVIDEO", | |
291 | u"title": "UHDTV TEST 8K VIDEO.mp4" | |
4919603f PH |
292 | }, |
293 | u"params": { | |
294 | u"youtube_include_dash_manifest": True, | |
295 | u"format": "141", | |
296 | }, | |
dd27fd17 | 297 | }, |
2eb88d95 PH |
298 | ] |
299 | ||
c5e8d7af PH |
300 | |
301 | @classmethod | |
302 | def suitable(cls, url): | |
303 | """Receives a URL and returns True if suitable for this IE.""" | |
e3ea4790 | 304 | if YoutubePlaylistIE.suitable(url): return False |
fccd3771 | 305 | return re.match(cls._VALID_URL, url) is not None |
c5e8d7af | 306 | |
e0df6211 PH |
307 | def __init__(self, *args, **kwargs): |
308 | super(YoutubeIE, self).__init__(*args, **kwargs) | |
83799698 | 309 | self._player_cache = {} |
e0df6211 | 310 | |
c5e8d7af PH |
311 | def report_video_info_webpage_download(self, video_id): |
312 | """Report attempt to download video info webpage.""" | |
313 | self.to_screen(u'%s: Downloading video info webpage' % video_id) | |
314 | ||
c5e8d7af PH |
315 | def report_information_extraction(self, video_id): |
316 | """Report attempt to extract video information.""" | |
317 | self.to_screen(u'%s: Extracting video information' % video_id) | |
318 | ||
319 | def report_unavailable_format(self, video_id, format): | |
320 | """Report extracted video URL.""" | |
321 | self.to_screen(u'%s: Format %s not available' % (video_id, format)) | |
322 | ||
323 | def report_rtmp_download(self): | |
324 | """Indicate the download will use the RTMP protocol.""" | |
325 | self.to_screen(u'RTMP download detected') | |
326 | ||
c4417ddb PH |
327 | def _extract_signature_function(self, video_id, player_url, slen): |
328 | id_m = re.match(r'.*-(?P<id>[a-zA-Z0-9_-]+)\.(?P<ext>[a-z]+)$', | |
83799698 | 329 | player_url) |
e0df6211 PH |
330 | player_type = id_m.group('ext') |
331 | player_id = id_m.group('id') | |
332 | ||
c4417ddb PH |
333 | # Read from filesystem cache |
334 | func_id = '%s_%s_%d' % (player_type, player_id, slen) | |
335 | assert os.path.basename(func_id) == func_id | |
c38b1e77 | 336 | cache_dir = get_cachedir(self._downloader.params) |
c4417ddb | 337 | |
c3c88a26 | 338 | cache_enabled = cache_dir is not None |
f8061589 | 339 | if cache_enabled: |
c4417ddb PH |
340 | cache_fn = os.path.join(os.path.expanduser(cache_dir), |
341 | u'youtube-sigfuncs', | |
342 | func_id + '.json') | |
343 | try: | |
edf3e38e | 344 | with io.open(cache_fn, 'r', encoding='utf-8') as cachef: |
c4417ddb PH |
345 | cache_spec = json.load(cachef) |
346 | return lambda s: u''.join(s[i] for i in cache_spec) | |
edf3e38e | 347 | except IOError: |
c4417ddb | 348 | pass # No cache available |
83799698 | 349 | |
e0df6211 PH |
350 | if player_type == 'js': |
351 | code = self._download_webpage( | |
352 | player_url, video_id, | |
83799698 | 353 | note=u'Downloading %s player %s' % (player_type, player_id), |
e0df6211 | 354 | errnote=u'Download of %s failed' % player_url) |
83799698 | 355 | res = self._parse_sig_js(code) |
c4417ddb | 356 | elif player_type == 'swf': |
e0df6211 PH |
357 | urlh = self._request_webpage( |
358 | player_url, video_id, | |
83799698 | 359 | note=u'Downloading %s player %s' % (player_type, player_id), |
e0df6211 PH |
360 | errnote=u'Download of %s failed' % player_url) |
361 | code = urlh.read() | |
83799698 | 362 | res = self._parse_sig_swf(code) |
e0df6211 PH |
363 | else: |
364 | assert False, 'Invalid player type %r' % player_type | |
365 | ||
f8061589 | 366 | if cache_enabled: |
edf3e38e | 367 | try: |
c705320f PH |
368 | test_string = u''.join(map(compat_chr, range(slen))) |
369 | cache_res = res(test_string) | |
edf3e38e PH |
370 | cache_spec = [ord(c) for c in cache_res] |
371 | try: | |
372 | os.makedirs(os.path.dirname(cache_fn)) | |
373 | except OSError as ose: | |
374 | if ose.errno != errno.EEXIST: | |
375 | raise | |
376 | write_json_file(cache_spec, cache_fn) | |
0ca96d48 | 377 | except Exception: |
edf3e38e PH |
378 | tb = traceback.format_exc() |
379 | self._downloader.report_warning( | |
380 | u'Writing cache to %r failed: %s' % (cache_fn, tb)) | |
83799698 PH |
381 | |
382 | return res | |
383 | ||
edf3e38e PH |
384 | def _print_sig_code(self, func, slen): |
385 | def gen_sig_code(idxs): | |
386 | def _genslice(start, end, step): | |
387 | starts = u'' if start == 0 else str(start) | |
e35e4ddc PH |
388 | ends = (u':%d' % (end+step)) if end + step >= 0 else u':' |
389 | steps = u'' if step == 1 else (u':%d' % step) | |
edf3e38e PH |
390 | return u's[%s%s%s]' % (starts, ends, steps) |
391 | ||
392 | step = None | |
0ca96d48 PH |
393 | start = '(Never used)' # Quelch pyflakes warnings - start will be |
394 | # set as soon as step is set | |
edf3e38e PH |
395 | for i, prev in zip(idxs[1:], idxs[:-1]): |
396 | if step is not None: | |
397 | if i - prev == step: | |
398 | continue | |
399 | yield _genslice(start, prev, step) | |
400 | step = None | |
401 | continue | |
402 | if i - prev in [-1, 1]: | |
403 | step = i - prev | |
404 | start = prev | |
405 | continue | |
406 | else: | |
407 | yield u's[%d]' % prev | |
408 | if step is None: | |
409 | yield u's[%d]' % i | |
410 | else: | |
411 | yield _genslice(start, i, step) | |
412 | ||
c705320f PH |
413 | test_string = u''.join(map(compat_chr, range(slen))) |
414 | cache_res = func(test_string) | |
edf3e38e PH |
415 | cache_spec = [ord(c) for c in cache_res] |
416 | expr_code = u' + '.join(gen_sig_code(cache_spec)) | |
417 | code = u'if len(s) == %d:\n return %s\n' % (slen, expr_code) | |
f8061589 | 418 | self.to_screen(u'Extracted signature function:\n' + code) |
edf3e38e | 419 | |
e0df6211 PH |
420 | def _parse_sig_js(self, jscode): |
421 | funcname = self._search_regex( | |
422 | r'signature=([a-zA-Z]+)', jscode, | |
423 | u'Initial JS player signature function name') | |
424 | ||
425 | functions = {} | |
426 | ||
427 | def argidx(varname): | |
428 | return string.lowercase.index(varname) | |
429 | ||
430 | def interpret_statement(stmt, local_vars, allow_recursion=20): | |
431 | if allow_recursion < 0: | |
0ca96d48 | 432 | raise ExtractorError(u'Recursion limit reached') |
e0df6211 PH |
433 | |
434 | if stmt.startswith(u'var '): | |
435 | stmt = stmt[len(u'var '):] | |
436 | ass_m = re.match(r'^(?P<out>[a-z]+)(?:\[(?P<index>[^\]]+)\])?' + | |
437 | r'=(?P<expr>.*)$', stmt) | |
438 | if ass_m: | |
439 | if ass_m.groupdict().get('index'): | |
440 | def assign(val): | |
441 | lvar = local_vars[ass_m.group('out')] | |
442 | idx = interpret_expression(ass_m.group('index'), | |
443 | local_vars, allow_recursion) | |
444 | assert isinstance(idx, int) | |
445 | lvar[idx] = val | |
446 | return val | |
447 | expr = ass_m.group('expr') | |
448 | else: | |
449 | def assign(val): | |
450 | local_vars[ass_m.group('out')] = val | |
451 | return val | |
452 | expr = ass_m.group('expr') | |
453 | elif stmt.startswith(u'return '): | |
454 | assign = lambda v: v | |
455 | expr = stmt[len(u'return '):] | |
456 | else: | |
457 | raise ExtractorError( | |
458 | u'Cannot determine left side of statement in %r' % stmt) | |
459 | ||
460 | v = interpret_expression(expr, local_vars, allow_recursion) | |
461 | return assign(v) | |
462 | ||
463 | def interpret_expression(expr, local_vars, allow_recursion): | |
464 | if expr.isdigit(): | |
465 | return int(expr) | |
466 | ||
467 | if expr.isalpha(): | |
468 | return local_vars[expr] | |
469 | ||
470 | m = re.match(r'^(?P<in>[a-z]+)\.(?P<member>.*)$', expr) | |
471 | if m: | |
472 | member = m.group('member') | |
473 | val = local_vars[m.group('in')] | |
474 | if member == 'split("")': | |
475 | return list(val) | |
476 | if member == 'join("")': | |
477 | return u''.join(val) | |
478 | if member == 'length': | |
479 | return len(val) | |
480 | if member == 'reverse()': | |
481 | return val[::-1] | |
482 | slice_m = re.match(r'slice\((?P<idx>.*)\)', member) | |
483 | if slice_m: | |
484 | idx = interpret_expression( | |
485 | slice_m.group('idx'), local_vars, allow_recursion-1) | |
486 | return val[idx:] | |
487 | ||
488 | m = re.match( | |
489 | r'^(?P<in>[a-z]+)\[(?P<idx>.+)\]$', expr) | |
490 | if m: | |
491 | val = local_vars[m.group('in')] | |
492 | idx = interpret_expression(m.group('idx'), local_vars, | |
493 | allow_recursion-1) | |
494 | return val[idx] | |
495 | ||
496 | m = re.match(r'^(?P<a>.+?)(?P<op>[%])(?P<b>.+?)$', expr) | |
497 | if m: | |
498 | a = interpret_expression(m.group('a'), | |
499 | local_vars, allow_recursion) | |
500 | b = interpret_expression(m.group('b'), | |
501 | local_vars, allow_recursion) | |
502 | return a % b | |
503 | ||
504 | m = re.match( | |
505 | r'^(?P<func>[a-zA-Z]+)\((?P<args>[a-z0-9,]+)\)$', expr) | |
506 | if m: | |
507 | fname = m.group('func') | |
508 | if fname not in functions: | |
509 | functions[fname] = extract_function(fname) | |
510 | argvals = [int(v) if v.isdigit() else local_vars[v] | |
511 | for v in m.group('args').split(',')] | |
512 | return functions[fname](argvals) | |
513 | raise ExtractorError(u'Unsupported JS expression %r' % expr) | |
514 | ||
515 | def extract_function(funcname): | |
516 | func_m = re.search( | |
517 | r'function ' + re.escape(funcname) + | |
518 | r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}', | |
519 | jscode) | |
520 | argnames = func_m.group('args').split(',') | |
521 | ||
522 | def resf(args): | |
523 | local_vars = dict(zip(argnames, args)) | |
524 | for stmt in func_m.group('code').split(';'): | |
525 | res = interpret_statement(stmt, local_vars) | |
526 | return res | |
527 | return resf | |
528 | ||
529 | initial_function = extract_function(funcname) | |
530 | return lambda s: initial_function([s]) | |
531 | ||
532 | def _parse_sig_swf(self, file_contents): | |
533 | if file_contents[1:3] != b'WS': | |
534 | raise ExtractorError( | |
535 | u'Not an SWF file; header is %r' % file_contents[:3]) | |
536 | if file_contents[:1] == b'C': | |
537 | content = zlib.decompress(file_contents[8:]) | |
538 | else: | |
539 | raise NotImplementedError(u'Unsupported compression format %r' % | |
540 | file_contents[:1]) | |
541 | ||
542 | def extract_tags(content): | |
543 | pos = 0 | |
544 | while pos < len(content): | |
545 | header16 = struct.unpack('<H', content[pos:pos+2])[0] | |
546 | pos += 2 | |
547 | tag_code = header16 >> 6 | |
548 | tag_len = header16 & 0x3f | |
549 | if tag_len == 0x3f: | |
550 | tag_len = struct.unpack('<I', content[pos:pos+4])[0] | |
551 | pos += 4 | |
552 | assert pos+tag_len <= len(content) | |
553 | yield (tag_code, content[pos:pos+tag_len]) | |
554 | pos += tag_len | |
555 | ||
556 | code_tag = next(tag | |
557 | for tag_code, tag in extract_tags(content) | |
558 | if tag_code == 82) | |
559 | p = code_tag.index(b'\0', 4) + 1 | |
ba552f54 | 560 | code_reader = io.BytesIO(code_tag[p:]) |
e0df6211 PH |
561 | |
562 | # Parse ABC (AVM2 ByteCode) | |
ba552f54 PH |
563 | def read_int(reader=None): |
564 | if reader is None: | |
565 | reader = code_reader | |
e0df6211 PH |
566 | res = 0 |
567 | shift = 0 | |
568 | for _ in range(5): | |
ba552f54 PH |
569 | buf = reader.read(1) |
570 | assert len(buf) == 1 | |
571 | b = struct.unpack('<B', buf)[0] | |
e0df6211 PH |
572 | res = res | ((b & 0x7f) << shift) |
573 | if b & 0x80 == 0: | |
574 | break | |
575 | shift += 7 | |
ba552f54 PH |
576 | return res |
577 | ||
578 | def u30(reader=None): | |
579 | res = read_int(reader) | |
580 | assert res & 0xf0000000 == 0 | |
e0df6211 PH |
581 | return res |
582 | u32 = read_int | |
583 | ||
ba552f54 PH |
584 | def s32(reader=None): |
585 | v = read_int(reader) | |
e0df6211 PH |
586 | if v & 0x80000000 != 0: |
587 | v = - ((v ^ 0xffffffff) + 1) | |
ba552f54 PH |
588 | return v |
589 | ||
0ca96d48 | 590 | def read_string(reader=None): |
ba552f54 PH |
591 | if reader is None: |
592 | reader = code_reader | |
593 | slen = u30(reader) | |
594 | resb = reader.read(slen) | |
595 | assert len(resb) == slen | |
596 | return resb.decode('utf-8') | |
597 | ||
598 | def read_bytes(count, reader=None): | |
599 | if reader is None: | |
600 | reader = code_reader | |
601 | resb = reader.read(count) | |
602 | assert len(resb) == count | |
603 | return resb | |
604 | ||
605 | def read_byte(reader=None): | |
606 | resb = read_bytes(1, reader=reader) | |
607 | res = struct.unpack('<B', resb)[0] | |
608 | return res | |
e0df6211 PH |
609 | |
610 | # minor_version + major_version | |
0ca96d48 | 611 | read_bytes(2 + 2) |
e0df6211 PH |
612 | |
613 | # Constant pool | |
ba552f54 | 614 | int_count = u30() |
e0df6211 | 615 | for _c in range(1, int_count): |
0ca96d48 | 616 | s32() |
ba552f54 | 617 | uint_count = u30() |
e0df6211 | 618 | for _c in range(1, uint_count): |
0ca96d48 | 619 | u32() |
ba552f54 | 620 | double_count = u30() |
0ca96d48 | 621 | read_bytes((double_count-1) * 8) |
ba552f54 | 622 | string_count = u30() |
e0df6211 PH |
623 | constant_strings = [u''] |
624 | for _c in range(1, string_count): | |
0ca96d48 | 625 | s = read_string() |
e0df6211 | 626 | constant_strings.append(s) |
ba552f54 | 627 | namespace_count = u30() |
e0df6211 | 628 | for _c in range(1, namespace_count): |
0ca96d48 PH |
629 | read_bytes(1) # kind |
630 | u30() # name | |
ba552f54 | 631 | ns_set_count = u30() |
e0df6211 | 632 | for _c in range(1, ns_set_count): |
ba552f54 | 633 | count = u30() |
e0df6211 | 634 | for _c2 in range(count): |
0ca96d48 | 635 | u30() |
ba552f54 | 636 | multiname_count = u30() |
e0df6211 PH |
637 | MULTINAME_SIZES = { |
638 | 0x07: 2, # QName | |
639 | 0x0d: 2, # QNameA | |
640 | 0x0f: 1, # RTQName | |
641 | 0x10: 1, # RTQNameA | |
642 | 0x11: 0, # RTQNameL | |
643 | 0x12: 0, # RTQNameLA | |
644 | 0x09: 2, # Multiname | |
645 | 0x0e: 2, # MultinameA | |
646 | 0x1b: 1, # MultinameL | |
647 | 0x1c: 1, # MultinameLA | |
648 | } | |
649 | multinames = [u''] | |
650 | for _c in range(1, multiname_count): | |
ba552f54 | 651 | kind = u30() |
e0df6211 PH |
652 | assert kind in MULTINAME_SIZES, u'Invalid multiname kind %r' % kind |
653 | if kind == 0x07: | |
0ca96d48 | 654 | u30() # namespace_idx |
ba552f54 | 655 | name_idx = u30() |
e0df6211 PH |
656 | multinames.append(constant_strings[name_idx]) |
657 | else: | |
658 | multinames.append('[MULTINAME kind: %d]' % kind) | |
659 | for _c2 in range(MULTINAME_SIZES[kind]): | |
0ca96d48 | 660 | u30() |
e0df6211 PH |
661 | |
662 | # Methods | |
ba552f54 | 663 | method_count = u30() |
e0df6211 PH |
664 | MethodInfo = collections.namedtuple( |
665 | 'MethodInfo', | |
666 | ['NEED_ARGUMENTS', 'NEED_REST']) | |
667 | method_infos = [] | |
668 | for method_id in range(method_count): | |
ba552f54 | 669 | param_count = u30() |
0ca96d48 | 670 | u30() # return type |
e0df6211 | 671 | for _ in range(param_count): |
0ca96d48 PH |
672 | u30() # param type |
673 | u30() # name index (always 0 for youtube) | |
ba552f54 | 674 | flags = read_byte() |
e0df6211 PH |
675 | if flags & 0x08 != 0: |
676 | # Options present | |
ba552f54 | 677 | option_count = u30() |
e0df6211 | 678 | for c in range(option_count): |
0ca96d48 PH |
679 | u30() # val |
680 | read_bytes(1) # kind | |
e0df6211 PH |
681 | if flags & 0x80 != 0: |
682 | # Param names present | |
683 | for _ in range(param_count): | |
0ca96d48 | 684 | u30() # param name |
e0df6211 PH |
685 | mi = MethodInfo(flags & 0x01 != 0, flags & 0x04 != 0) |
686 | method_infos.append(mi) | |
687 | ||
688 | # Metadata | |
ba552f54 | 689 | metadata_count = u30() |
e0df6211 | 690 | for _c in range(metadata_count): |
0ca96d48 | 691 | u30() # name |
ba552f54 | 692 | item_count = u30() |
e0df6211 | 693 | for _c2 in range(item_count): |
0ca96d48 PH |
694 | u30() # key |
695 | u30() # value | |
ba552f54 PH |
696 | |
697 | def parse_traits_info(): | |
698 | trait_name_idx = u30() | |
699 | kind_full = read_byte() | |
e0df6211 PH |
700 | kind = kind_full & 0x0f |
701 | attrs = kind_full >> 4 | |
702 | methods = {} | |
703 | if kind in [0x00, 0x06]: # Slot or Const | |
0ca96d48 PH |
704 | u30() # Slot id |
705 | u30() # type_name_idx | |
ba552f54 | 706 | vindex = u30() |
e0df6211 | 707 | if vindex != 0: |
0ca96d48 | 708 | read_byte() # vkind |
e0df6211 | 709 | elif kind in [0x01, 0x02, 0x03]: # Method / Getter / Setter |
0ca96d48 | 710 | u30() # disp_id |
ba552f54 | 711 | method_idx = u30() |
e0df6211 PH |
712 | methods[multinames[trait_name_idx]] = method_idx |
713 | elif kind == 0x04: # Class | |
0ca96d48 PH |
714 | u30() # slot_id |
715 | u30() # classi | |
e0df6211 | 716 | elif kind == 0x05: # Function |
0ca96d48 | 717 | u30() # slot_id |
ba552f54 | 718 | function_idx = u30() |
e0df6211 PH |
719 | methods[function_idx] = multinames[trait_name_idx] |
720 | else: | |
721 | raise ExtractorError(u'Unsupported trait kind %d' % kind) | |
722 | ||
723 | if attrs & 0x4 != 0: # Metadata present | |
ba552f54 | 724 | metadata_count = u30() |
e0df6211 | 725 | for _c3 in range(metadata_count): |
0ca96d48 | 726 | u30() # metadata index |
e0df6211 | 727 | |
ba552f54 | 728 | return methods |
e0df6211 PH |
729 | |
730 | # Classes | |
731 | TARGET_CLASSNAME = u'SignatureDecipher' | |
732 | searched_idx = multinames.index(TARGET_CLASSNAME) | |
733 | searched_class_id = None | |
ba552f54 | 734 | class_count = u30() |
e0df6211 | 735 | for class_id in range(class_count): |
ba552f54 | 736 | name_idx = u30() |
e0df6211 PH |
737 | if name_idx == searched_idx: |
738 | # We found the class we're looking for! | |
739 | searched_class_id = class_id | |
0ca96d48 | 740 | u30() # super_name idx |
ba552f54 | 741 | flags = read_byte() |
e0df6211 | 742 | if flags & 0x08 != 0: # Protected namespace is present |
0ca96d48 | 743 | u30() # protected_ns_idx |
ba552f54 | 744 | intrf_count = u30() |
e0df6211 | 745 | for _c2 in range(intrf_count): |
0ca96d48 PH |
746 | u30() |
747 | u30() # iinit | |
ba552f54 | 748 | trait_count = u30() |
e0df6211 | 749 | for _c2 in range(trait_count): |
0ca96d48 | 750 | parse_traits_info() |
e0df6211 PH |
751 | |
752 | if searched_class_id is None: | |
753 | raise ExtractorError(u'Target class %r not found' % | |
754 | TARGET_CLASSNAME) | |
755 | ||
756 | method_names = {} | |
757 | method_idxs = {} | |
758 | for class_id in range(class_count): | |
0ca96d48 | 759 | u30() # cinit |
ba552f54 | 760 | trait_count = u30() |
e0df6211 | 761 | for _c2 in range(trait_count): |
ba552f54 | 762 | trait_methods = parse_traits_info() |
e0df6211 PH |
763 | if class_id == searched_class_id: |
764 | method_names.update(trait_methods.items()) | |
765 | method_idxs.update(dict( | |
766 | (idx, name) | |
767 | for name, idx in trait_methods.items())) | |
768 | ||
769 | # Scripts | |
ba552f54 | 770 | script_count = u30() |
e0df6211 | 771 | for _c in range(script_count): |
0ca96d48 | 772 | u30() # init |
ba552f54 | 773 | trait_count = u30() |
e0df6211 | 774 | for _c2 in range(trait_count): |
0ca96d48 | 775 | parse_traits_info() |
e0df6211 PH |
776 | |
777 | # Method bodies | |
ba552f54 | 778 | method_body_count = u30() |
e0df6211 PH |
779 | Method = collections.namedtuple('Method', ['code', 'local_count']) |
780 | methods = {} | |
781 | for _c in range(method_body_count): | |
ba552f54 | 782 | method_idx = u30() |
0ca96d48 | 783 | u30() # max_stack |
ba552f54 | 784 | local_count = u30() |
0ca96d48 PH |
785 | u30() # init_scope_depth |
786 | u30() # max_scope_depth | |
ba552f54 PH |
787 | code_length = u30() |
788 | code = read_bytes(code_length) | |
e0df6211 | 789 | if method_idx in method_idxs: |
ba552f54 | 790 | m = Method(code, local_count) |
e0df6211 | 791 | methods[method_idxs[method_idx]] = m |
ba552f54 | 792 | exception_count = u30() |
e0df6211 | 793 | for _c2 in range(exception_count): |
0ca96d48 PH |
794 | u30() # from |
795 | u30() # to | |
796 | u30() # target | |
797 | u30() # exc_type | |
798 | u30() # var_name | |
ba552f54 | 799 | trait_count = u30() |
e0df6211 | 800 | for _c2 in range(trait_count): |
0ca96d48 | 801 | parse_traits_info() |
e0df6211 | 802 | |
ba552f54 | 803 | assert p + code_reader.tell() == len(code_tag) |
e0df6211 PH |
804 | assert len(methods) == len(method_idxs) |
805 | ||
806 | method_pyfunctions = {} | |
807 | ||
808 | def extract_function(func_name): | |
809 | if func_name in method_pyfunctions: | |
810 | return method_pyfunctions[func_name] | |
811 | if func_name not in methods: | |
812 | raise ExtractorError(u'Cannot find function %r' % func_name) | |
813 | m = methods[func_name] | |
814 | ||
815 | def resfunc(args): | |
e0df6211 PH |
816 | registers = ['(this)'] + list(args) + [None] * m.local_count |
817 | stack = [] | |
818 | coder = io.BytesIO(m.code) | |
819 | while True: | |
820 | opcode = struct.unpack('!B', coder.read(1))[0] | |
a7177865 | 821 | if opcode == 36: # pushbyte |
e0df6211 PH |
822 | v = struct.unpack('!B', coder.read(1))[0] |
823 | stack.append(v) | |
824 | elif opcode == 44: # pushstring | |
825 | idx = u30(coder) | |
826 | stack.append(constant_strings[idx]) | |
827 | elif opcode == 48: # pushscope | |
828 | # We don't implement the scope register, so we'll just | |
829 | # ignore the popped value | |
830 | stack.pop() | |
831 | elif opcode == 70: # callproperty | |
832 | index = u30(coder) | |
833 | mname = multinames[index] | |
834 | arg_count = u30(coder) | |
835 | args = list(reversed( | |
836 | [stack.pop() for _ in range(arg_count)])) | |
837 | obj = stack.pop() | |
838 | if mname == u'split': | |
839 | assert len(args) == 1 | |
840 | assert isinstance(args[0], compat_str) | |
841 | assert isinstance(obj, compat_str) | |
842 | if args[0] == u'': | |
843 | res = list(obj) | |
844 | else: | |
845 | res = obj.split(args[0]) | |
846 | stack.append(res) | |
a7177865 PH |
847 | elif mname == u'slice': |
848 | assert len(args) == 1 | |
849 | assert isinstance(args[0], int) | |
850 | assert isinstance(obj, list) | |
851 | res = obj[args[0]:] | |
852 | stack.append(res) | |
853 | elif mname == u'join': | |
854 | assert len(args) == 1 | |
855 | assert isinstance(args[0], compat_str) | |
856 | assert isinstance(obj, list) | |
857 | res = args[0].join(obj) | |
858 | stack.append(res) | |
e0df6211 PH |
859 | elif mname in method_pyfunctions: |
860 | stack.append(method_pyfunctions[mname](args)) | |
861 | else: | |
862 | raise NotImplementedError( | |
863 | u'Unsupported property %r on %r' | |
864 | % (mname, obj)) | |
a7177865 PH |
865 | elif opcode == 72: # returnvalue |
866 | res = stack.pop() | |
867 | return res | |
868 | elif opcode == 79: # callpropvoid | |
869 | index = u30(coder) | |
870 | mname = multinames[index] | |
871 | arg_count = u30(coder) | |
872 | args = list(reversed( | |
873 | [stack.pop() for _ in range(arg_count)])) | |
874 | obj = stack.pop() | |
875 | if mname == u'reverse': | |
876 | assert isinstance(obj, list) | |
877 | obj.reverse() | |
878 | else: | |
879 | raise NotImplementedError( | |
880 | u'Unsupported (void) property %r on %r' | |
881 | % (mname, obj)) | |
e0df6211 PH |
882 | elif opcode == 93: # findpropstrict |
883 | index = u30(coder) | |
884 | mname = multinames[index] | |
885 | res = extract_function(mname) | |
886 | stack.append(res) | |
887 | elif opcode == 97: # setproperty | |
888 | index = u30(coder) | |
889 | value = stack.pop() | |
890 | idx = stack.pop() | |
891 | obj = stack.pop() | |
892 | assert isinstance(obj, list) | |
893 | assert isinstance(idx, int) | |
894 | obj[idx] = value | |
895 | elif opcode == 98: # getlocal | |
896 | index = u30(coder) | |
897 | stack.append(registers[index]) | |
898 | elif opcode == 99: # setlocal | |
899 | index = u30(coder) | |
900 | value = stack.pop() | |
901 | registers[index] = value | |
902 | elif opcode == 102: # getproperty | |
903 | index = u30(coder) | |
904 | pname = multinames[index] | |
905 | if pname == u'length': | |
906 | obj = stack.pop() | |
907 | assert isinstance(obj, list) | |
908 | stack.append(len(obj)) | |
909 | else: # Assume attribute access | |
910 | idx = stack.pop() | |
911 | assert isinstance(idx, int) | |
912 | obj = stack.pop() | |
913 | assert isinstance(obj, list) | |
914 | stack.append(obj[idx]) | |
915 | elif opcode == 128: # coerce | |
0ca96d48 | 916 | u30(coder) |
e0df6211 PH |
917 | elif opcode == 133: # coerce_s |
918 | assert isinstance(stack[-1], (type(None), compat_str)) | |
919 | elif opcode == 164: # modulo | |
920 | value2 = stack.pop() | |
921 | value1 = stack.pop() | |
922 | res = value1 % value2 | |
923 | stack.append(res) | |
a7177865 PH |
924 | elif opcode == 208: # getlocal_0 |
925 | stack.append(registers[0]) | |
926 | elif opcode == 209: # getlocal_1 | |
927 | stack.append(registers[1]) | |
928 | elif opcode == 210: # getlocal_2 | |
929 | stack.append(registers[2]) | |
930 | elif opcode == 211: # getlocal_3 | |
931 | stack.append(registers[3]) | |
e0df6211 PH |
932 | elif opcode == 214: # setlocal_2 |
933 | registers[2] = stack.pop() | |
934 | elif opcode == 215: # setlocal_3 | |
935 | registers[3] = stack.pop() | |
936 | else: | |
937 | raise NotImplementedError( | |
938 | u'Unsupported opcode %d' % opcode) | |
939 | ||
940 | method_pyfunctions[func_name] = resfunc | |
941 | return resfunc | |
942 | ||
943 | initial_function = extract_function(u'decipher') | |
944 | return lambda s: initial_function([s]) | |
945 | ||
83799698 | 946 | def _decrypt_signature(self, s, video_id, player_url, age_gate=False): |
257a2501 | 947 | """Turn the encrypted s field into a working signature""" |
6b37f0be | 948 | |
83799698 | 949 | if player_url is not None: |
9f9be844 PH |
950 | if player_url.startswith(u'//'): |
951 | player_url = u'https:' + player_url | |
e0df6211 | 952 | try: |
7f8ae73a PH |
953 | player_id = (player_url, len(s)) |
954 | if player_id not in self._player_cache: | |
83799698 | 955 | func = self._extract_signature_function( |
c4417ddb | 956 | video_id, player_url, len(s) |
e0df6211 | 957 | ) |
7f8ae73a PH |
958 | self._player_cache[player_id] = func |
959 | func = self._player_cache[player_id] | |
edf3e38e PH |
960 | if self._downloader.params.get('youtube_print_sig_code'): |
961 | self._print_sig_code(func, len(s)) | |
962 | return func(s) | |
0ca96d48 | 963 | except Exception: |
e0df6211 | 964 | tb = traceback.format_exc() |
83799698 PH |
965 | self._downloader.report_warning( |
966 | u'Automatic signature extraction failed: ' + tb) | |
e0df6211 | 967 | |
d2d8f895 PH |
968 | self._downloader.report_warning( |
969 | u'Warning: Falling back to static signature algorithm') | |
920de7a2 | 970 | |
2f2ffea9 PH |
971 | return self._static_decrypt_signature( |
972 | s, video_id, player_url, age_gate) | |
e0df6211 | 973 | |
2f2ffea9 | 974 | def _static_decrypt_signature(self, s, video_id, player_url, age_gate): |
e0df6211 PH |
975 | if age_gate: |
976 | # The videos with age protection use another player, so the | |
977 | # algorithms can be different. | |
978 | if len(s) == 86: | |
979 | return s[2:63] + s[82] + s[64:82] + s[63] | |
980 | ||
bc4b9008 | 981 | if len(s) == 93: |
982 | return s[86:29:-1] + s[88] + s[28:5:-1] | |
983 | elif len(s) == 92: | |
444b1165 | 984 | return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83] |
38d025b3 JMF |
985 | elif len(s) == 91: |
986 | return s[84:27:-1] + s[86] + s[26:5:-1] | |
444b1165 JMF |
987 | elif len(s) == 90: |
988 | return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81] | |
8a9d86a2 | 989 | elif len(s) == 89: |
990 | return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1] | |
444b1165 | 991 | elif len(s) == 88: |
3e223834 | 992 | return s[7:28] + s[87] + s[29:45] + s[55] + s[46:55] + s[2] + s[56:87] + s[28] |
be547e1d | 993 | elif len(s) == 87: |
3a725669 | 994 | return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:] |
be547e1d | 995 | elif len(s) == 86: |
f2c327fd | 996 | return s[80:72:-1] + s[16] + s[71:39:-1] + s[72] + s[38:16:-1] + s[82] + s[15::-1] |
be547e1d | 997 | elif len(s) == 85: |
6ae8ee3f | 998 | return s[3:11] + s[0] + s[12:55] + s[84] + s[56:84] |
be547e1d | 999 | elif len(s) == 84: |
6f56389b | 1000 | return s[78:70:-1] + s[14] + s[69:37:-1] + s[70] + s[36:14:-1] + s[80] + s[:14][::-1] |
be547e1d | 1001 | elif len(s) == 83: |
920de7a2 | 1002 | return s[80:63:-1] + s[0] + s[62:0:-1] + s[63] |
be547e1d | 1003 | elif len(s) == 82: |
c21315f2 | 1004 | return s[80:37:-1] + s[7] + s[36:7:-1] + s[0] + s[6:0:-1] + s[37] |
be547e1d | 1005 | elif len(s) == 81: |
aedd6bb9 | 1006 | 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 |
1007 | elif len(s) == 80: |
1008 | return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80] | |
5c468ca8 JMF |
1009 | elif len(s) == 79: |
1010 | 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 |
1011 | |
1012 | else: | |
1013 | raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) | |
c5e8d7af | 1014 | |
1f343eaa | 1015 | def _get_available_subtitles(self, video_id, webpage): |
de7f3446 | 1016 | try: |
7fad1c63 | 1017 | sub_list = self._download_webpage( |
38c2e5b8 | 1018 | 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, |
7fad1c63 JMF |
1019 | video_id, note=False) |
1020 | except ExtractorError as err: | |
de7f3446 JMF |
1021 | self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) |
1022 | return {} | |
1023 | lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
1024 | ||
1025 | sub_lang_list = {} | |
1026 | for l in lang_list: | |
1027 | lang = l[1] | |
1028 | params = compat_urllib_parse.urlencode({ | |
1029 | 'lang': lang, | |
1030 | 'v': video_id, | |
ca715127 | 1031 | 'fmt': self._downloader.params.get('subtitlesformat', 'srt'), |
c3197e3e | 1032 | 'name': unescapeHTML(l[0]).encode('utf-8'), |
de7f3446 | 1033 | }) |
38c2e5b8 | 1034 | url = u'https://www.youtube.com/api/timedtext?' + params |
de7f3446 JMF |
1035 | sub_lang_list[lang] = url |
1036 | if not sub_lang_list: | |
1037 | self._downloader.report_warning(u'video doesn\'t have subtitles') | |
1038 | return {} | |
1039 | return sub_lang_list | |
1040 | ||
055e6f36 | 1041 | def _get_available_automatic_caption(self, video_id, webpage): |
de7f3446 JMF |
1042 | """We need the webpage for getting the captions url, pass it as an |
1043 | argument to speed up the process.""" | |
ca715127 | 1044 | sub_format = self._downloader.params.get('subtitlesformat', 'srt') |
de7f3446 JMF |
1045 | self.to_screen(u'%s: Looking for automatic captions' % video_id) |
1046 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
055e6f36 | 1047 | err_msg = u'Couldn\'t find automatic captions for %s' % video_id |
de7f3446 JMF |
1048 | if mobj is None: |
1049 | self._downloader.report_warning(err_msg) | |
1050 | return {} | |
1051 | player_config = json.loads(mobj.group(1)) | |
1052 | try: | |
1053 | args = player_config[u'args'] | |
1054 | caption_url = args[u'ttsurl'] | |
1055 | timestamp = args[u'timestamp'] | |
055e6f36 JMF |
1056 | # We get the available subtitles |
1057 | list_params = compat_urllib_parse.urlencode({ | |
1058 | 'type': 'list', | |
1059 | 'tlangs': 1, | |
1060 | 'asrs': 1, | |
de7f3446 | 1061 | }) |
055e6f36 | 1062 | list_url = caption_url + '&' + list_params |
e26f8712 | 1063 | caption_list = self._download_xml(list_url, video_id) |
e3dc22ca | 1064 | original_lang_node = caption_list.find('track') |
f6a54188 | 1065 | if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' : |
e3dc22ca JMF |
1066 | self._downloader.report_warning(u'Video doesn\'t have automatic captions') |
1067 | return {} | |
1068 | original_lang = original_lang_node.attrib['lang_code'] | |
055e6f36 JMF |
1069 | |
1070 | sub_lang_list = {} | |
1071 | for lang_node in caption_list.findall('target'): | |
1072 | sub_lang = lang_node.attrib['lang_code'] | |
1073 | params = compat_urllib_parse.urlencode({ | |
1074 | 'lang': original_lang, | |
1075 | 'tlang': sub_lang, | |
1076 | 'fmt': sub_format, | |
1077 | 'ts': timestamp, | |
1078 | 'kind': 'asr', | |
1079 | }) | |
1080 | sub_lang_list[sub_lang] = caption_url + '&' + params | |
1081 | return sub_lang_list | |
de7f3446 JMF |
1082 | # An extractor error can be raise by the download process if there are |
1083 | # no automatic captions but there are subtitles | |
1084 | except (KeyError, ExtractorError): | |
1085 | self._downloader.report_warning(err_msg) | |
1086 | return {} | |
1087 | ||
c5e8d7af PH |
1088 | def _extract_id(self, url): |
1089 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
1090 | if mobj is None: | |
1091 | raise ExtractorError(u'Invalid URL: %s' % url) | |
1092 | video_id = mobj.group(2) | |
1093 | return video_id | |
1094 | ||
1d043b93 JMF |
1095 | def _extract_from_m3u8(self, manifest_url, video_id): |
1096 | url_map = {} | |
1097 | def _get_urls(_manifest): | |
1098 | lines = _manifest.split('\n') | |
1099 | urls = filter(lambda l: l and not l.startswith('#'), | |
1100 | lines) | |
1101 | return urls | |
1102 | manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') | |
1103 | formats_urls = _get_urls(manifest) | |
1104 | for format_url in formats_urls: | |
890f62e8 | 1105 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') |
1d043b93 JMF |
1106 | url_map[itag] = format_url |
1107 | return url_map | |
1108 | ||
1fb07d10 JG |
1109 | def _extract_annotations(self, video_id): |
1110 | url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id | |
1111 | return self._download_webpage(url, video_id, note=u'Searching for annotations.', errnote=u'Unable to download video annotations.') | |
1112 | ||
c5e8d7af PH |
1113 | def _real_extract(self, url): |
1114 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter | |
1115 | mobj = re.search(self._NEXT_URL_RE, url) | |
1116 | if mobj: | |
1117 | url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
1118 | video_id = self._extract_id(url) | |
1119 | ||
1120 | # Get video webpage | |
c5e8d7af | 1121 | url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id |
336c3a69 | 1122 | video_webpage = self._download_webpage(url, video_id) |
c5e8d7af PH |
1123 | |
1124 | # Attempt to extract SWF player URL | |
e0df6211 | 1125 | mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) |
c5e8d7af PH |
1126 | if mobj is not None: |
1127 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
1128 | else: | |
1129 | player_url = None | |
1130 | ||
1131 | # Get video info | |
1132 | self.report_video_info_webpage_download(video_id) | |
c108eb73 JMF |
1133 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
1134 | self.report_age_confirmation() | |
1135 | age_gate = True | |
1136 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
1137 | # this can be viewed without login into Youtube | |
1138 | data = compat_urllib_parse.urlencode({'video_id': video_id, | |
fccd3771 | 1139 | 'el': 'player_embedded', |
c108eb73 JMF |
1140 | 'gl': 'US', |
1141 | 'hl': 'en', | |
1142 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
1143 | 'asv': 3, | |
1144 | 'sts':'1588', | |
1145 | }) | |
1146 | video_info_url = 'https://www.youtube.com/get_video_info?' + data | |
c5e8d7af PH |
1147 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
1148 | note=False, | |
1149 | errnote='unable to download video info webpage') | |
1150 | video_info = compat_parse_qs(video_info_webpage) | |
c108eb73 JMF |
1151 | else: |
1152 | age_gate = False | |
1153 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
1154 | video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
1155 | % (video_id, el_type)) | |
1156 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
1157 | note=False, | |
1158 | errnote='unable to download video info webpage') | |
1159 | video_info = compat_parse_qs(video_info_webpage) | |
1160 | if 'token' in video_info: | |
1161 | break | |
c5e8d7af PH |
1162 | if 'token' not in video_info: |
1163 | if 'reason' in video_info: | |
9a82b238 | 1164 | raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True) |
c5e8d7af PH |
1165 | else: |
1166 | raise ExtractorError(u'"token" parameter not in video info for unknown reason') | |
1167 | ||
1d699755 PH |
1168 | if 'view_count' in video_info: |
1169 | view_count = int(video_info['view_count'][0]) | |
1170 | else: | |
1171 | view_count = None | |
1172 | ||
c5e8d7af PH |
1173 | # Check for "rental" videos |
1174 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
1175 | raise ExtractorError(u'"rental" videos not supported') | |
1176 | ||
1177 | # Start extracting information | |
1178 | self.report_information_extraction(video_id) | |
1179 | ||
1180 | # uploader | |
1181 | if 'author' not in video_info: | |
1182 | raise ExtractorError(u'Unable to extract uploader name') | |
1183 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
1184 | ||
1185 | # uploader_id | |
1186 | video_uploader_id = None | |
1187 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
1188 | if mobj is not None: | |
1189 | video_uploader_id = mobj.group(1) | |
1190 | else: | |
1191 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
1192 | ||
1193 | # title | |
a8c6b241 PH |
1194 | if 'title' in video_info: |
1195 | video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) | |
1196 | else: | |
1197 | self._downloader.report_warning(u'Unable to extract video title') | |
1198 | video_title = u'_' | |
c5e8d7af PH |
1199 | |
1200 | # thumbnail image | |
7763b04e JMF |
1201 | # We try first to get a high quality image: |
1202 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
1203 | video_webpage, re.DOTALL) | |
1204 | if m_thumb is not None: | |
1205 | video_thumbnail = m_thumb.group(1) | |
1206 | elif 'thumbnail_url' not in video_info: | |
c5e8d7af | 1207 | self._downloader.report_warning(u'unable to extract video thumbnail') |
f490e77e | 1208 | video_thumbnail = None |
c5e8d7af PH |
1209 | else: # don't panic if we can't find it |
1210 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
1211 | ||
1212 | # upload date | |
1213 | upload_date = None | |
1214 | mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL) | |
1215 | if mobj is not None: | |
1216 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
1217 | upload_date = unified_strdate(upload_date) | |
1218 | ||
1219 | # description | |
1220 | video_description = get_element_by_id("eow-description", video_webpage) | |
1221 | if video_description: | |
27dcce19 PH |
1222 | video_description = re.sub(r'''(?x) |
1223 | <a\s+ | |
1224 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
1225 | title="([^"]+)"\s+ | |
1226 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
1227 | class="yt-uix-redirect-link"\s*> | |
1228 | [^<]+ | |
1229 | </a> | |
1230 | ''', r'\1', video_description) | |
c5e8d7af PH |
1231 | video_description = clean_html(video_description) |
1232 | else: | |
1233 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
1234 | if fd_mobj: | |
1235 | video_description = unescapeHTML(fd_mobj.group(1)) | |
1236 | else: | |
1237 | video_description = u'' | |
1238 | ||
336c3a69 | 1239 | def _extract_count(klass): |
46374a56 PH |
1240 | count = self._search_regex( |
1241 | r'class="%s">([\d,]+)</span>' % re.escape(klass), | |
1242 | video_webpage, klass, default=None) | |
336c3a69 JMF |
1243 | if count is not None: |
1244 | return int(count.replace(',', '')) | |
1245 | return None | |
1246 | like_count = _extract_count(u'likes-count') | |
1247 | dislike_count = _extract_count(u'dislikes-count') | |
1248 | ||
c5e8d7af | 1249 | # subtitles |
d82134c3 | 1250 | video_subtitles = self.extract_subtitles(video_id, video_webpage) |
c5e8d7af | 1251 | |
c5e8d7af | 1252 | if self._downloader.params.get('listsubtitles', False): |
d665f8d3 | 1253 | self._list_available_subtitles(video_id, video_webpage) |
c5e8d7af PH |
1254 | return |
1255 | ||
1256 | if 'length_seconds' not in video_info: | |
1257 | self._downloader.report_warning(u'unable to extract video duration') | |
b466b702 | 1258 | video_duration = None |
c5e8d7af | 1259 | else: |
b466b702 | 1260 | video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) |
c5e8d7af | 1261 | |
1fb07d10 JG |
1262 | # annotations |
1263 | video_annotations = None | |
1264 | if self._downloader.params.get('writeannotations', False): | |
1265 | video_annotations = self._extract_annotations(video_id) | |
1266 | ||
c5e8d7af | 1267 | # Decide which formats to download |
c5e8d7af PH |
1268 | try: |
1269 | mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) | |
50be92c1 PH |
1270 | if not mobj: |
1271 | raise ValueError('Could not find vevo ID') | |
c5e8d7af PH |
1272 | info = json.loads(mobj.group(1)) |
1273 | args = info['args'] | |
7ce7e394 JMF |
1274 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
1275 | # this signatures are encrypted | |
44d46655 | 1276 | if 'url_encoded_fmt_stream_map' not in args: |
f10503db | 1277 | raise ValueError(u'No stream_map present') # caught below |
00fe14fc JMF |
1278 | re_signature = re.compile(r'[&,]s=') |
1279 | m_s = re_signature.search(args['url_encoded_fmt_stream_map']) | |
7ce7e394 JMF |
1280 | if m_s is not None: |
1281 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
c5e8d7af | 1282 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
00fe14fc | 1283 | m_s = re_signature.search(args.get('adaptive_fmts', u'')) |
b7a68384 | 1284 | if m_s is not None: |
00fe14fc JMF |
1285 | if 'adaptive_fmts' in video_info: |
1286 | video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts'] | |
37b6d5f6 | 1287 | else: |
00fe14fc | 1288 | video_info['adaptive_fmts'] = [args['adaptive_fmts']] |
c5e8d7af PH |
1289 | except ValueError: |
1290 | pass | |
1291 | ||
dd27fd17 PH |
1292 | def _map_to_format_list(urlmap): |
1293 | formats = [] | |
1294 | for itag, video_real_url in urlmap.items(): | |
1295 | dct = { | |
1296 | 'format_id': itag, | |
1297 | 'url': video_real_url, | |
1298 | 'player_url': player_url, | |
1299 | } | |
0b65e5d4 PH |
1300 | if itag in self._formats: |
1301 | dct.update(self._formats[itag]) | |
dd27fd17 PH |
1302 | formats.append(dct) |
1303 | return formats | |
1304 | ||
c5e8d7af PH |
1305 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): |
1306 | self.report_rtmp_download() | |
dd27fd17 PH |
1307 | formats = [{ |
1308 | 'format_id': '_rtmp', | |
1309 | 'protocol': 'rtmp', | |
1310 | 'url': video_info['conn'][0], | |
1311 | 'player_url': player_url, | |
1312 | }] | |
00fe14fc JMF |
1313 | elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1: |
1314 | encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0] | |
1315 | if 'rtmpe%3Dyes' in encoded_url_map: | |
a7055eb9 | 1316 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) |
c5e8d7af | 1317 | url_map = {} |
00fe14fc | 1318 | for url_data_str in encoded_url_map.split(','): |
c5e8d7af PH |
1319 | url_data = compat_parse_qs(url_data_str) |
1320 | if 'itag' in url_data and 'url' in url_data: | |
1321 | url = url_data['url'][0] | |
1322 | if 'sig' in url_data: | |
1323 | url += '&signature=' + url_data['sig'][0] | |
1324 | elif 's' in url_data: | |
e0df6211 | 1325 | encrypted_sig = url_data['s'][0] |
769fda3c | 1326 | if self._downloader.params.get('verbose'): |
c108eb73 | 1327 | if age_gate: |
bdde940e PH |
1328 | if player_url is None: |
1329 | player_version = 'unknown' | |
1330 | else: | |
1331 | player_version = self._search_regex( | |
1332 | r'-(.+)\.swf$', player_url, | |
1333 | u'flash player', fatal=False) | |
e0df6211 | 1334 | player_desc = 'flash player %s' % player_version |
c108eb73 | 1335 | else: |
83799698 PH |
1336 | player_version = self._search_regex( |
1337 | r'html5player-(.+?)\.js', video_webpage, | |
c108eb73 | 1338 | 'html5 player', fatal=False) |
e0df6211 PH |
1339 | player_desc = u'html5 player %s' % player_version |
1340 | ||
1341 | parts_sizes = u'.'.join(compat_str(len(part)) for part in encrypted_sig.split('.')) | |
5a76c651 | 1342 | self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % |
e0df6211 PH |
1343 | (len(encrypted_sig), parts_sizes, url_data['itag'][0], player_desc)) |
1344 | ||
83799698 | 1345 | if not age_gate: |
e0df6211 PH |
1346 | jsplayer_url_json = self._search_regex( |
1347 | r'"assets":.+?"js":\s*("[^"]+")', | |
1348 | video_webpage, u'JS player URL') | |
83799698 | 1349 | player_url = json.loads(jsplayer_url_json) |
e0df6211 | 1350 | |
83799698 PH |
1351 | signature = self._decrypt_signature( |
1352 | encrypted_sig, video_id, player_url, age_gate) | |
c5e8d7af PH |
1353 | url += '&signature=' + signature |
1354 | if 'ratebypass' not in url: | |
1355 | url += '&ratebypass=yes' | |
1356 | url_map[url_data['itag'][0]] = url | |
dd27fd17 | 1357 | formats = _map_to_format_list(url_map) |
1d043b93 JMF |
1358 | elif video_info.get('hlsvp'): |
1359 | manifest_url = video_info['hlsvp'][0] | |
1360 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
dd27fd17 | 1361 | formats = _map_to_format_list(url_map) |
c5e8d7af | 1362 | else: |
9abb3204 | 1363 | raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') |
c5e8d7af | 1364 | |
dd27fd17 PH |
1365 | # Look for the DASH manifest |
1366 | dash_manifest_url_lst = video_info.get('dashmpd') | |
4919603f PH |
1367 | if (dash_manifest_url_lst and dash_manifest_url_lst[0] and |
1368 | self._downloader.params.get('youtube_include_dash_manifest', False)): | |
dd27fd17 PH |
1369 | try: |
1370 | dash_doc = self._download_xml( | |
1371 | dash_manifest_url_lst[0], video_id, | |
1372 | note=u'Downloading DASH manifest', | |
1373 | errnote=u'Could not download DASH manifest') | |
1374 | for r in dash_doc.findall(u'.//{urn:mpeg:DASH:schema:MPD:2011}Representation'): | |
1375 | url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') | |
1376 | if url_el is None: | |
1377 | continue | |
1378 | format_id = r.attrib['id'] | |
1379 | video_url = url_el.text | |
1380 | filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) | |
1381 | f = { | |
1382 | 'format_id': format_id, | |
1383 | 'url': video_url, | |
1384 | 'width': int_or_none(r.attrib.get('width')), | |
1385 | 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), | |
1386 | 'asr': int_or_none(r.attrib.get('audioSamplingRate')), | |
1387 | 'filesize': filesize, | |
1388 | } | |
1389 | try: | |
1390 | existing_format = next( | |
1391 | fo for fo in formats | |
1392 | if fo['format_id'] == format_id) | |
1393 | except StopIteration: | |
1394 | f.update(self._formats.get(format_id, {})) | |
1395 | formats.append(f) | |
1396 | else: | |
1397 | existing_format.update(f) | |
1398 | ||
1399 | except (ExtractorError, KeyError) as e: | |
1400 | self.report_warning(u'Skipping DASH manifest: %s' % e, video_id) | |
d80044c2 | 1401 | |
4bcc7bd1 | 1402 | self._sort_formats(formats) |
4ea3be0a | 1403 | |
1404 | return { | |
1405 | 'id': video_id, | |
1406 | 'uploader': video_uploader, | |
1407 | 'uploader_id': video_uploader_id, | |
1408 | 'upload_date': upload_date, | |
1409 | 'title': video_title, | |
1410 | 'thumbnail': video_thumbnail, | |
1411 | 'description': video_description, | |
1412 | 'subtitles': video_subtitles, | |
1413 | 'duration': video_duration, | |
1414 | 'age_limit': 18 if age_gate else 0, | |
1415 | 'annotations': video_annotations, | |
1416 | 'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id, | |
1417 | 'view_count': view_count, | |
1418 | 'like_count': like_count, | |
1419 | 'dislike_count': dislike_count, | |
1420 | 'formats': formats, | |
1421 | } | |
c5e8d7af | 1422 | |
880e1c52 | 1423 | class YoutubePlaylistIE(YoutubeBaseInfoExtractor): |
0f818663 | 1424 | IE_DESC = u'YouTube.com playlists' |
c5e8d7af PH |
1425 | _VALID_URL = r"""(?: |
1426 | (?:https?://)? | |
1427 | (?:\w+\.)? | |
1428 | youtube\.com/ | |
1429 | (?: | |
1430 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
1431 | \? (?:.*?&)*? (?:p|a|list)= | |
1432 | | p/ | |
1433 | ) | |
715c8e7b | 1434 | ((?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}) |
c5e8d7af PH |
1435 | .* |
1436 | | | |
715c8e7b | 1437 | ((?:PL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,}) |
c5e8d7af | 1438 | )""" |
dcbb4580 JMF |
1439 | _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s' |
1440 | _MORE_PAGES_INDICATOR = r'data-link-type="next"' | |
6e47b51e | 1441 | _VIDEO_RE = r'href="/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' |
c5e8d7af PH |
1442 | IE_NAME = u'youtube:playlist' |
1443 | ||
1444 | @classmethod | |
1445 | def suitable(cls, url): | |
1446 | """Receives a URL and returns True if suitable for this IE.""" | |
1447 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
1448 | ||
880e1c52 JMF |
1449 | def _real_initialize(self): |
1450 | self._login() | |
1451 | ||
652cdaa2 JMF |
1452 | def _ids_to_results(self, ids): |
1453 | return [self.url_result(vid_id, 'Youtube', video_id=vid_id) | |
1454 | for vid_id in ids] | |
1455 | ||
1456 | def _extract_mix(self, playlist_id): | |
1457 | # The mixes are generated from a a single video | |
1458 | # the id of the playlist is just 'RD' + video_id | |
7d4afc55 | 1459 | url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id) |
652cdaa2 | 1460 | webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix') |
76d1700b JMF |
1461 | title_span = (get_element_by_attribute('class', 'title long-title', webpage) or |
1462 | get_element_by_attribute('class', 'title ', webpage)) | |
1463 | title = clean_html(title_span) | |
652cdaa2 JMF |
1464 | video_re = r'data-index="\d+".*?href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s' % re.escape(playlist_id) |
1465 | ids = orderedSet(re.findall(video_re, webpage)) | |
1466 | url_results = self._ids_to_results(ids) | |
1467 | ||
1468 | return self.playlist_result(url_results, playlist_id, title) | |
1469 | ||
c5e8d7af PH |
1470 | def _real_extract(self, url): |
1471 | # Extract playlist id | |
1472 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
1473 | if mobj is None: | |
1474 | raise ExtractorError(u'Invalid URL: %s' % url) | |
47192f92 FV |
1475 | playlist_id = mobj.group(1) or mobj.group(2) |
1476 | ||
1477 | # Check if it's a video-specific URL | |
7c61bd36 | 1478 | query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) |
47192f92 FV |
1479 | if 'v' in query_dict: |
1480 | video_id = query_dict['v'][0] | |
1481 | if self._downloader.params.get('noplaylist'): | |
1482 | self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id) | |
7012b23c | 1483 | return self.url_result(video_id, 'Youtube', video_id=video_id) |
47192f92 FV |
1484 | else: |
1485 | self.to_screen(u'Downloading playlist PL%s - add --no-playlist to just download video %s' % (playlist_id, video_id)) | |
c5e8d7af | 1486 | |
7d4afc55 | 1487 | if playlist_id.startswith('RD'): |
652cdaa2 JMF |
1488 | # Mixes require a custom extraction process |
1489 | return self._extract_mix(playlist_id) | |
0a688bc0 JMF |
1490 | if playlist_id.startswith('TL'): |
1491 | raise ExtractorError(u'For downloading YouTube.com top lists, use ' | |
1492 | u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) | |
652cdaa2 | 1493 | |
dcbb4580 JMF |
1494 | # Extract the video ids from the playlist pages |
1495 | ids = [] | |
c5e8d7af | 1496 | |
755eb032 | 1497 | for page_num in itertools.count(1): |
dcbb4580 | 1498 | url = self._TEMPLATE_URL % (playlist_id, page_num) |
c5e8d7af | 1499 | page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) |
6e47b51e JMF |
1500 | matches = re.finditer(self._VIDEO_RE, page) |
1501 | # We remove the duplicates and the link with index 0 | |
1502 | # (it's not the first video of the playlist) | |
1503 | new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') | |
dcbb4580 | 1504 | ids.extend(new_ids) |
c5e8d7af | 1505 | |
dcbb4580 | 1506 | if re.search(self._MORE_PAGES_INDICATOR, page) is None: |
c5e8d7af PH |
1507 | break |
1508 | ||
c91778f8 PH |
1509 | try: |
1510 | playlist_title = self._og_search_title(page) | |
1511 | except RegexNotFoundError: | |
1512 | self.report_warning( | |
1513 | u'Playlist page is missing OpenGraph title, falling back ...', | |
1514 | playlist_id) | |
1515 | playlist_title = self._html_search_regex( | |
1516 | r'<h1 class="pl-header-title">(.*?)</h1>', page, u'title') | |
c5e8d7af | 1517 | |
652cdaa2 | 1518 | url_results = self._ids_to_results(ids) |
dcbb4580 | 1519 | return self.playlist_result(url_results, playlist_id, playlist_title) |
c5e8d7af PH |
1520 | |
1521 | ||
0a688bc0 JMF |
1522 | class YoutubeTopListIE(YoutubePlaylistIE): |
1523 | IE_NAME = u'youtube:toplist' | |
1524 | IE_DESC = (u'YouTube.com top lists, "yttoplist:{channel}:{list title}"' | |
1525 | u' (Example: "yttoplist:music:Top Tracks")') | |
1526 | _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$' | |
1527 | ||
1528 | def _real_extract(self, url): | |
1529 | mobj = re.match(self._VALID_URL, url) | |
1530 | channel = mobj.group('chann') | |
1531 | title = mobj.group('title') | |
1532 | query = compat_urllib_parse.urlencode({'title': title}) | |
beddbc2a | 1533 | playlist_re = 'href="([^"]+?%s.*?)"' % re.escape(query) |
0a688bc0 JMF |
1534 | channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title) |
1535 | link = self._html_search_regex(playlist_re, channel_page, u'list') | |
1536 | url = compat_urlparse.urljoin('https://www.youtube.com/', link) | |
1537 | ||
1538 | video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"' | |
1539 | ids = [] | |
1540 | # sometimes the webpage doesn't contain the videos | |
1541 | # retry until we get them | |
1542 | for i in itertools.count(0): | |
1543 | msg = u'Downloading Youtube mix' | |
1544 | if i > 0: | |
1545 | msg += ', retry #%d' % i | |
1546 | webpage = self._download_webpage(url, title, msg) | |
1547 | ids = orderedSet(re.findall(video_re, webpage)) | |
1548 | if ids: | |
1549 | break | |
1550 | url_results = self._ids_to_results(ids) | |
1551 | return self.playlist_result(url_results, playlist_title=title) | |
1552 | ||
1553 | ||
c5e8d7af | 1554 | class YoutubeChannelIE(InfoExtractor): |
0f818663 | 1555 | IE_DESC = u'YouTube.com channels' |
c5e8d7af | 1556 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
c5e8d7af | 1557 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' |
38c2e5b8 | 1558 | _MORE_PAGES_URL = 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' |
c5e8d7af PH |
1559 | IE_NAME = u'youtube:channel' |
1560 | ||
1561 | def extract_videos_from_page(self, page): | |
1562 | ids_in_page = [] | |
1563 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
1564 | if mobj.group(1) not in ids_in_page: | |
1565 | ids_in_page.append(mobj.group(1)) | |
1566 | return ids_in_page | |
1567 | ||
1568 | def _real_extract(self, url): | |
1569 | # Extract channel id | |
1570 | mobj = re.match(self._VALID_URL, url) | |
1571 | if mobj is None: | |
1572 | raise ExtractorError(u'Invalid URL: %s' % url) | |
1573 | ||
1574 | # Download channel page | |
1575 | channel_id = mobj.group(1) | |
1576 | video_ids = [] | |
b9643eed JMF |
1577 | url = 'https://www.youtube.com/channel/%s/videos' % channel_id |
1578 | channel_page = self._download_webpage(url, channel_id) | |
31812a9e PH |
1579 | autogenerated = re.search(r'''(?x) |
1580 | class="[^"]*?(?: | |
1581 | channel-header-autogenerated-label| | |
1582 | yt-channel-title-autogenerated | |
1583 | )[^"]*"''', channel_page) is not None | |
c5e8d7af | 1584 | |
b9643eed JMF |
1585 | if autogenerated: |
1586 | # The videos are contained in a single page | |
1587 | # the ajax pages can't be used, they are empty | |
1588 | video_ids = self.extract_videos_from_page(channel_page) | |
1589 | else: | |
1590 | # Download all channel pages using the json-based channel_ajax query | |
1591 | for pagenum in itertools.count(1): | |
1592 | url = self._MORE_PAGES_URL % (pagenum, channel_id) | |
1593 | page = self._download_webpage(url, channel_id, | |
1594 | u'Downloading page #%s' % pagenum) | |
1595 | ||
1596 | page = json.loads(page) | |
1597 | ||
1598 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
1599 | video_ids.extend(ids_in_page) | |
1600 | ||
1601 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
1602 | break | |
c5e8d7af PH |
1603 | |
1604 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
1605 | ||
7012b23c PH |
1606 | url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id) |
1607 | for video_id in video_ids] | |
1608 | return self.playlist_result(url_entries, channel_id) | |
c5e8d7af PH |
1609 | |
1610 | ||
1611 | class YoutubeUserIE(InfoExtractor): | |
0f818663 | 1612 | IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' |
57da92b7 | 1613 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)' |
38c2e5b8 | 1614 | _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s' |
c5e8d7af | 1615 | _GDATA_PAGE_SIZE = 50 |
38c2e5b8 | 1616 | _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' |
c5e8d7af PH |
1617 | IE_NAME = u'youtube:user' |
1618 | ||
e3ea4790 | 1619 | @classmethod |
f4b05232 | 1620 | def suitable(cls, url): |
e3ea4790 JMF |
1621 | # Don't return True if the url can be extracted with other youtube |
1622 | # extractor, the regex would is too permissive and it would match. | |
1623 | other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) | |
1624 | if any(ie.suitable(url) for ie in other_ies): return False | |
f4b05232 JMF |
1625 | else: return super(YoutubeUserIE, cls).suitable(url) |
1626 | ||
c5e8d7af PH |
1627 | def _real_extract(self, url): |
1628 | # Extract username | |
1629 | mobj = re.match(self._VALID_URL, url) | |
1630 | if mobj is None: | |
1631 | raise ExtractorError(u'Invalid URL: %s' % url) | |
1632 | ||
1633 | username = mobj.group(1) | |
1634 | ||
1635 | # Download video ids using YouTube Data API. Result size per | |
1636 | # query is limited (currently to 50 videos) so we need to query | |
1637 | # page by page until there are no video ids - it means we got | |
1638 | # all of them. | |
1639 | ||
b7ab0590 | 1640 | def download_page(pagenum): |
c5e8d7af PH |
1641 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 |
1642 | ||
1643 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
b7ab0590 PH |
1644 | page = self._download_webpage( |
1645 | gdata_url, username, | |
1646 | u'Downloading video ids from %d to %d' % ( | |
1647 | start_index, start_index + self._GDATA_PAGE_SIZE)) | |
c5e8d7af | 1648 | |
fd9cf738 JMF |
1649 | try: |
1650 | response = json.loads(page) | |
1651 | except ValueError as err: | |
1652 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
71c82637 | 1653 | if 'entry' not in response['feed']: |
b7ab0590 | 1654 | return |
fd9cf738 | 1655 | |
c5e8d7af | 1656 | # Extract video identifiers |
e302f9ce PH |
1657 | entries = response['feed']['entry'] |
1658 | for entry in entries: | |
1659 | title = entry['title']['$t'] | |
1660 | video_id = entry['id']['$t'].split('/')[-1] | |
b7ab0590 | 1661 | yield { |
e302f9ce PH |
1662 | '_type': 'url', |
1663 | 'url': video_id, | |
1664 | 'ie_key': 'Youtube', | |
1665 | 'id': 'video_id', | |
1666 | 'title': title, | |
b7ab0590 PH |
1667 | } |
1668 | url_results = PagedList(download_page, self._GDATA_PAGE_SIZE) | |
c5e8d7af | 1669 | |
7012b23c PH |
1670 | return self.playlist_result(url_results, playlist_title=username) |
1671 | ||
b05654f0 PH |
1672 | |
1673 | class YoutubeSearchIE(SearchInfoExtractor): | |
0f818663 | 1674 | IE_DESC = u'YouTube.com searches' |
b05654f0 PH |
1675 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' |
1676 | _MAX_RESULTS = 1000 | |
1677 | IE_NAME = u'youtube:search' | |
1678 | _SEARCH_KEY = 'ytsearch' | |
1679 | ||
b05654f0 PH |
1680 | def _get_n_results(self, query, n): |
1681 | """Get a specified number of results for a query""" | |
1682 | ||
1683 | video_ids = [] | |
1684 | pagenum = 0 | |
1685 | limit = n | |
1686 | ||
1687 | while (50 * pagenum) < limit: | |
b05654f0 | 1688 | result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) |
7cc3570e PH |
1689 | data_json = self._download_webpage( |
1690 | result_url, video_id=u'query "%s"' % query, | |
1691 | note=u'Downloading page %s' % (pagenum + 1), | |
1692 | errnote=u'Unable to download API page') | |
1693 | data = json.loads(data_json) | |
1694 | api_response = data['data'] | |
1695 | ||
1696 | if 'items' not in api_response: | |
b05654f0 PH |
1697 | raise ExtractorError(u'[youtube] No video results') |
1698 | ||
1699 | new_ids = list(video['id'] for video in api_response['items']) | |
1700 | video_ids += new_ids | |
1701 | ||
1702 | limit = min(n, api_response['totalItems']) | |
1703 | pagenum += 1 | |
1704 | ||
1705 | if len(video_ids) > n: | |
1706 | video_ids = video_ids[:n] | |
7012b23c PH |
1707 | videos = [self.url_result(video_id, 'Youtube', video_id=video_id) |
1708 | for video_id in video_ids] | |
b05654f0 | 1709 | return self.playlist_result(videos, query) |
75dff0ee | 1710 | |
a3dd9248 | 1711 | class YoutubeSearchDateIE(YoutubeSearchIE): |
cb7fb546 | 1712 | IE_NAME = YoutubeSearchIE.IE_NAME + ':date' |
a3dd9248 CM |
1713 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published' |
1714 | _SEARCH_KEY = 'ytsearchdate' | |
08fb86c4 | 1715 | IE_DESC = u'YouTube.com searches, newest videos first' |
75dff0ee JMF |
1716 | |
1717 | class YoutubeShowIE(InfoExtractor): | |
0f818663 | 1718 | IE_DESC = u'YouTube.com (multi-season) shows' |
75dff0ee JMF |
1719 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
1720 | IE_NAME = u'youtube:show' | |
1721 | ||
1722 | def _real_extract(self, url): | |
1723 | mobj = re.match(self._VALID_URL, url) | |
1724 | show_name = mobj.group(1) | |
1725 | webpage = self._download_webpage(url, show_name, u'Downloading show webpage') | |
1726 | # There's one playlist for each season of the show | |
1727 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
1728 | self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) | |
1729 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] | |
04cc9617 JMF |
1730 | |
1731 | ||
b2e8bc1b | 1732 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): |
d7ae0639 JMF |
1733 | """ |
1734 | Base class for extractors that fetch info from | |
1735 | http://www.youtube.com/feed_ajax | |
1736 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
1737 | """ | |
b2e8bc1b | 1738 | _LOGIN_REQUIRED = True |
43ba5456 JMF |
1739 | # use action_load_personal_feed instead of action_load_system_feed |
1740 | _PERSONAL_FEED = False | |
04cc9617 | 1741 | |
d7ae0639 JMF |
1742 | @property |
1743 | def _FEED_TEMPLATE(self): | |
43ba5456 JMF |
1744 | action = 'action_load_system_feed' |
1745 | if self._PERSONAL_FEED: | |
1746 | action = 'action_load_personal_feed' | |
38c2e5b8 | 1747 | return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) |
d7ae0639 JMF |
1748 | |
1749 | @property | |
1750 | def IE_NAME(self): | |
1751 | return u'youtube:%s' % self._FEED_NAME | |
04cc9617 | 1752 | |
81f0259b | 1753 | def _real_initialize(self): |
b2e8bc1b | 1754 | self._login() |
81f0259b | 1755 | |
04cc9617 JMF |
1756 | def _real_extract(self, url): |
1757 | feed_entries = [] | |
0e44d838 JMF |
1758 | paging = 0 |
1759 | for i in itertools.count(1): | |
d7ae0639 JMF |
1760 | info = self._download_webpage(self._FEED_TEMPLATE % paging, |
1761 | u'%s feed' % self._FEED_NAME, | |
04cc9617 JMF |
1762 | u'Downloading page %s' % i) |
1763 | info = json.loads(info) | |
1764 | feed_html = info['feed_html'] | |
43ba5456 | 1765 | m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) |
04cc9617 | 1766 | ids = orderedSet(m.group(1) for m in m_ids) |
7012b23c PH |
1767 | feed_entries.extend( |
1768 | self.url_result(video_id, 'Youtube', video_id=video_id) | |
1769 | for video_id in ids) | |
04cc9617 JMF |
1770 | if info['paging'] is None: |
1771 | break | |
0e44d838 | 1772 | paging = info['paging'] |
d7ae0639 JMF |
1773 | return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) |
1774 | ||
1775 | class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): | |
1776 | IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' | |
1777 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
1778 | _FEED_NAME = 'subscriptions' | |
1779 | _PLAYLIST_TITLE = u'Youtube Subscriptions' | |
1780 | ||
1781 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): | |
1782 | IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' | |
1783 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' | |
1784 | _FEED_NAME = 'recommended' | |
1785 | _PLAYLIST_TITLE = u'Youtube Recommended videos' | |
c626a3d9 | 1786 | |
43ba5456 JMF |
1787 | class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): |
1788 | IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' | |
1789 | _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' | |
1790 | _FEED_NAME = 'watch_later' | |
1791 | _PLAYLIST_TITLE = u'Youtube Watch Later' | |
43ba5456 | 1792 | _PERSONAL_FEED = True |
c626a3d9 | 1793 | |
f459d170 JMF |
1794 | class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): |
1795 | IE_DESC = u'Youtube watch history, "ythistory" keyword (requires authentication)' | |
1796 | _VALID_URL = u'https?://www\.youtube\.com/feed/history|:ythistory' | |
1797 | _FEED_NAME = 'history' | |
1798 | _PERSONAL_FEED = True | |
1799 | _PLAYLIST_TITLE = u'Youtube Watch History' | |
1800 | ||
c626a3d9 JMF |
1801 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): |
1802 | IE_NAME = u'youtube:favorites' | |
1803 | IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' | |
c7a7750d | 1804 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?' |
c626a3d9 JMF |
1805 | _LOGIN_REQUIRED = True |
1806 | ||
1807 | def _real_extract(self, url): | |
1808 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
1809 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id') | |
1810 | return self.url_result(playlist_id, 'YoutubePlaylist') | |
15870e90 PH |
1811 | |
1812 | ||
1813 | class YoutubeTruncatedURLIE(InfoExtractor): | |
1814 | IE_NAME = 'youtube:truncated_url' | |
1815 | IE_DESC = False # Do not list | |
975d35db PH |
1816 | _VALID_URL = r'''(?x) |
1817 | (?:https?://)?[^/]+/watch\?feature=[a-z_]+$| | |
1818 | (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$ | |
1819 | ''' | |
15870e90 PH |
1820 | |
1821 | def _real_extract(self, url): | |
1822 | raise ExtractorError( | |
1823 | u'Did you forget to quote the URL? Remember that & is a meta ' | |
1824 | u'character in most shells, so you want to put the URL in quotes, ' | |
1825 | u'like youtube-dl ' | |
b4622a32 PH |
1826 | u'"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' |
1827 | u' or simply youtube-dl BaW_jenozKc .', | |
15870e90 | 1828 | expected=True) |