]>
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 struct |
11 | import traceback | |
12 | import zlib | |
c5e8d7af | 13 | |
b05654f0 | 14 | from .common import InfoExtractor, SearchInfoExtractor |
54d39d8b | 15 | from .subtitles import SubtitlesInfoExtractor |
2b25cb5d | 16 | from ..jsinterp import JSInterpreter |
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, |
c5e8d7af PH |
32 | unescapeHTML, |
33 | unified_strdate, | |
04cc9617 | 34 | orderedSet, |
edf3e38e | 35 | write_json_file, |
81c2f20b | 36 | uppercase_escape, |
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 | 138 | (?:www\.)?deturl\.com/www\.youtube\.com/| |
e70dc1d1 | 139 | (?:www\.)?pwnyoutube\.com/| |
f7000f3a | 140 | (?:www\.)?yourepeat\.com/| |
e69ae5b9 JMF |
141 | tube\.majestyc\.net/| |
142 | youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains | |
c5e8d7af PH |
143 | (?:.*?\#/)? # handle anchor (#/) redirect urls |
144 | (?: # the various things that can precede the ID: | |
145 | (?:(?:v|embed|e)/) # v/ or embed/ or e/ | |
146 | |(?: # or the v= param in all its forms | |
f7000f3a | 147 | (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) |
c5e8d7af PH |
148 | (?:\?|\#!?) # the params delimiter ? or # or #! |
149 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
150 | v= | |
151 | ) | |
f4b05232 JMF |
152 | )) |
153 | |youtu\.be/ # just youtu.be/xxxx | |
b9c76aa1 | 154 | |https?://(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= |
f4b05232 | 155 | ) |
c5e8d7af | 156 | )? # all until now is optional -> you can pass the naked ID |
8963d9c2 | 157 | ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID |
c5e8d7af PH |
158 | (?(1).+)? # if we found the ID, everything can follow |
159 | $""" | |
c5e8d7af | 160 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' |
2c62dc26 PH |
161 | _formats = { |
162 | '5': {'ext': 'flv', 'width': 400, 'height': 240}, | |
163 | '6': {'ext': 'flv', 'width': 450, 'height': 270}, | |
164 | '13': {'ext': '3gp'}, | |
165 | '17': {'ext': '3gp', 'width': 176, 'height': 144}, | |
166 | '18': {'ext': 'mp4', 'width': 640, 'height': 360}, | |
167 | '22': {'ext': 'mp4', 'width': 1280, 'height': 720}, | |
168 | '34': {'ext': 'flv', 'width': 640, 'height': 360}, | |
169 | '35': {'ext': 'flv', 'width': 854, 'height': 480}, | |
170 | '36': {'ext': '3gp', 'width': 320, 'height': 240}, | |
171 | '37': {'ext': 'mp4', 'width': 1920, 'height': 1080}, | |
172 | '38': {'ext': 'mp4', 'width': 4096, 'height': 3072}, | |
173 | '43': {'ext': 'webm', 'width': 640, 'height': 360}, | |
174 | '44': {'ext': 'webm', 'width': 854, 'height': 480}, | |
175 | '45': {'ext': 'webm', 'width': 1280, 'height': 720}, | |
176 | '46': {'ext': 'webm', 'width': 1920, 'height': 1080}, | |
177 | ||
1d043b93 | 178 | |
86fe61c8 | 179 | # 3d videos |
43b81eb9 PH |
180 | '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20}, |
181 | '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20}, | |
182 | '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20}, | |
183 | '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20}, | |
184 | '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20}, | |
185 | '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20}, | |
186 | '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20}, | |
836a086c | 187 | |
96fb5605 | 188 | # Apple HTTP Live Streaming |
43b81eb9 PH |
189 | '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}, |
190 | '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10}, | |
191 | '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10}, | |
192 | '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10}, | |
193 | '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10}, | |
194 | '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}, | |
195 | '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10}, | |
2c62dc26 PH |
196 | |
197 | # DASH mp4 video | |
43b81eb9 PH |
198 | '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, |
199 | '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
200 | '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
201 | '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
202 | '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
203 | '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
204 | '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
205 | '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
836a086c | 206 | |
f6f1fc92 | 207 | # Dash mp4 audio |
2c62dc26 PH |
208 | '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50}, |
209 | '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50}, | |
210 | '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50}, | |
836a086c AZ |
211 | |
212 | # Dash webm | |
e75cafe9 A |
213 | '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, |
214 | '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
215 | '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
216 | '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
217 | '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
218 | '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
219 | '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
220 | '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
221 | '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
222 | '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
223 | '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
224 | '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
225 | '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
3c80377b | 226 | '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, |
18061bba | 227 | '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, |
2c62dc26 PH |
228 | |
229 | # Dash webm audio | |
e75cafe9 A |
230 | '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 48, 'preference': -50}, |
231 | '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50}, | |
ce6b9a2d PH |
232 | |
233 | # RTMP (unnamed) | |
234 | '_rtmp': {'protocol': 'rtmp'}, | |
c5e8d7af | 235 | } |
836a086c | 236 | |
c5e8d7af | 237 | IE_NAME = u'youtube' |
2eb88d95 PH |
238 | _TESTS = [ |
239 | { | |
0e853ca4 PH |
240 | u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", |
241 | u"file": u"BaW_jenozKc.mp4", | |
242 | u"info_dict": { | |
243 | u"title": u"youtube-dl test video \"'/\\ä↭𝕐", | |
244 | u"uploader": u"Philipp Hagemeister", | |
245 | u"uploader_id": u"phihag", | |
246 | u"upload_date": u"20121002", | |
ad3bc6ac PH |
247 | 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 .", |
248 | u"categories": [u'Science & Technology'], | |
2eb88d95 | 249 | } |
0e853ca4 | 250 | }, |
0e853ca4 PH |
251 | { |
252 | u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", | |
253 | u"file": u"UxxajLWwzqY.mp4", | |
254 | u"note": u"Test generic use_cipher_signature video (#897)", | |
255 | u"info_dict": { | |
256 | u"upload_date": u"20120506", | |
257 | u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", | |
ba60a3eb | 258 | u"description": u"md5:fea86fda2d5a5784273df5c7cc994d9f", |
45ed795c | 259 | u"uploader": u"Icona Pop", |
0e853ca4 | 260 | u"uploader_id": u"IconaPop" |
2eb88d95 | 261 | } |
c108eb73 JMF |
262 | }, |
263 | { | |
264 | u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", | |
265 | u"file": u"07FYdnEawAQ.mp4", | |
266 | u"note": u"Test VEVO video with age protection (#956)", | |
267 | u"info_dict": { | |
268 | u"upload_date": u"20130703", | |
269 | u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", | |
270 | u"description": u"md5:64249768eec3bc4276236606ea996373", | |
271 | u"uploader": u"justintimberlakeVEVO", | |
272 | u"uploader_id": u"justintimberlakeVEVO" | |
273 | } | |
274 | }, | |
fccd3771 | 275 | { |
83aa5293 | 276 | u"url": u"//www.YouTube.com/watch?v=yZIXLfi8CZQ", |
fccd3771 PH |
277 | u"file": u"yZIXLfi8CZQ.mp4", |
278 | u"note": u"Embed-only video (#1746)", | |
279 | u"info_dict": { | |
280 | u"upload_date": u"20120608", | |
281 | u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012", | |
282 | u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7", | |
283 | u"uploader": u"SET India", | |
284 | u"uploader_id": u"setindia" | |
285 | } | |
286 | }, | |
dd27fd17 PH |
287 | { |
288 | u"url": u"http://www.youtube.com/watch?v=a9LDPn-MO4I", | |
289 | u"file": u"a9LDPn-MO4I.m4a", | |
290 | u"note": u"256k DASH audio (format 141) via DASH manifest", | |
dd27fd17 PH |
291 | u"info_dict": { |
292 | u"upload_date": "20121002", | |
293 | u"uploader_id": "8KVIDEO", | |
294 | u"description": "No description available.", | |
295 | u"uploader": "8KVIDEO", | |
296 | u"title": "UHDTV TEST 8K VIDEO.mp4" | |
4919603f PH |
297 | }, |
298 | u"params": { | |
299 | u"youtube_include_dash_manifest": True, | |
300 | u"format": "141", | |
301 | }, | |
dd27fd17 | 302 | }, |
3489b7d2 JMF |
303 | # DASH manifest with encrypted signature |
304 | { | |
305 | u'url': u'https://www.youtube.com/watch?v=IB3lcPjvWLA', | |
306 | u'info_dict': { | |
307 | u'id': u'IB3lcPjvWLA', | |
308 | u'ext': u'm4a', | |
309 | u'title': u'Afrojack - The Spark ft. Spree Wilson', | |
e00c9cf5 | 310 | u'description': u'md5:9717375db5a9a3992be4668bbf3bc0a8', |
3489b7d2 JMF |
311 | u'uploader': u'AfrojackVEVO', |
312 | u'uploader_id': u'AfrojackVEVO', | |
313 | u'upload_date': u'20131011', | |
314 | }, | |
315 | u"params": { | |
316 | u'youtube_include_dash_manifest': True, | |
317 | u'format': '141', | |
318 | }, | |
319 | }, | |
2eb88d95 PH |
320 | ] |
321 | ||
c5e8d7af PH |
322 | |
323 | @classmethod | |
324 | def suitable(cls, url): | |
325 | """Receives a URL and returns True if suitable for this IE.""" | |
e3ea4790 | 326 | if YoutubePlaylistIE.suitable(url): return False |
fccd3771 | 327 | return re.match(cls._VALID_URL, url) is not None |
c5e8d7af | 328 | |
e0df6211 PH |
329 | def __init__(self, *args, **kwargs): |
330 | super(YoutubeIE, self).__init__(*args, **kwargs) | |
83799698 | 331 | self._player_cache = {} |
e0df6211 | 332 | |
c5e8d7af PH |
333 | def report_video_info_webpage_download(self, video_id): |
334 | """Report attempt to download video info webpage.""" | |
335 | self.to_screen(u'%s: Downloading video info webpage' % video_id) | |
336 | ||
c5e8d7af PH |
337 | def report_information_extraction(self, video_id): |
338 | """Report attempt to extract video information.""" | |
339 | self.to_screen(u'%s: Extracting video information' % video_id) | |
340 | ||
341 | def report_unavailable_format(self, video_id, format): | |
342 | """Report extracted video URL.""" | |
343 | self.to_screen(u'%s: Format %s not available' % (video_id, format)) | |
344 | ||
345 | def report_rtmp_download(self): | |
346 | """Indicate the download will use the RTMP protocol.""" | |
347 | self.to_screen(u'RTMP download detected') | |
348 | ||
c4417ddb PH |
349 | def _extract_signature_function(self, video_id, player_url, slen): |
350 | id_m = re.match(r'.*-(?P<id>[a-zA-Z0-9_-]+)\.(?P<ext>[a-z]+)$', | |
83799698 | 351 | player_url) |
e0df6211 PH |
352 | player_type = id_m.group('ext') |
353 | player_id = id_m.group('id') | |
354 | ||
c4417ddb PH |
355 | # Read from filesystem cache |
356 | func_id = '%s_%s_%d' % (player_type, player_id, slen) | |
357 | assert os.path.basename(func_id) == func_id | |
c38b1e77 | 358 | cache_dir = get_cachedir(self._downloader.params) |
c4417ddb | 359 | |
c3c88a26 | 360 | cache_enabled = cache_dir is not None |
f8061589 | 361 | if cache_enabled: |
c4417ddb PH |
362 | cache_fn = os.path.join(os.path.expanduser(cache_dir), |
363 | u'youtube-sigfuncs', | |
364 | func_id + '.json') | |
365 | try: | |
edf3e38e | 366 | with io.open(cache_fn, 'r', encoding='utf-8') as cachef: |
c4417ddb PH |
367 | cache_spec = json.load(cachef) |
368 | return lambda s: u''.join(s[i] for i in cache_spec) | |
edf3e38e | 369 | except IOError: |
c4417ddb | 370 | pass # No cache available |
83799698 | 371 | |
e0df6211 PH |
372 | if player_type == 'js': |
373 | code = self._download_webpage( | |
374 | player_url, video_id, | |
83799698 | 375 | note=u'Downloading %s player %s' % (player_type, player_id), |
e0df6211 | 376 | errnote=u'Download of %s failed' % player_url) |
83799698 | 377 | res = self._parse_sig_js(code) |
c4417ddb | 378 | elif player_type == 'swf': |
e0df6211 PH |
379 | urlh = self._request_webpage( |
380 | player_url, video_id, | |
83799698 | 381 | note=u'Downloading %s player %s' % (player_type, player_id), |
e0df6211 PH |
382 | errnote=u'Download of %s failed' % player_url) |
383 | code = urlh.read() | |
83799698 | 384 | res = self._parse_sig_swf(code) |
e0df6211 PH |
385 | else: |
386 | assert False, 'Invalid player type %r' % player_type | |
387 | ||
f8061589 | 388 | if cache_enabled: |
edf3e38e | 389 | try: |
c705320f PH |
390 | test_string = u''.join(map(compat_chr, range(slen))) |
391 | cache_res = res(test_string) | |
edf3e38e PH |
392 | cache_spec = [ord(c) for c in cache_res] |
393 | try: | |
394 | os.makedirs(os.path.dirname(cache_fn)) | |
395 | except OSError as ose: | |
396 | if ose.errno != errno.EEXIST: | |
397 | raise | |
398 | write_json_file(cache_spec, cache_fn) | |
0ca96d48 | 399 | except Exception: |
edf3e38e PH |
400 | tb = traceback.format_exc() |
401 | self._downloader.report_warning( | |
402 | u'Writing cache to %r failed: %s' % (cache_fn, tb)) | |
83799698 PH |
403 | |
404 | return res | |
405 | ||
edf3e38e PH |
406 | def _print_sig_code(self, func, slen): |
407 | def gen_sig_code(idxs): | |
408 | def _genslice(start, end, step): | |
409 | starts = u'' if start == 0 else str(start) | |
e35e4ddc PH |
410 | ends = (u':%d' % (end+step)) if end + step >= 0 else u':' |
411 | steps = u'' if step == 1 else (u':%d' % step) | |
edf3e38e PH |
412 | return u's[%s%s%s]' % (starts, ends, steps) |
413 | ||
414 | step = None | |
0ca96d48 PH |
415 | start = '(Never used)' # Quelch pyflakes warnings - start will be |
416 | # set as soon as step is set | |
edf3e38e PH |
417 | for i, prev in zip(idxs[1:], idxs[:-1]): |
418 | if step is not None: | |
419 | if i - prev == step: | |
420 | continue | |
421 | yield _genslice(start, prev, step) | |
422 | step = None | |
423 | continue | |
424 | if i - prev in [-1, 1]: | |
425 | step = i - prev | |
426 | start = prev | |
427 | continue | |
428 | else: | |
429 | yield u's[%d]' % prev | |
430 | if step is None: | |
431 | yield u's[%d]' % i | |
432 | else: | |
433 | yield _genslice(start, i, step) | |
434 | ||
c705320f PH |
435 | test_string = u''.join(map(compat_chr, range(slen))) |
436 | cache_res = func(test_string) | |
edf3e38e PH |
437 | cache_spec = [ord(c) for c in cache_res] |
438 | expr_code = u' + '.join(gen_sig_code(cache_spec)) | |
439 | code = u'if len(s) == %d:\n return %s\n' % (slen, expr_code) | |
f8061589 | 440 | self.to_screen(u'Extracted signature function:\n' + code) |
edf3e38e | 441 | |
e0df6211 PH |
442 | def _parse_sig_js(self, jscode): |
443 | funcname = self._search_regex( | |
c26e9ac4 | 444 | r'signature=([$a-zA-Z]+)', jscode, |
2b25cb5d PH |
445 | u'Initial JS player signature function name') |
446 | ||
447 | jsi = JSInterpreter(jscode) | |
448 | initial_function = jsi.extract_function(funcname) | |
e0df6211 PH |
449 | return lambda s: initial_function([s]) |
450 | ||
451 | def _parse_sig_swf(self, file_contents): | |
452 | if file_contents[1:3] != b'WS': | |
453 | raise ExtractorError( | |
454 | u'Not an SWF file; header is %r' % file_contents[:3]) | |
455 | if file_contents[:1] == b'C': | |
456 | content = zlib.decompress(file_contents[8:]) | |
457 | else: | |
458 | raise NotImplementedError(u'Unsupported compression format %r' % | |
459 | file_contents[:1]) | |
460 | ||
461 | def extract_tags(content): | |
462 | pos = 0 | |
463 | while pos < len(content): | |
464 | header16 = struct.unpack('<H', content[pos:pos+2])[0] | |
465 | pos += 2 | |
466 | tag_code = header16 >> 6 | |
467 | tag_len = header16 & 0x3f | |
468 | if tag_len == 0x3f: | |
469 | tag_len = struct.unpack('<I', content[pos:pos+4])[0] | |
470 | pos += 4 | |
471 | assert pos+tag_len <= len(content) | |
472 | yield (tag_code, content[pos:pos+tag_len]) | |
473 | pos += tag_len | |
474 | ||
475 | code_tag = next(tag | |
476 | for tag_code, tag in extract_tags(content) | |
477 | if tag_code == 82) | |
478 | p = code_tag.index(b'\0', 4) + 1 | |
ba552f54 | 479 | code_reader = io.BytesIO(code_tag[p:]) |
e0df6211 PH |
480 | |
481 | # Parse ABC (AVM2 ByteCode) | |
ba552f54 PH |
482 | def read_int(reader=None): |
483 | if reader is None: | |
484 | reader = code_reader | |
e0df6211 PH |
485 | res = 0 |
486 | shift = 0 | |
487 | for _ in range(5): | |
ba552f54 PH |
488 | buf = reader.read(1) |
489 | assert len(buf) == 1 | |
490 | b = struct.unpack('<B', buf)[0] | |
e0df6211 PH |
491 | res = res | ((b & 0x7f) << shift) |
492 | if b & 0x80 == 0: | |
493 | break | |
494 | shift += 7 | |
ba552f54 PH |
495 | return res |
496 | ||
497 | def u30(reader=None): | |
498 | res = read_int(reader) | |
499 | assert res & 0xf0000000 == 0 | |
e0df6211 PH |
500 | return res |
501 | u32 = read_int | |
502 | ||
ba552f54 PH |
503 | def s32(reader=None): |
504 | v = read_int(reader) | |
e0df6211 PH |
505 | if v & 0x80000000 != 0: |
506 | v = - ((v ^ 0xffffffff) + 1) | |
ba552f54 PH |
507 | return v |
508 | ||
0ca96d48 | 509 | def read_string(reader=None): |
ba552f54 PH |
510 | if reader is None: |
511 | reader = code_reader | |
512 | slen = u30(reader) | |
513 | resb = reader.read(slen) | |
514 | assert len(resb) == slen | |
515 | return resb.decode('utf-8') | |
516 | ||
517 | def read_bytes(count, reader=None): | |
518 | if reader is None: | |
519 | reader = code_reader | |
520 | resb = reader.read(count) | |
521 | assert len(resb) == count | |
522 | return resb | |
523 | ||
524 | def read_byte(reader=None): | |
525 | resb = read_bytes(1, reader=reader) | |
526 | res = struct.unpack('<B', resb)[0] | |
527 | return res | |
e0df6211 PH |
528 | |
529 | # minor_version + major_version | |
0ca96d48 | 530 | read_bytes(2 + 2) |
e0df6211 PH |
531 | |
532 | # Constant pool | |
ba552f54 | 533 | int_count = u30() |
e0df6211 | 534 | for _c in range(1, int_count): |
0ca96d48 | 535 | s32() |
ba552f54 | 536 | uint_count = u30() |
e0df6211 | 537 | for _c in range(1, uint_count): |
0ca96d48 | 538 | u32() |
ba552f54 | 539 | double_count = u30() |
0ca96d48 | 540 | read_bytes((double_count-1) * 8) |
ba552f54 | 541 | string_count = u30() |
e0df6211 PH |
542 | constant_strings = [u''] |
543 | for _c in range(1, string_count): | |
0ca96d48 | 544 | s = read_string() |
e0df6211 | 545 | constant_strings.append(s) |
ba552f54 | 546 | namespace_count = u30() |
e0df6211 | 547 | for _c in range(1, namespace_count): |
0ca96d48 PH |
548 | read_bytes(1) # kind |
549 | u30() # name | |
ba552f54 | 550 | ns_set_count = u30() |
e0df6211 | 551 | for _c in range(1, ns_set_count): |
ba552f54 | 552 | count = u30() |
e0df6211 | 553 | for _c2 in range(count): |
0ca96d48 | 554 | u30() |
ba552f54 | 555 | multiname_count = u30() |
e0df6211 PH |
556 | MULTINAME_SIZES = { |
557 | 0x07: 2, # QName | |
558 | 0x0d: 2, # QNameA | |
559 | 0x0f: 1, # RTQName | |
560 | 0x10: 1, # RTQNameA | |
561 | 0x11: 0, # RTQNameL | |
562 | 0x12: 0, # RTQNameLA | |
563 | 0x09: 2, # Multiname | |
564 | 0x0e: 2, # MultinameA | |
565 | 0x1b: 1, # MultinameL | |
566 | 0x1c: 1, # MultinameLA | |
567 | } | |
568 | multinames = [u''] | |
569 | for _c in range(1, multiname_count): | |
ba552f54 | 570 | kind = u30() |
e0df6211 PH |
571 | assert kind in MULTINAME_SIZES, u'Invalid multiname kind %r' % kind |
572 | if kind == 0x07: | |
0ca96d48 | 573 | u30() # namespace_idx |
ba552f54 | 574 | name_idx = u30() |
e0df6211 PH |
575 | multinames.append(constant_strings[name_idx]) |
576 | else: | |
577 | multinames.append('[MULTINAME kind: %d]' % kind) | |
578 | for _c2 in range(MULTINAME_SIZES[kind]): | |
0ca96d48 | 579 | u30() |
e0df6211 PH |
580 | |
581 | # Methods | |
ba552f54 | 582 | method_count = u30() |
e0df6211 PH |
583 | MethodInfo = collections.namedtuple( |
584 | 'MethodInfo', | |
585 | ['NEED_ARGUMENTS', 'NEED_REST']) | |
586 | method_infos = [] | |
587 | for method_id in range(method_count): | |
ba552f54 | 588 | param_count = u30() |
0ca96d48 | 589 | u30() # return type |
e0df6211 | 590 | for _ in range(param_count): |
0ca96d48 PH |
591 | u30() # param type |
592 | u30() # name index (always 0 for youtube) | |
ba552f54 | 593 | flags = read_byte() |
e0df6211 PH |
594 | if flags & 0x08 != 0: |
595 | # Options present | |
ba552f54 | 596 | option_count = u30() |
e0df6211 | 597 | for c in range(option_count): |
0ca96d48 PH |
598 | u30() # val |
599 | read_bytes(1) # kind | |
e0df6211 PH |
600 | if flags & 0x80 != 0: |
601 | # Param names present | |
602 | for _ in range(param_count): | |
0ca96d48 | 603 | u30() # param name |
e0df6211 PH |
604 | mi = MethodInfo(flags & 0x01 != 0, flags & 0x04 != 0) |
605 | method_infos.append(mi) | |
606 | ||
607 | # Metadata | |
ba552f54 | 608 | metadata_count = u30() |
e0df6211 | 609 | for _c in range(metadata_count): |
0ca96d48 | 610 | u30() # name |
ba552f54 | 611 | item_count = u30() |
e0df6211 | 612 | for _c2 in range(item_count): |
0ca96d48 PH |
613 | u30() # key |
614 | u30() # value | |
ba552f54 PH |
615 | |
616 | def parse_traits_info(): | |
617 | trait_name_idx = u30() | |
618 | kind_full = read_byte() | |
e0df6211 PH |
619 | kind = kind_full & 0x0f |
620 | attrs = kind_full >> 4 | |
621 | methods = {} | |
622 | if kind in [0x00, 0x06]: # Slot or Const | |
0ca96d48 PH |
623 | u30() # Slot id |
624 | u30() # type_name_idx | |
ba552f54 | 625 | vindex = u30() |
e0df6211 | 626 | if vindex != 0: |
0ca96d48 | 627 | read_byte() # vkind |
e0df6211 | 628 | elif kind in [0x01, 0x02, 0x03]: # Method / Getter / Setter |
0ca96d48 | 629 | u30() # disp_id |
ba552f54 | 630 | method_idx = u30() |
e0df6211 PH |
631 | methods[multinames[trait_name_idx]] = method_idx |
632 | elif kind == 0x04: # Class | |
0ca96d48 PH |
633 | u30() # slot_id |
634 | u30() # classi | |
e0df6211 | 635 | elif kind == 0x05: # Function |
0ca96d48 | 636 | u30() # slot_id |
ba552f54 | 637 | function_idx = u30() |
e0df6211 PH |
638 | methods[function_idx] = multinames[trait_name_idx] |
639 | else: | |
640 | raise ExtractorError(u'Unsupported trait kind %d' % kind) | |
641 | ||
642 | if attrs & 0x4 != 0: # Metadata present | |
ba552f54 | 643 | metadata_count = u30() |
e0df6211 | 644 | for _c3 in range(metadata_count): |
0ca96d48 | 645 | u30() # metadata index |
e0df6211 | 646 | |
ba552f54 | 647 | return methods |
e0df6211 PH |
648 | |
649 | # Classes | |
650 | TARGET_CLASSNAME = u'SignatureDecipher' | |
651 | searched_idx = multinames.index(TARGET_CLASSNAME) | |
652 | searched_class_id = None | |
ba552f54 | 653 | class_count = u30() |
e0df6211 | 654 | for class_id in range(class_count): |
ba552f54 | 655 | name_idx = u30() |
e0df6211 PH |
656 | if name_idx == searched_idx: |
657 | # We found the class we're looking for! | |
658 | searched_class_id = class_id | |
0ca96d48 | 659 | u30() # super_name idx |
ba552f54 | 660 | flags = read_byte() |
e0df6211 | 661 | if flags & 0x08 != 0: # Protected namespace is present |
0ca96d48 | 662 | u30() # protected_ns_idx |
ba552f54 | 663 | intrf_count = u30() |
e0df6211 | 664 | for _c2 in range(intrf_count): |
0ca96d48 PH |
665 | u30() |
666 | u30() # iinit | |
ba552f54 | 667 | trait_count = u30() |
e0df6211 | 668 | for _c2 in range(trait_count): |
0ca96d48 | 669 | parse_traits_info() |
e0df6211 PH |
670 | |
671 | if searched_class_id is None: | |
672 | raise ExtractorError(u'Target class %r not found' % | |
673 | TARGET_CLASSNAME) | |
674 | ||
675 | method_names = {} | |
676 | method_idxs = {} | |
677 | for class_id in range(class_count): | |
0ca96d48 | 678 | u30() # cinit |
ba552f54 | 679 | trait_count = u30() |
e0df6211 | 680 | for _c2 in range(trait_count): |
ba552f54 | 681 | trait_methods = parse_traits_info() |
e0df6211 PH |
682 | if class_id == searched_class_id: |
683 | method_names.update(trait_methods.items()) | |
684 | method_idxs.update(dict( | |
685 | (idx, name) | |
686 | for name, idx in trait_methods.items())) | |
687 | ||
688 | # Scripts | |
ba552f54 | 689 | script_count = u30() |
e0df6211 | 690 | for _c in range(script_count): |
0ca96d48 | 691 | u30() # init |
ba552f54 | 692 | trait_count = u30() |
e0df6211 | 693 | for _c2 in range(trait_count): |
0ca96d48 | 694 | parse_traits_info() |
e0df6211 PH |
695 | |
696 | # Method bodies | |
ba552f54 | 697 | method_body_count = u30() |
e0df6211 PH |
698 | Method = collections.namedtuple('Method', ['code', 'local_count']) |
699 | methods = {} | |
700 | for _c in range(method_body_count): | |
ba552f54 | 701 | method_idx = u30() |
0ca96d48 | 702 | u30() # max_stack |
ba552f54 | 703 | local_count = u30() |
0ca96d48 PH |
704 | u30() # init_scope_depth |
705 | u30() # max_scope_depth | |
ba552f54 PH |
706 | code_length = u30() |
707 | code = read_bytes(code_length) | |
e0df6211 | 708 | if method_idx in method_idxs: |
ba552f54 | 709 | m = Method(code, local_count) |
e0df6211 | 710 | methods[method_idxs[method_idx]] = m |
ba552f54 | 711 | exception_count = u30() |
e0df6211 | 712 | for _c2 in range(exception_count): |
0ca96d48 PH |
713 | u30() # from |
714 | u30() # to | |
715 | u30() # target | |
716 | u30() # exc_type | |
717 | u30() # var_name | |
ba552f54 | 718 | trait_count = u30() |
e0df6211 | 719 | for _c2 in range(trait_count): |
0ca96d48 | 720 | parse_traits_info() |
e0df6211 | 721 | |
ba552f54 | 722 | assert p + code_reader.tell() == len(code_tag) |
e0df6211 PH |
723 | assert len(methods) == len(method_idxs) |
724 | ||
725 | method_pyfunctions = {} | |
726 | ||
727 | def extract_function(func_name): | |
728 | if func_name in method_pyfunctions: | |
729 | return method_pyfunctions[func_name] | |
730 | if func_name not in methods: | |
731 | raise ExtractorError(u'Cannot find function %r' % func_name) | |
732 | m = methods[func_name] | |
733 | ||
734 | def resfunc(args): | |
e0df6211 PH |
735 | registers = ['(this)'] + list(args) + [None] * m.local_count |
736 | stack = [] | |
737 | coder = io.BytesIO(m.code) | |
738 | while True: | |
739 | opcode = struct.unpack('!B', coder.read(1))[0] | |
a7177865 | 740 | if opcode == 36: # pushbyte |
e0df6211 PH |
741 | v = struct.unpack('!B', coder.read(1))[0] |
742 | stack.append(v) | |
743 | elif opcode == 44: # pushstring | |
744 | idx = u30(coder) | |
745 | stack.append(constant_strings[idx]) | |
746 | elif opcode == 48: # pushscope | |
747 | # We don't implement the scope register, so we'll just | |
748 | # ignore the popped value | |
749 | stack.pop() | |
750 | elif opcode == 70: # callproperty | |
751 | index = u30(coder) | |
752 | mname = multinames[index] | |
753 | arg_count = u30(coder) | |
754 | args = list(reversed( | |
755 | [stack.pop() for _ in range(arg_count)])) | |
756 | obj = stack.pop() | |
757 | if mname == u'split': | |
758 | assert len(args) == 1 | |
759 | assert isinstance(args[0], compat_str) | |
760 | assert isinstance(obj, compat_str) | |
761 | if args[0] == u'': | |
762 | res = list(obj) | |
763 | else: | |
764 | res = obj.split(args[0]) | |
765 | stack.append(res) | |
a7177865 PH |
766 | elif mname == u'slice': |
767 | assert len(args) == 1 | |
768 | assert isinstance(args[0], int) | |
769 | assert isinstance(obj, list) | |
770 | res = obj[args[0]:] | |
771 | stack.append(res) | |
772 | elif mname == u'join': | |
773 | assert len(args) == 1 | |
774 | assert isinstance(args[0], compat_str) | |
775 | assert isinstance(obj, list) | |
776 | res = args[0].join(obj) | |
777 | stack.append(res) | |
e0df6211 PH |
778 | elif mname in method_pyfunctions: |
779 | stack.append(method_pyfunctions[mname](args)) | |
780 | else: | |
781 | raise NotImplementedError( | |
782 | u'Unsupported property %r on %r' | |
783 | % (mname, obj)) | |
a7177865 PH |
784 | elif opcode == 72: # returnvalue |
785 | res = stack.pop() | |
786 | return res | |
787 | elif opcode == 79: # callpropvoid | |
788 | index = u30(coder) | |
789 | mname = multinames[index] | |
790 | arg_count = u30(coder) | |
791 | args = list(reversed( | |
792 | [stack.pop() for _ in range(arg_count)])) | |
793 | obj = stack.pop() | |
794 | if mname == u'reverse': | |
795 | assert isinstance(obj, list) | |
796 | obj.reverse() | |
797 | else: | |
798 | raise NotImplementedError( | |
799 | u'Unsupported (void) property %r on %r' | |
800 | % (mname, obj)) | |
e0df6211 PH |
801 | elif opcode == 93: # findpropstrict |
802 | index = u30(coder) | |
803 | mname = multinames[index] | |
804 | res = extract_function(mname) | |
805 | stack.append(res) | |
806 | elif opcode == 97: # setproperty | |
807 | index = u30(coder) | |
808 | value = stack.pop() | |
809 | idx = stack.pop() | |
810 | obj = stack.pop() | |
811 | assert isinstance(obj, list) | |
812 | assert isinstance(idx, int) | |
813 | obj[idx] = value | |
814 | elif opcode == 98: # getlocal | |
815 | index = u30(coder) | |
816 | stack.append(registers[index]) | |
817 | elif opcode == 99: # setlocal | |
818 | index = u30(coder) | |
819 | value = stack.pop() | |
820 | registers[index] = value | |
821 | elif opcode == 102: # getproperty | |
822 | index = u30(coder) | |
823 | pname = multinames[index] | |
824 | if pname == u'length': | |
825 | obj = stack.pop() | |
826 | assert isinstance(obj, list) | |
827 | stack.append(len(obj)) | |
828 | else: # Assume attribute access | |
829 | idx = stack.pop() | |
830 | assert isinstance(idx, int) | |
831 | obj = stack.pop() | |
832 | assert isinstance(obj, list) | |
833 | stack.append(obj[idx]) | |
834 | elif opcode == 128: # coerce | |
0ca96d48 | 835 | u30(coder) |
e0df6211 PH |
836 | elif opcode == 133: # coerce_s |
837 | assert isinstance(stack[-1], (type(None), compat_str)) | |
838 | elif opcode == 164: # modulo | |
839 | value2 = stack.pop() | |
840 | value1 = stack.pop() | |
841 | res = value1 % value2 | |
842 | stack.append(res) | |
a7177865 PH |
843 | elif opcode == 208: # getlocal_0 |
844 | stack.append(registers[0]) | |
845 | elif opcode == 209: # getlocal_1 | |
846 | stack.append(registers[1]) | |
847 | elif opcode == 210: # getlocal_2 | |
848 | stack.append(registers[2]) | |
849 | elif opcode == 211: # getlocal_3 | |
850 | stack.append(registers[3]) | |
e0df6211 PH |
851 | elif opcode == 214: # setlocal_2 |
852 | registers[2] = stack.pop() | |
853 | elif opcode == 215: # setlocal_3 | |
854 | registers[3] = stack.pop() | |
855 | else: | |
856 | raise NotImplementedError( | |
857 | u'Unsupported opcode %d' % opcode) | |
858 | ||
859 | method_pyfunctions[func_name] = resfunc | |
860 | return resfunc | |
861 | ||
862 | initial_function = extract_function(u'decipher') | |
863 | return lambda s: initial_function([s]) | |
864 | ||
83799698 | 865 | def _decrypt_signature(self, s, video_id, player_url, age_gate=False): |
257a2501 | 866 | """Turn the encrypted s field into a working signature""" |
6b37f0be | 867 | |
c8bf86d5 PH |
868 | if player_url is None: |
869 | raise ExtractorError(u'Cannot decrypt signature without player_url') | |
920de7a2 | 870 | |
c8bf86d5 PH |
871 | if player_url.startswith(u'//'): |
872 | player_url = u'https:' + player_url | |
873 | try: | |
874 | player_id = (player_url, len(s)) | |
875 | if player_id not in self._player_cache: | |
876 | func = self._extract_signature_function( | |
877 | video_id, player_url, len(s) | |
878 | ) | |
879 | self._player_cache[player_id] = func | |
880 | func = self._player_cache[player_id] | |
881 | if self._downloader.params.get('youtube_print_sig_code'): | |
882 | self._print_sig_code(func, len(s)) | |
883 | return func(s) | |
884 | except Exception as e: | |
885 | tb = traceback.format_exc() | |
886 | raise ExtractorError( | |
887 | u'Automatic signature extraction failed: ' + tb, cause=e) | |
e0df6211 | 888 | |
1f343eaa | 889 | def _get_available_subtitles(self, video_id, webpage): |
de7f3446 | 890 | try: |
7fad1c63 | 891 | sub_list = self._download_webpage( |
38c2e5b8 | 892 | 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, |
7fad1c63 JMF |
893 | video_id, note=False) |
894 | except ExtractorError as err: | |
de7f3446 JMF |
895 | self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) |
896 | return {} | |
897 | lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
898 | ||
899 | sub_lang_list = {} | |
900 | for l in lang_list: | |
901 | lang = l[1] | |
902 | params = compat_urllib_parse.urlencode({ | |
903 | 'lang': lang, | |
904 | 'v': video_id, | |
ca715127 | 905 | 'fmt': self._downloader.params.get('subtitlesformat', 'srt'), |
c3197e3e | 906 | 'name': unescapeHTML(l[0]).encode('utf-8'), |
de7f3446 | 907 | }) |
38c2e5b8 | 908 | url = u'https://www.youtube.com/api/timedtext?' + params |
de7f3446 JMF |
909 | sub_lang_list[lang] = url |
910 | if not sub_lang_list: | |
911 | self._downloader.report_warning(u'video doesn\'t have subtitles') | |
912 | return {} | |
913 | return sub_lang_list | |
914 | ||
055e6f36 | 915 | def _get_available_automatic_caption(self, video_id, webpage): |
de7f3446 JMF |
916 | """We need the webpage for getting the captions url, pass it as an |
917 | argument to speed up the process.""" | |
ca715127 | 918 | sub_format = self._downloader.params.get('subtitlesformat', 'srt') |
de7f3446 JMF |
919 | self.to_screen(u'%s: Looking for automatic captions' % video_id) |
920 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
055e6f36 | 921 | err_msg = u'Couldn\'t find automatic captions for %s' % video_id |
de7f3446 JMF |
922 | if mobj is None: |
923 | self._downloader.report_warning(err_msg) | |
924 | return {} | |
925 | player_config = json.loads(mobj.group(1)) | |
926 | try: | |
927 | args = player_config[u'args'] | |
928 | caption_url = args[u'ttsurl'] | |
929 | timestamp = args[u'timestamp'] | |
055e6f36 JMF |
930 | # We get the available subtitles |
931 | list_params = compat_urllib_parse.urlencode({ | |
932 | 'type': 'list', | |
933 | 'tlangs': 1, | |
934 | 'asrs': 1, | |
de7f3446 | 935 | }) |
055e6f36 | 936 | list_url = caption_url + '&' + list_params |
e26f8712 | 937 | caption_list = self._download_xml(list_url, video_id) |
e3dc22ca | 938 | original_lang_node = caption_list.find('track') |
f6a54188 | 939 | if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' : |
e3dc22ca JMF |
940 | self._downloader.report_warning(u'Video doesn\'t have automatic captions') |
941 | return {} | |
942 | original_lang = original_lang_node.attrib['lang_code'] | |
055e6f36 JMF |
943 | |
944 | sub_lang_list = {} | |
945 | for lang_node in caption_list.findall('target'): | |
946 | sub_lang = lang_node.attrib['lang_code'] | |
947 | params = compat_urllib_parse.urlencode({ | |
948 | 'lang': original_lang, | |
949 | 'tlang': sub_lang, | |
950 | 'fmt': sub_format, | |
951 | 'ts': timestamp, | |
952 | 'kind': 'asr', | |
953 | }) | |
954 | sub_lang_list[sub_lang] = caption_url + '&' + params | |
955 | return sub_lang_list | |
de7f3446 JMF |
956 | # An extractor error can be raise by the download process if there are |
957 | # no automatic captions but there are subtitles | |
958 | except (KeyError, ExtractorError): | |
959 | self._downloader.report_warning(err_msg) | |
960 | return {} | |
961 | ||
97665381 PH |
962 | @classmethod |
963 | def extract_id(cls, url): | |
964 | mobj = re.match(cls._VALID_URL, url, re.VERBOSE) | |
c5e8d7af PH |
965 | if mobj is None: |
966 | raise ExtractorError(u'Invalid URL: %s' % url) | |
967 | video_id = mobj.group(2) | |
968 | return video_id | |
969 | ||
1d043b93 JMF |
970 | def _extract_from_m3u8(self, manifest_url, video_id): |
971 | url_map = {} | |
972 | def _get_urls(_manifest): | |
973 | lines = _manifest.split('\n') | |
974 | urls = filter(lambda l: l and not l.startswith('#'), | |
975 | lines) | |
976 | return urls | |
977 | manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') | |
978 | formats_urls = _get_urls(manifest) | |
979 | for format_url in formats_urls: | |
890f62e8 | 980 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') |
1d043b93 JMF |
981 | url_map[itag] = format_url |
982 | return url_map | |
983 | ||
1fb07d10 JG |
984 | def _extract_annotations(self, video_id): |
985 | url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id | |
986 | return self._download_webpage(url, video_id, note=u'Searching for annotations.', errnote=u'Unable to download video annotations.') | |
987 | ||
c5e8d7af | 988 | def _real_extract(self, url): |
7e8c0af0 PH |
989 | proto = ( |
990 | u'http' if self._downloader.params.get('prefer_insecure', False) | |
991 | else u'https') | |
992 | ||
c5e8d7af PH |
993 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter |
994 | mobj = re.search(self._NEXT_URL_RE, url) | |
995 | if mobj: | |
7e8c0af0 | 996 | url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') |
97665381 | 997 | video_id = self.extract_id(url) |
c5e8d7af PH |
998 | |
999 | # Get video webpage | |
7e8c0af0 | 1000 | url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id |
336c3a69 | 1001 | video_webpage = self._download_webpage(url, video_id) |
c5e8d7af PH |
1002 | |
1003 | # Attempt to extract SWF player URL | |
e0df6211 | 1004 | mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) |
c5e8d7af PH |
1005 | if mobj is not None: |
1006 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
1007 | else: | |
1008 | player_url = None | |
1009 | ||
1010 | # Get video info | |
1011 | self.report_video_info_webpage_download(video_id) | |
c108eb73 JMF |
1012 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
1013 | self.report_age_confirmation() | |
1014 | age_gate = True | |
1015 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
1016 | # this can be viewed without login into Youtube | |
1017 | data = compat_urllib_parse.urlencode({'video_id': video_id, | |
fccd3771 | 1018 | 'el': 'player_embedded', |
c108eb73 JMF |
1019 | 'gl': 'US', |
1020 | 'hl': 'en', | |
1021 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
1022 | 'asv': 3, | |
1023 | 'sts':'1588', | |
1024 | }) | |
7e8c0af0 | 1025 | video_info_url = proto + '://www.youtube.com/get_video_info?' + data |
c5e8d7af PH |
1026 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
1027 | note=False, | |
1028 | errnote='unable to download video info webpage') | |
1029 | video_info = compat_parse_qs(video_info_webpage) | |
c108eb73 JMF |
1030 | else: |
1031 | age_gate = False | |
1032 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
7e8c0af0 | 1033 | video_info_url = (proto + '://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' |
c108eb73 JMF |
1034 | % (video_id, el_type)) |
1035 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
1036 | note=False, | |
1037 | errnote='unable to download video info webpage') | |
1038 | video_info = compat_parse_qs(video_info_webpage) | |
1039 | if 'token' in video_info: | |
1040 | break | |
c5e8d7af PH |
1041 | if 'token' not in video_info: |
1042 | if 'reason' in video_info: | |
d11271dd PH |
1043 | raise ExtractorError( |
1044 | u'YouTube said: %s' % video_info['reason'][0], | |
1045 | expected=True, video_id=video_id) | |
c5e8d7af | 1046 | else: |
d11271dd PH |
1047 | raise ExtractorError( |
1048 | u'"token" parameter not in video info for unknown reason', | |
1049 | video_id=video_id) | |
c5e8d7af | 1050 | |
1d699755 PH |
1051 | if 'view_count' in video_info: |
1052 | view_count = int(video_info['view_count'][0]) | |
1053 | else: | |
1054 | view_count = None | |
1055 | ||
c5e8d7af PH |
1056 | # Check for "rental" videos |
1057 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
1058 | raise ExtractorError(u'"rental" videos not supported') | |
1059 | ||
1060 | # Start extracting information | |
1061 | self.report_information_extraction(video_id) | |
1062 | ||
1063 | # uploader | |
1064 | if 'author' not in video_info: | |
1065 | raise ExtractorError(u'Unable to extract uploader name') | |
1066 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
1067 | ||
1068 | # uploader_id | |
1069 | video_uploader_id = None | |
1070 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
1071 | if mobj is not None: | |
1072 | video_uploader_id = mobj.group(1) | |
1073 | else: | |
1074 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
1075 | ||
1076 | # title | |
a8c6b241 | 1077 | if 'title' in video_info: |
aa92f063 | 1078 | video_title = video_info['title'][0] |
a8c6b241 PH |
1079 | else: |
1080 | self._downloader.report_warning(u'Unable to extract video title') | |
1081 | video_title = u'_' | |
c5e8d7af PH |
1082 | |
1083 | # thumbnail image | |
7763b04e JMF |
1084 | # We try first to get a high quality image: |
1085 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
1086 | video_webpage, re.DOTALL) | |
1087 | if m_thumb is not None: | |
1088 | video_thumbnail = m_thumb.group(1) | |
1089 | elif 'thumbnail_url' not in video_info: | |
c5e8d7af | 1090 | self._downloader.report_warning(u'unable to extract video thumbnail') |
f490e77e | 1091 | video_thumbnail = None |
c5e8d7af PH |
1092 | else: # don't panic if we can't find it |
1093 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
1094 | ||
1095 | # upload date | |
1096 | upload_date = None | |
ad3bc6ac | 1097 | mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage) |
beee53de PH |
1098 | if mobj is None: |
1099 | mobj = re.search( | |
263bd4ec | 1100 | r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>', |
beee53de | 1101 | video_webpage) |
c5e8d7af PH |
1102 | if mobj is not None: |
1103 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
1104 | upload_date = unified_strdate(upload_date) | |
1105 | ||
ec8deefc DG |
1106 | m_cat_container = get_element_by_id("eow-category", video_webpage) |
1107 | if m_cat_container: | |
ad3bc6ac | 1108 | category = self._html_search_regex( |
01ed5c9b | 1109 | r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category', |
ad3bc6ac PH |
1110 | default=None) |
1111 | video_categories = None if category is None else [category] | |
1112 | else: | |
1113 | video_categories = None | |
ec8deefc | 1114 | |
c5e8d7af PH |
1115 | # description |
1116 | video_description = get_element_by_id("eow-description", video_webpage) | |
1117 | if video_description: | |
27dcce19 PH |
1118 | video_description = re.sub(r'''(?x) |
1119 | <a\s+ | |
1120 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
1121 | title="([^"]+)"\s+ | |
1122 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
1123 | class="yt-uix-redirect-link"\s*> | |
1124 | [^<]+ | |
1125 | </a> | |
1126 | ''', r'\1', video_description) | |
c5e8d7af PH |
1127 | video_description = clean_html(video_description) |
1128 | else: | |
1129 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
1130 | if fd_mobj: | |
1131 | video_description = unescapeHTML(fd_mobj.group(1)) | |
1132 | else: | |
1133 | video_description = u'' | |
1134 | ||
336c3a69 | 1135 | def _extract_count(klass): |
46374a56 PH |
1136 | count = self._search_regex( |
1137 | r'class="%s">([\d,]+)</span>' % re.escape(klass), | |
1138 | video_webpage, klass, default=None) | |
336c3a69 JMF |
1139 | if count is not None: |
1140 | return int(count.replace(',', '')) | |
1141 | return None | |
1142 | like_count = _extract_count(u'likes-count') | |
1143 | dislike_count = _extract_count(u'dislikes-count') | |
1144 | ||
c5e8d7af | 1145 | # subtitles |
d82134c3 | 1146 | video_subtitles = self.extract_subtitles(video_id, video_webpage) |
c5e8d7af | 1147 | |
c5e8d7af | 1148 | if self._downloader.params.get('listsubtitles', False): |
d665f8d3 | 1149 | self._list_available_subtitles(video_id, video_webpage) |
c5e8d7af PH |
1150 | return |
1151 | ||
1152 | if 'length_seconds' not in video_info: | |
1153 | self._downloader.report_warning(u'unable to extract video duration') | |
b466b702 | 1154 | video_duration = None |
c5e8d7af | 1155 | else: |
b466b702 | 1156 | video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) |
c5e8d7af | 1157 | |
1fb07d10 JG |
1158 | # annotations |
1159 | video_annotations = None | |
1160 | if self._downloader.params.get('writeannotations', False): | |
1161 | video_annotations = self._extract_annotations(video_id) | |
1162 | ||
c5e8d7af | 1163 | # Decide which formats to download |
c5e8d7af | 1164 | try: |
ae7ed920 | 1165 | mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage) |
50be92c1 PH |
1166 | if not mobj: |
1167 | raise ValueError('Could not find vevo ID') | |
ae7ed920 PH |
1168 | json_code = uppercase_escape(mobj.group(1)) |
1169 | ytplayer_config = json.loads(json_code) | |
3489b7d2 | 1170 | args = ytplayer_config['args'] |
7ce7e394 JMF |
1171 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
1172 | # this signatures are encrypted | |
44d46655 | 1173 | if 'url_encoded_fmt_stream_map' not in args: |
f10503db | 1174 | raise ValueError(u'No stream_map present') # caught below |
00fe14fc JMF |
1175 | re_signature = re.compile(r'[&,]s=') |
1176 | m_s = re_signature.search(args['url_encoded_fmt_stream_map']) | |
7ce7e394 JMF |
1177 | if m_s is not None: |
1178 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
c5e8d7af | 1179 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
00fe14fc | 1180 | m_s = re_signature.search(args.get('adaptive_fmts', u'')) |
b7a68384 | 1181 | if m_s is not None: |
00fe14fc JMF |
1182 | if 'adaptive_fmts' in video_info: |
1183 | video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts'] | |
37b6d5f6 | 1184 | else: |
00fe14fc | 1185 | video_info['adaptive_fmts'] = [args['adaptive_fmts']] |
c5e8d7af PH |
1186 | except ValueError: |
1187 | pass | |
1188 | ||
dd27fd17 PH |
1189 | def _map_to_format_list(urlmap): |
1190 | formats = [] | |
1191 | for itag, video_real_url in urlmap.items(): | |
1192 | dct = { | |
1193 | 'format_id': itag, | |
1194 | 'url': video_real_url, | |
1195 | 'player_url': player_url, | |
1196 | } | |
0b65e5d4 PH |
1197 | if itag in self._formats: |
1198 | dct.update(self._formats[itag]) | |
dd27fd17 PH |
1199 | formats.append(dct) |
1200 | return formats | |
1201 | ||
c5e8d7af PH |
1202 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): |
1203 | self.report_rtmp_download() | |
dd27fd17 PH |
1204 | formats = [{ |
1205 | 'format_id': '_rtmp', | |
1206 | 'protocol': 'rtmp', | |
1207 | 'url': video_info['conn'][0], | |
1208 | 'player_url': player_url, | |
1209 | }] | |
00fe14fc JMF |
1210 | elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1: |
1211 | encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0] | |
1212 | if 'rtmpe%3Dyes' in encoded_url_map: | |
a7055eb9 | 1213 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) |
c5e8d7af | 1214 | url_map = {} |
00fe14fc | 1215 | for url_data_str in encoded_url_map.split(','): |
c5e8d7af PH |
1216 | url_data = compat_parse_qs(url_data_str) |
1217 | if 'itag' in url_data and 'url' in url_data: | |
1218 | url = url_data['url'][0] | |
1219 | if 'sig' in url_data: | |
1220 | url += '&signature=' + url_data['sig'][0] | |
1221 | elif 's' in url_data: | |
e0df6211 | 1222 | encrypted_sig = url_data['s'][0] |
769fda3c | 1223 | if self._downloader.params.get('verbose'): |
c108eb73 | 1224 | if age_gate: |
bdde940e PH |
1225 | if player_url is None: |
1226 | player_version = 'unknown' | |
1227 | else: | |
1228 | player_version = self._search_regex( | |
1229 | r'-(.+)\.swf$', player_url, | |
1230 | u'flash player', fatal=False) | |
e0df6211 | 1231 | player_desc = 'flash player %s' % player_version |
c108eb73 | 1232 | else: |
83799698 PH |
1233 | player_version = self._search_regex( |
1234 | r'html5player-(.+?)\.js', video_webpage, | |
c108eb73 | 1235 | 'html5 player', fatal=False) |
e0df6211 PH |
1236 | player_desc = u'html5 player %s' % player_version |
1237 | ||
1238 | parts_sizes = u'.'.join(compat_str(len(part)) for part in encrypted_sig.split('.')) | |
5a76c651 | 1239 | self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % |
e0df6211 PH |
1240 | (len(encrypted_sig), parts_sizes, url_data['itag'][0], player_desc)) |
1241 | ||
83799698 | 1242 | if not age_gate: |
e0df6211 PH |
1243 | jsplayer_url_json = self._search_regex( |
1244 | r'"assets":.+?"js":\s*("[^"]+")', | |
1245 | video_webpage, u'JS player URL') | |
83799698 | 1246 | player_url = json.loads(jsplayer_url_json) |
e0df6211 | 1247 | |
83799698 PH |
1248 | signature = self._decrypt_signature( |
1249 | encrypted_sig, video_id, player_url, age_gate) | |
c5e8d7af PH |
1250 | url += '&signature=' + signature |
1251 | if 'ratebypass' not in url: | |
1252 | url += '&ratebypass=yes' | |
1253 | url_map[url_data['itag'][0]] = url | |
dd27fd17 | 1254 | formats = _map_to_format_list(url_map) |
1d043b93 JMF |
1255 | elif video_info.get('hlsvp'): |
1256 | manifest_url = video_info['hlsvp'][0] | |
1257 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
dd27fd17 | 1258 | formats = _map_to_format_list(url_map) |
c5e8d7af | 1259 | else: |
9abb3204 | 1260 | raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') |
c5e8d7af | 1261 | |
dd27fd17 | 1262 | # Look for the DASH manifest |
d68f0cdb | 1263 | if (self._downloader.params.get('youtube_include_dash_manifest', False)): |
dd27fd17 | 1264 | try: |
d68f0cdb | 1265 | # The DASH manifest used needs to be the one from the original video_webpage. |
1266 | # The one found in get_video_info seems to be using different signatures. | |
1267 | # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage. | |
1268 | # Luckily, it seems, this case uses some kind of default signature (len == 86), so the | |
1269 | # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here. | |
1270 | if age_gate: | |
3489b7d2 | 1271 | dash_manifest_url = video_info.get('dashmpd')[0] |
d68f0cdb | 1272 | else: |
3489b7d2 | 1273 | dash_manifest_url = ytplayer_config['args']['dashmpd'] |
d68f0cdb | 1274 | def decrypt_sig(mobj): |
1275 | s = mobj.group(1) | |
1276 | dec_s = self._decrypt_signature(s, video_id, player_url, age_gate) | |
1277 | return '/signature/%s' % dec_s | |
1278 | dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url) | |
dd27fd17 | 1279 | dash_doc = self._download_xml( |
d68f0cdb | 1280 | dash_manifest_url, video_id, |
dd27fd17 PH |
1281 | note=u'Downloading DASH manifest', |
1282 | errnote=u'Could not download DASH manifest') | |
1283 | for r in dash_doc.findall(u'.//{urn:mpeg:DASH:schema:MPD:2011}Representation'): | |
1284 | url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') | |
1285 | if url_el is None: | |
1286 | continue | |
1287 | format_id = r.attrib['id'] | |
1288 | video_url = url_el.text | |
1289 | filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) | |
1290 | f = { | |
1291 | 'format_id': format_id, | |
1292 | 'url': video_url, | |
1293 | 'width': int_or_none(r.attrib.get('width')), | |
1294 | 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), | |
1295 | 'asr': int_or_none(r.attrib.get('audioSamplingRate')), | |
1296 | 'filesize': filesize, | |
1297 | } | |
1298 | try: | |
1299 | existing_format = next( | |
1300 | fo for fo in formats | |
1301 | if fo['format_id'] == format_id) | |
1302 | except StopIteration: | |
1303 | f.update(self._formats.get(format_id, {})) | |
1304 | formats.append(f) | |
1305 | else: | |
1306 | existing_format.update(f) | |
1307 | ||
1308 | except (ExtractorError, KeyError) as e: | |
1309 | self.report_warning(u'Skipping DASH manifest: %s' % e, video_id) | |
d80044c2 | 1310 | |
4bcc7bd1 | 1311 | self._sort_formats(formats) |
4ea3be0a | 1312 | |
1313 | return { | |
1314 | 'id': video_id, | |
1315 | 'uploader': video_uploader, | |
1316 | 'uploader_id': video_uploader_id, | |
1317 | 'upload_date': upload_date, | |
1318 | 'title': video_title, | |
1319 | 'thumbnail': video_thumbnail, | |
1320 | 'description': video_description, | |
ec8deefc | 1321 | 'categories': video_categories, |
4ea3be0a | 1322 | 'subtitles': video_subtitles, |
1323 | 'duration': video_duration, | |
1324 | 'age_limit': 18 if age_gate else 0, | |
1325 | 'annotations': video_annotations, | |
7e8c0af0 | 1326 | 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id, |
4ea3be0a | 1327 | 'view_count': view_count, |
1328 | 'like_count': like_count, | |
1329 | 'dislike_count': dislike_count, | |
1330 | 'formats': formats, | |
1331 | } | |
c5e8d7af | 1332 | |
880e1c52 | 1333 | class YoutubePlaylistIE(YoutubeBaseInfoExtractor): |
0f818663 | 1334 | IE_DESC = u'YouTube.com playlists' |
d67cc9fa | 1335 | _VALID_URL = r"""(?x)(?: |
c5e8d7af PH |
1336 | (?:https?://)? |
1337 | (?:\w+\.)? | |
1338 | youtube\.com/ | |
1339 | (?: | |
1340 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
1341 | \? (?:.*?&)*? (?:p|a|list)= | |
1342 | | p/ | |
1343 | ) | |
d67cc9fa | 1344 | ( |
7d568f5a | 1345 | (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,} |
d67cc9fa JMF |
1346 | # Top tracks, they can also include dots |
1347 | |(?:MC)[\w\.]* | |
1348 | ) | |
c5e8d7af PH |
1349 | .* |
1350 | | | |
7d568f5a | 1351 | ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,}) |
c5e8d7af | 1352 | )""" |
dbb94fb0 | 1353 | _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' |
dcbb4580 | 1354 | _MORE_PAGES_INDICATOR = r'data-link-type="next"' |
dbb94fb0 | 1355 | _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' |
c5e8d7af PH |
1356 | IE_NAME = u'youtube:playlist' |
1357 | ||
880e1c52 JMF |
1358 | def _real_initialize(self): |
1359 | self._login() | |
1360 | ||
652cdaa2 JMF |
1361 | def _ids_to_results(self, ids): |
1362 | return [self.url_result(vid_id, 'Youtube', video_id=vid_id) | |
1363 | for vid_id in ids] | |
1364 | ||
1365 | def _extract_mix(self, playlist_id): | |
1366 | # The mixes are generated from a a single video | |
1367 | # the id of the playlist is just 'RD' + video_id | |
7d4afc55 | 1368 | url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id) |
652cdaa2 | 1369 | webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix') |
bc2f773b JMF |
1370 | search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage) |
1371 | title_span = (search_title('playlist-title') or | |
1372 | search_title('title long-title') or search_title('title')) | |
76d1700b | 1373 | title = clean_html(title_span) |
70e32269 | 1374 | video_re = r'''(?x)data-video-username=".*?".*? |
bc2f773b | 1375 | href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id) |
70e32269 | 1376 | ids = orderedSet(re.findall(video_re, webpage, flags=re.DOTALL)) |
652cdaa2 JMF |
1377 | url_results = self._ids_to_results(ids) |
1378 | ||
1379 | return self.playlist_result(url_results, playlist_id, title) | |
1380 | ||
c5e8d7af PH |
1381 | def _real_extract(self, url): |
1382 | # Extract playlist id | |
d67cc9fa | 1383 | mobj = re.match(self._VALID_URL, url) |
c5e8d7af PH |
1384 | if mobj is None: |
1385 | raise ExtractorError(u'Invalid URL: %s' % url) | |
47192f92 FV |
1386 | playlist_id = mobj.group(1) or mobj.group(2) |
1387 | ||
1388 | # Check if it's a video-specific URL | |
7c61bd36 | 1389 | query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) |
47192f92 FV |
1390 | if 'v' in query_dict: |
1391 | video_id = query_dict['v'][0] | |
1392 | if self._downloader.params.get('noplaylist'): | |
1393 | self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id) | |
7012b23c | 1394 | return self.url_result(video_id, 'Youtube', video_id=video_id) |
47192f92 | 1395 | else: |
1db26669 | 1396 | self.to_screen(u'Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) |
c5e8d7af | 1397 | |
7d4afc55 | 1398 | if playlist_id.startswith('RD'): |
652cdaa2 JMF |
1399 | # Mixes require a custom extraction process |
1400 | return self._extract_mix(playlist_id) | |
0a688bc0 JMF |
1401 | if playlist_id.startswith('TL'): |
1402 | raise ExtractorError(u'For downloading YouTube.com top lists, use ' | |
1403 | u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) | |
652cdaa2 | 1404 | |
dbb94fb0 S |
1405 | url = self._TEMPLATE_URL % playlist_id |
1406 | page = self._download_webpage(url, playlist_id) | |
1407 | more_widget_html = content_html = page | |
1408 | ||
10c0e2d8 | 1409 | # Check if the playlist exists or is private |
e399853d | 1410 | if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None: |
10c0e2d8 JMF |
1411 | raise ExtractorError( |
1412 | u'The playlist doesn\'t exist or is private, use --username or ' | |
1413 | '--netrc to access it.', | |
1414 | expected=True) | |
1415 | ||
dcbb4580 JMF |
1416 | # Extract the video ids from the playlist pages |
1417 | ids = [] | |
c5e8d7af | 1418 | |
755eb032 | 1419 | for page_num in itertools.count(1): |
dbb94fb0 | 1420 | matches = re.finditer(self._VIDEO_RE, content_html) |
6e47b51e JMF |
1421 | # We remove the duplicates and the link with index 0 |
1422 | # (it's not the first video of the playlist) | |
1423 | new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') | |
dcbb4580 | 1424 | ids.extend(new_ids) |
c5e8d7af | 1425 | |
dbb94fb0 S |
1426 | mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) |
1427 | if not mobj: | |
c5e8d7af PH |
1428 | break |
1429 | ||
dbb94fb0 | 1430 | more = self._download_json( |
5912c639 PH |
1431 | 'https://youtube.com/%s' % mobj.group('more'), playlist_id, |
1432 | 'Downloading page #%s' % page_num, | |
1433 | transform_source=uppercase_escape) | |
dbb94fb0 S |
1434 | content_html = more['content_html'] |
1435 | more_widget_html = more['load_more_widget_html'] | |
1436 | ||
1437 | playlist_title = self._html_search_regex( | |
68eb8e90 PH |
1438 | r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>', |
1439 | page, u'title') | |
c5e8d7af | 1440 | |
652cdaa2 | 1441 | url_results = self._ids_to_results(ids) |
dcbb4580 | 1442 | return self.playlist_result(url_results, playlist_id, playlist_title) |
c5e8d7af PH |
1443 | |
1444 | ||
0a688bc0 JMF |
1445 | class YoutubeTopListIE(YoutubePlaylistIE): |
1446 | IE_NAME = u'youtube:toplist' | |
1447 | IE_DESC = (u'YouTube.com top lists, "yttoplist:{channel}:{list title}"' | |
1448 | u' (Example: "yttoplist:music:Top Tracks")') | |
1449 | _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$' | |
1450 | ||
1451 | def _real_extract(self, url): | |
1452 | mobj = re.match(self._VALID_URL, url) | |
1453 | channel = mobj.group('chann') | |
1454 | title = mobj.group('title') | |
1455 | query = compat_urllib_parse.urlencode({'title': title}) | |
beddbc2a | 1456 | playlist_re = 'href="([^"]+?%s.*?)"' % re.escape(query) |
0a688bc0 JMF |
1457 | channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title) |
1458 | link = self._html_search_regex(playlist_re, channel_page, u'list') | |
1459 | url = compat_urlparse.urljoin('https://www.youtube.com/', link) | |
1460 | ||
1461 | video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"' | |
1462 | ids = [] | |
1463 | # sometimes the webpage doesn't contain the videos | |
1464 | # retry until we get them | |
1465 | for i in itertools.count(0): | |
1466 | msg = u'Downloading Youtube mix' | |
1467 | if i > 0: | |
1468 | msg += ', retry #%d' % i | |
1469 | webpage = self._download_webpage(url, title, msg) | |
1470 | ids = orderedSet(re.findall(video_re, webpage)) | |
1471 | if ids: | |
1472 | break | |
1473 | url_results = self._ids_to_results(ids) | |
1474 | return self.playlist_result(url_results, playlist_title=title) | |
1475 | ||
1476 | ||
c5e8d7af | 1477 | class YoutubeChannelIE(InfoExtractor): |
0f818663 | 1478 | IE_DESC = u'YouTube.com channels' |
c5e8d7af | 1479 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
c5e8d7af | 1480 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' |
38c2e5b8 | 1481 | _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 |
1482 | IE_NAME = u'youtube:channel' |
1483 | ||
1484 | def extract_videos_from_page(self, page): | |
1485 | ids_in_page = [] | |
1486 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
1487 | if mobj.group(1) not in ids_in_page: | |
1488 | ids_in_page.append(mobj.group(1)) | |
1489 | return ids_in_page | |
1490 | ||
1491 | def _real_extract(self, url): | |
1492 | # Extract channel id | |
1493 | mobj = re.match(self._VALID_URL, url) | |
1494 | if mobj is None: | |
1495 | raise ExtractorError(u'Invalid URL: %s' % url) | |
1496 | ||
1497 | # Download channel page | |
1498 | channel_id = mobj.group(1) | |
1499 | video_ids = [] | |
b9643eed JMF |
1500 | url = 'https://www.youtube.com/channel/%s/videos' % channel_id |
1501 | channel_page = self._download_webpage(url, channel_id) | |
31812a9e PH |
1502 | autogenerated = re.search(r'''(?x) |
1503 | class="[^"]*?(?: | |
1504 | channel-header-autogenerated-label| | |
1505 | yt-channel-title-autogenerated | |
1506 | )[^"]*"''', channel_page) is not None | |
c5e8d7af | 1507 | |
b9643eed JMF |
1508 | if autogenerated: |
1509 | # The videos are contained in a single page | |
1510 | # the ajax pages can't be used, they are empty | |
1511 | video_ids = self.extract_videos_from_page(channel_page) | |
1512 | else: | |
1513 | # Download all channel pages using the json-based channel_ajax query | |
1514 | for pagenum in itertools.count(1): | |
1515 | url = self._MORE_PAGES_URL % (pagenum, channel_id) | |
81c2f20b PH |
1516 | page = self._download_json( |
1517 | url, channel_id, note=u'Downloading page #%s' % pagenum, | |
1518 | transform_source=uppercase_escape) | |
1519 | ||
b9643eed JMF |
1520 | ids_in_page = self.extract_videos_from_page(page['content_html']) |
1521 | video_ids.extend(ids_in_page) | |
1522 | ||
1523 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
1524 | break | |
c5e8d7af PH |
1525 | |
1526 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
1527 | ||
7012b23c PH |
1528 | url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id) |
1529 | for video_id in video_ids] | |
1530 | return self.playlist_result(url_entries, channel_id) | |
c5e8d7af PH |
1531 | |
1532 | ||
1533 | class YoutubeUserIE(InfoExtractor): | |
0f818663 | 1534 | IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' |
c9ae7b95 | 1535 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)' |
38c2e5b8 | 1536 | _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s' |
c5e8d7af | 1537 | _GDATA_PAGE_SIZE = 50 |
38c2e5b8 | 1538 | _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' |
c5e8d7af PH |
1539 | IE_NAME = u'youtube:user' |
1540 | ||
e3ea4790 | 1541 | @classmethod |
f4b05232 | 1542 | def suitable(cls, url): |
e3ea4790 JMF |
1543 | # Don't return True if the url can be extracted with other youtube |
1544 | # extractor, the regex would is too permissive and it would match. | |
1545 | other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) | |
1546 | if any(ie.suitable(url) for ie in other_ies): return False | |
f4b05232 JMF |
1547 | else: return super(YoutubeUserIE, cls).suitable(url) |
1548 | ||
c5e8d7af PH |
1549 | def _real_extract(self, url): |
1550 | # Extract username | |
1551 | mobj = re.match(self._VALID_URL, url) | |
1552 | if mobj is None: | |
1553 | raise ExtractorError(u'Invalid URL: %s' % url) | |
1554 | ||
1555 | username = mobj.group(1) | |
1556 | ||
1557 | # Download video ids using YouTube Data API. Result size per | |
1558 | # query is limited (currently to 50 videos) so we need to query | |
1559 | # page by page until there are no video ids - it means we got | |
1560 | # all of them. | |
1561 | ||
b7ab0590 | 1562 | def download_page(pagenum): |
c5e8d7af PH |
1563 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 |
1564 | ||
1565 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
b7ab0590 PH |
1566 | page = self._download_webpage( |
1567 | gdata_url, username, | |
1568 | u'Downloading video ids from %d to %d' % ( | |
1569 | start_index, start_index + self._GDATA_PAGE_SIZE)) | |
c5e8d7af | 1570 | |
fd9cf738 JMF |
1571 | try: |
1572 | response = json.loads(page) | |
1573 | except ValueError as err: | |
1574 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
71c82637 | 1575 | if 'entry' not in response['feed']: |
b7ab0590 | 1576 | return |
fd9cf738 | 1577 | |
c5e8d7af | 1578 | # Extract video identifiers |
e302f9ce PH |
1579 | entries = response['feed']['entry'] |
1580 | for entry in entries: | |
1581 | title = entry['title']['$t'] | |
1582 | video_id = entry['id']['$t'].split('/')[-1] | |
b7ab0590 | 1583 | yield { |
e302f9ce PH |
1584 | '_type': 'url', |
1585 | 'url': video_id, | |
1586 | 'ie_key': 'Youtube', | |
b11cec41 | 1587 | 'id': video_id, |
e302f9ce | 1588 | 'title': title, |
b7ab0590 PH |
1589 | } |
1590 | url_results = PagedList(download_page, self._GDATA_PAGE_SIZE) | |
c5e8d7af | 1591 | |
7012b23c PH |
1592 | return self.playlist_result(url_results, playlist_title=username) |
1593 | ||
b05654f0 PH |
1594 | |
1595 | class YoutubeSearchIE(SearchInfoExtractor): | |
0f818663 | 1596 | IE_DESC = u'YouTube.com searches' |
83d548ef | 1597 | _API_URL = u'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' |
b05654f0 PH |
1598 | _MAX_RESULTS = 1000 |
1599 | IE_NAME = u'youtube:search' | |
1600 | _SEARCH_KEY = 'ytsearch' | |
1601 | ||
b05654f0 PH |
1602 | def _get_n_results(self, query, n): |
1603 | """Get a specified number of results for a query""" | |
1604 | ||
1605 | video_ids = [] | |
1606 | pagenum = 0 | |
1607 | limit = n | |
83d548ef | 1608 | PAGE_SIZE = 50 |
b05654f0 | 1609 | |
83d548ef PH |
1610 | while (PAGE_SIZE * pagenum) < limit: |
1611 | result_url = self._API_URL % ( | |
1612 | compat_urllib_parse.quote_plus(query.encode('utf-8')), | |
1613 | (PAGE_SIZE * pagenum) + 1) | |
7cc3570e PH |
1614 | data_json = self._download_webpage( |
1615 | result_url, video_id=u'query "%s"' % query, | |
1616 | note=u'Downloading page %s' % (pagenum + 1), | |
1617 | errnote=u'Unable to download API page') | |
1618 | data = json.loads(data_json) | |
1619 | api_response = data['data'] | |
1620 | ||
1621 | if 'items' not in api_response: | |
07ad22b8 PH |
1622 | raise ExtractorError( |
1623 | u'[youtube] No video results', expected=True) | |
b05654f0 PH |
1624 | |
1625 | new_ids = list(video['id'] for video in api_response['items']) | |
1626 | video_ids += new_ids | |
1627 | ||
1628 | limit = min(n, api_response['totalItems']) | |
1629 | pagenum += 1 | |
1630 | ||
1631 | if len(video_ids) > n: | |
1632 | video_ids = video_ids[:n] | |
7012b23c PH |
1633 | videos = [self.url_result(video_id, 'Youtube', video_id=video_id) |
1634 | for video_id in video_ids] | |
b05654f0 | 1635 | return self.playlist_result(videos, query) |
75dff0ee | 1636 | |
c9ae7b95 | 1637 | |
a3dd9248 | 1638 | class YoutubeSearchDateIE(YoutubeSearchIE): |
cb7fb546 | 1639 | IE_NAME = YoutubeSearchIE.IE_NAME + ':date' |
a3dd9248 CM |
1640 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published' |
1641 | _SEARCH_KEY = 'ytsearchdate' | |
08fb86c4 | 1642 | IE_DESC = u'YouTube.com searches, newest videos first' |
75dff0ee | 1643 | |
c9ae7b95 PH |
1644 | |
1645 | class YoutubeSearchURLIE(InfoExtractor): | |
1646 | IE_DESC = u'YouTube.com search URLs' | |
1647 | IE_NAME = u'youtube:search_url' | |
1648 | _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)' | |
1649 | ||
1650 | def _real_extract(self, url): | |
1651 | mobj = re.match(self._VALID_URL, url) | |
1652 | query = compat_urllib_parse.unquote_plus(mobj.group('query')) | |
1653 | ||
1654 | webpage = self._download_webpage(url, query) | |
1655 | result_code = self._search_regex( | |
6feb2d5e | 1656 | r'(?s)<ol class="item-section"(.*?)</ol>', webpage, u'result HTML') |
c9ae7b95 PH |
1657 | |
1658 | part_codes = re.findall( | |
1659 | r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code) | |
1660 | entries = [] | |
1661 | for part_code in part_codes: | |
1662 | part_title = self._html_search_regex( | |
6feb2d5e | 1663 | [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False) |
c9ae7b95 PH |
1664 | part_url_snippet = self._html_search_regex( |
1665 | r'(?s)href="([^"]+)"', part_code, 'item URL') | |
1666 | part_url = compat_urlparse.urljoin( | |
1667 | 'https://www.youtube.com/', part_url_snippet) | |
1668 | entries.append({ | |
1669 | '_type': 'url', | |
1670 | 'url': part_url, | |
1671 | 'title': part_title, | |
1672 | }) | |
1673 | ||
1674 | return { | |
1675 | '_type': 'playlist', | |
1676 | 'entries': entries, | |
1677 | 'title': query, | |
1678 | } | |
1679 | ||
1680 | ||
75dff0ee | 1681 | class YoutubeShowIE(InfoExtractor): |
0f818663 | 1682 | IE_DESC = u'YouTube.com (multi-season) shows' |
75dff0ee JMF |
1683 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
1684 | IE_NAME = u'youtube:show' | |
1685 | ||
1686 | def _real_extract(self, url): | |
1687 | mobj = re.match(self._VALID_URL, url) | |
1688 | show_name = mobj.group(1) | |
1689 | webpage = self._download_webpage(url, show_name, u'Downloading show webpage') | |
1690 | # There's one playlist for each season of the show | |
1691 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
1692 | self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) | |
1693 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] | |
04cc9617 JMF |
1694 | |
1695 | ||
b2e8bc1b | 1696 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): |
d7ae0639 JMF |
1697 | """ |
1698 | Base class for extractors that fetch info from | |
1699 | http://www.youtube.com/feed_ajax | |
1700 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
1701 | """ | |
b2e8bc1b | 1702 | _LOGIN_REQUIRED = True |
43ba5456 JMF |
1703 | # use action_load_personal_feed instead of action_load_system_feed |
1704 | _PERSONAL_FEED = False | |
04cc9617 | 1705 | |
d7ae0639 JMF |
1706 | @property |
1707 | def _FEED_TEMPLATE(self): | |
43ba5456 JMF |
1708 | action = 'action_load_system_feed' |
1709 | if self._PERSONAL_FEED: | |
1710 | action = 'action_load_personal_feed' | |
38c2e5b8 | 1711 | return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) |
d7ae0639 JMF |
1712 | |
1713 | @property | |
1714 | def IE_NAME(self): | |
1715 | return u'youtube:%s' % self._FEED_NAME | |
04cc9617 | 1716 | |
81f0259b | 1717 | def _real_initialize(self): |
b2e8bc1b | 1718 | self._login() |
81f0259b | 1719 | |
04cc9617 JMF |
1720 | def _real_extract(self, url): |
1721 | feed_entries = [] | |
0e44d838 JMF |
1722 | paging = 0 |
1723 | for i in itertools.count(1): | |
f6177462 | 1724 | info = self._download_json(self._FEED_TEMPLATE % paging, |
d7ae0639 | 1725 | u'%s feed' % self._FEED_NAME, |
04cc9617 | 1726 | u'Downloading page %s' % i) |
f6177462 | 1727 | feed_html = info.get('feed_html') or info.get('content_html') |
43ba5456 | 1728 | m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) |
04cc9617 | 1729 | ids = orderedSet(m.group(1) for m in m_ids) |
7012b23c PH |
1730 | feed_entries.extend( |
1731 | self.url_result(video_id, 'Youtube', video_id=video_id) | |
1732 | for video_id in ids) | |
05ee2b6d JMF |
1733 | mobj = re.search( |
1734 | r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)', | |
1735 | feed_html) | |
1736 | if mobj is None: | |
04cc9617 | 1737 | break |
05ee2b6d | 1738 | paging = mobj.group('paging') |
d7ae0639 JMF |
1739 | return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) |
1740 | ||
1741 | class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): | |
e45d40b1 | 1742 | IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)' |
d7ae0639 JMF |
1743 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' |
1744 | _FEED_NAME = 'subscriptions' | |
1745 | _PLAYLIST_TITLE = u'Youtube Subscriptions' | |
1746 | ||
1747 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): | |
1748 | IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' | |
1749 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' | |
1750 | _FEED_NAME = 'recommended' | |
1751 | _PLAYLIST_TITLE = u'Youtube Recommended videos' | |
c626a3d9 | 1752 | |
43ba5456 JMF |
1753 | class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): |
1754 | IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' | |
1755 | _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' | |
1756 | _FEED_NAME = 'watch_later' | |
1757 | _PLAYLIST_TITLE = u'Youtube Watch Later' | |
43ba5456 | 1758 | _PERSONAL_FEED = True |
c626a3d9 | 1759 | |
f459d170 JMF |
1760 | class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): |
1761 | IE_DESC = u'Youtube watch history, "ythistory" keyword (requires authentication)' | |
1762 | _VALID_URL = u'https?://www\.youtube\.com/feed/history|:ythistory' | |
1763 | _FEED_NAME = 'history' | |
1764 | _PERSONAL_FEED = True | |
1765 | _PLAYLIST_TITLE = u'Youtube Watch History' | |
1766 | ||
c626a3d9 JMF |
1767 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): |
1768 | IE_NAME = u'youtube:favorites' | |
1769 | IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' | |
c7a7750d | 1770 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?' |
c626a3d9 JMF |
1771 | _LOGIN_REQUIRED = True |
1772 | ||
1773 | def _real_extract(self, url): | |
1774 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
1775 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id') | |
1776 | return self.url_result(playlist_id, 'YoutubePlaylist') | |
15870e90 PH |
1777 | |
1778 | ||
1779 | class YoutubeTruncatedURLIE(InfoExtractor): | |
1780 | IE_NAME = 'youtube:truncated_url' | |
1781 | IE_DESC = False # Do not list | |
975d35db | 1782 | _VALID_URL = r'''(?x) |
c4808c60 PH |
1783 | (?:https?://)?[^/]+/watch\?(?: |
1784 | feature=[a-z_]+| | |
1785 | annotation_id=annotation_[^&]+ | |
1786 | )?$| | |
975d35db PH |
1787 | (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$ |
1788 | ''' | |
15870e90 | 1789 | |
c4808c60 PH |
1790 | _TESTS = [{ |
1791 | 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041', | |
1792 | 'only_matching': True, | |
dc2fc736 PH |
1793 | }, { |
1794 | 'url': 'http://www.youtube.com/watch?', | |
1795 | 'only_matching': True, | |
c4808c60 PH |
1796 | }] |
1797 | ||
15870e90 PH |
1798 | def _real_extract(self, url): |
1799 | raise ExtractorError( | |
1800 | u'Did you forget to quote the URL? Remember that & is a meta ' | |
1801 | u'character in most shells, so you want to put the URL in quotes, ' | |
1802 | u'like youtube-dl ' | |
b4622a32 PH |
1803 | u'"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' |
1804 | u' or simply youtube-dl BaW_jenozKc .', | |
15870e90 | 1805 | expected=True) |