]>
Commit | Line | Data |
---|---|---|
c5e8d7af | 1 | # coding: utf-8 |
c5e8d7af PH |
2 | |
3 | import json | |
4 | import netrc | |
5 | import re | |
6 | import socket | |
04cc9617 | 7 | import itertools |
055e6f36 | 8 | import xml.etree.ElementTree |
c5e8d7af | 9 | |
b05654f0 | 10 | from .common import InfoExtractor, SearchInfoExtractor |
54d39d8b | 11 | from .subtitles import SubtitlesInfoExtractor |
c5e8d7af PH |
12 | from ..utils import ( |
13 | compat_http_client, | |
14 | compat_parse_qs, | |
15 | compat_urllib_error, | |
16 | compat_urllib_parse, | |
17 | compat_urllib_request, | |
18 | compat_str, | |
19 | ||
20 | clean_html, | |
21 | get_element_by_id, | |
22 | ExtractorError, | |
23 | unescapeHTML, | |
24 | unified_strdate, | |
04cc9617 | 25 | orderedSet, |
c5e8d7af PH |
26 | ) |
27 | ||
de7f3446 | 28 | class YoutubeBaseInfoExtractor(InfoExtractor): |
b2e8bc1b JMF |
29 | """Provide base functions for Youtube extractors""" |
30 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
31 | _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' | |
32 | _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' | |
33 | _NETRC_MACHINE = 'youtube' | |
34 | # If True it will raise an error if no login info is provided | |
35 | _LOGIN_REQUIRED = False | |
36 | ||
37 | def report_lang(self): | |
38 | """Report attempt to set language.""" | |
39 | self.to_screen(u'Setting language') | |
40 | ||
41 | def _set_language(self): | |
42 | request = compat_urllib_request.Request(self._LANG_URL) | |
43 | try: | |
44 | self.report_lang() | |
45 | compat_urllib_request.urlopen(request).read() | |
46 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
47 | self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) | |
48 | return False | |
49 | return True | |
50 | ||
51 | def _login(self): | |
52 | (username, password) = self._get_login_info() | |
53 | # No authentication to be performed | |
54 | if username is None: | |
55 | if self._LOGIN_REQUIRED: | |
56 | raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) | |
57 | return False | |
58 | ||
59 | request = compat_urllib_request.Request(self._LOGIN_URL) | |
60 | try: | |
61 | login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
62 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
63 | self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) | |
64 | return False | |
65 | ||
66 | galx = None | |
67 | dsh = None | |
68 | match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page) | |
69 | if match: | |
70 | galx = match.group(1) | |
71 | match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page) | |
72 | if match: | |
73 | dsh = match.group(1) | |
c5e8d7af | 74 | |
b2e8bc1b JMF |
75 | # Log in |
76 | login_form_strs = { | |
77 | u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
78 | u'Email': username, | |
79 | u'GALX': galx, | |
80 | u'Passwd': password, | |
81 | u'PersistentCookie': u'yes', | |
82 | u'_utf8': u'霱', | |
83 | u'bgresponse': u'js_disabled', | |
84 | u'checkConnection': u'', | |
85 | u'checkedDomains': u'youtube', | |
86 | u'dnConn': u'', | |
87 | u'dsh': dsh, | |
88 | u'pstMsg': u'0', | |
89 | u'rmShown': u'1', | |
90 | u'secTok': u'', | |
91 | u'signIn': u'Sign in', | |
92 | u'timeStmp': u'', | |
93 | u'service': u'youtube', | |
94 | u'uilel': u'3', | |
95 | u'hl': u'en_US', | |
96 | } | |
97 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
98 | # chokes on unicode | |
99 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) | |
100 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
101 | request = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
102 | try: | |
103 | self.report_login() | |
104 | login_results = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
105 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
106 | self._downloader.report_warning(u'unable to log in: bad username or password') | |
107 | return False | |
108 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
109 | self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) | |
110 | return False | |
111 | return True | |
112 | ||
113 | def _confirm_age(self): | |
114 | age_form = { | |
115 | 'next_url': '/', | |
116 | 'action_confirm': 'Confirm', | |
117 | } | |
118 | request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) | |
119 | try: | |
120 | self.report_age_confirmation() | |
121 | compat_urllib_request.urlopen(request).read().decode('utf-8') | |
122 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
123 | raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) | |
124 | return True | |
125 | ||
126 | def _real_initialize(self): | |
127 | if self._downloader is None: | |
128 | return | |
129 | if not self._set_language(): | |
130 | return | |
131 | if not self._login(): | |
132 | return | |
133 | self._confirm_age() | |
c5e8d7af | 134 | |
8377574c | 135 | |
de7f3446 | 136 | class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): |
0f818663 | 137 | IE_DESC = u'YouTube.com' |
c5e8d7af PH |
138 | _VALID_URL = r"""^ |
139 | ( | |
140 | (?:https?://)? # http(s):// (optional) | |
f4b05232 | 141 | (?:(?:(?:(?:\w+\.)?youtube(?:-nocookie)?\.com/| |
c5e8d7af PH |
142 | tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains |
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 | |
d741e55a | 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 | |
154 | ) | |
c5e8d7af | 155 | )? # all until now is optional -> you can pass the naked ID |
8963d9c2 | 156 | ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID |
c5e8d7af PH |
157 | (?(1).+)? # if we found the ID, everything can follow |
158 | $""" | |
c5e8d7af | 159 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' |
c5e8d7af | 160 | # Listed in order of quality |
bdc6b3fc | 161 | _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '36', '17', '13', |
96fb5605 | 162 | # Apple HTTP Live Streaming |
bdc6b3fc | 163 | '96', '95', '94', '93', '92', '132', '151', |
939fbd26 JMF |
164 | # 3D |
165 | '85', '84', '102', '83', '101', '82', '100', | |
166 | # Dash video | |
167 | '138', '137', '248', '136', '247', '135', '246', | |
168 | '245', '244', '134', '243', '133', '242', '160', | |
169 | # Dash audio | |
170 | '141', '172', '140', '171', '139', | |
1d043b93 | 171 | ] |
bdc6b3fc | 172 | _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '36', '17', '13', |
96fb5605 | 173 | # Apple HTTP Live Streaming |
bdc6b3fc AZ |
174 | '96', '95', '94', '93', '92', '132', '151', |
175 | # 3D | |
86fe61c8 | 176 | '85', '102', '84', '101', '83', '100', '82', |
939fbd26 JMF |
177 | # Dash video |
178 | '138', '248', '137', '247', '136', '246', '245', | |
179 | '244', '135', '243', '134', '242', '133', '160', | |
180 | # Dash audio | |
181 | '172', '141', '171', '140', '139', | |
1d043b93 | 182 | ] |
bdc6b3fc AZ |
183 | _video_formats_map = { |
184 | 'flv': ['35', '34', '6', '5'], | |
185 | '3gp': ['36', '17', '13'], | |
186 | 'mp4': ['38', '37', '22', '18'], | |
187 | 'webm': ['46', '45', '44', '43'], | |
188 | } | |
c5e8d7af PH |
189 | _video_extensions = { |
190 | '13': '3gp', | |
bdc6b3fc | 191 | '17': '3gp', |
c5e8d7af PH |
192 | '18': 'mp4', |
193 | '22': 'mp4', | |
bdc6b3fc | 194 | '36': '3gp', |
c5e8d7af | 195 | '37': 'mp4', |
d69cf69a | 196 | '38': 'mp4', |
c5e8d7af PH |
197 | '43': 'webm', |
198 | '44': 'webm', | |
199 | '45': 'webm', | |
200 | '46': 'webm', | |
1d043b93 | 201 | |
86fe61c8 AZ |
202 | # 3d videos |
203 | '82': 'mp4', | |
204 | '83': 'mp4', | |
205 | '84': 'mp4', | |
206 | '85': 'mp4', | |
207 | '100': 'webm', | |
208 | '101': 'webm', | |
209 | '102': 'webm', | |
836a086c | 210 | |
96fb5605 | 211 | # Apple HTTP Live Streaming |
1d043b93 JMF |
212 | '92': 'mp4', |
213 | '93': 'mp4', | |
214 | '94': 'mp4', | |
215 | '95': 'mp4', | |
216 | '96': 'mp4', | |
217 | '132': 'mp4', | |
218 | '151': 'mp4', | |
836a086c AZ |
219 | |
220 | # Dash mp4 | |
221 | '133': 'mp4', | |
222 | '134': 'mp4', | |
223 | '135': 'mp4', | |
224 | '136': 'mp4', | |
225 | '137': 'mp4', | |
226 | '138': 'mp4', | |
227 | '139': 'mp4', | |
228 | '140': 'mp4', | |
229 | '141': 'mp4', | |
230 | '160': 'mp4', | |
231 | ||
232 | # Dash webm | |
233 | '171': 'webm', | |
234 | '172': 'webm', | |
235 | '242': 'webm', | |
236 | '243': 'webm', | |
237 | '244': 'webm', | |
238 | '245': 'webm', | |
239 | '246': 'webm', | |
240 | '247': 'webm', | |
241 | '248': 'webm', | |
c5e8d7af PH |
242 | } |
243 | _video_dimensions = { | |
244 | '5': '240x400', | |
245 | '6': '???', | |
246 | '13': '???', | |
247 | '17': '144x176', | |
248 | '18': '360x640', | |
249 | '22': '720x1280', | |
250 | '34': '360x640', | |
251 | '35': '480x854', | |
bdc6b3fc | 252 | '36': '240x320', |
c5e8d7af PH |
253 | '37': '1080x1920', |
254 | '38': '3072x4096', | |
255 | '43': '360x640', | |
256 | '44': '480x854', | |
257 | '45': '720x1280', | |
258 | '46': '1080x1920', | |
86fe61c8 AZ |
259 | '82': '360p', |
260 | '83': '480p', | |
261 | '84': '720p', | |
262 | '85': '1080p', | |
1d043b93 JMF |
263 | '92': '240p', |
264 | '93': '360p', | |
265 | '94': '480p', | |
266 | '95': '720p', | |
267 | '96': '1080p', | |
86fe61c8 AZ |
268 | '100': '360p', |
269 | '101': '480p', | |
836a086c | 270 | '102': '720p', |
1d043b93 JMF |
271 | '132': '240p', |
272 | '151': '72p', | |
836a086c AZ |
273 | '133': '240p', |
274 | '134': '360p', | |
275 | '135': '480p', | |
276 | '136': '720p', | |
277 | '137': '1080p', | |
278 | '138': '>1080p', | |
279 | '139': '48k', | |
280 | '140': '128k', | |
281 | '141': '256k', | |
282 | '160': '192p', | |
283 | '171': '128k', | |
284 | '172': '256k', | |
285 | '242': '240p', | |
286 | '243': '360p', | |
287 | '244': '480p', | |
288 | '245': '480p', | |
289 | '246': '480p', | |
290 | '247': '720p', | |
291 | '248': '1080p', | |
c5e8d7af | 292 | } |
836a086c AZ |
293 | _special_itags = { |
294 | '82': '3D', | |
295 | '83': '3D', | |
296 | '84': '3D', | |
297 | '85': '3D', | |
298 | '100': '3D', | |
299 | '101': '3D', | |
300 | '102': '3D', | |
301 | '133': 'DASH Video', | |
302 | '134': 'DASH Video', | |
303 | '135': 'DASH Video', | |
304 | '136': 'DASH Video', | |
305 | '137': 'DASH Video', | |
306 | '138': 'DASH Video', | |
307 | '139': 'DASH Audio', | |
308 | '140': 'DASH Audio', | |
309 | '141': 'DASH Audio', | |
310 | '160': 'DASH Video', | |
311 | '171': 'DASH Audio', | |
312 | '172': 'DASH Audio', | |
313 | '242': 'DASH Video', | |
314 | '243': 'DASH Video', | |
315 | '244': 'DASH Video', | |
316 | '245': 'DASH Video', | |
317 | '246': 'DASH Video', | |
318 | '247': 'DASH Video', | |
319 | '248': 'DASH Video', | |
c5e8d7af | 320 | } |
836a086c | 321 | |
c5e8d7af | 322 | IE_NAME = u'youtube' |
2eb88d95 PH |
323 | _TESTS = [ |
324 | { | |
0e853ca4 PH |
325 | u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", |
326 | u"file": u"BaW_jenozKc.mp4", | |
327 | u"info_dict": { | |
328 | u"title": u"youtube-dl test video \"'/\\ä↭𝕐", | |
329 | u"uploader": u"Philipp Hagemeister", | |
330 | u"uploader_id": u"phihag", | |
331 | u"upload_date": u"20121002", | |
332 | u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ." | |
2eb88d95 | 333 | } |
0e853ca4 PH |
334 | }, |
335 | { | |
336 | u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U", | |
337 | u"file": u"1ltcDfZMA3U.flv", | |
338 | u"note": u"Test VEVO video (#897)", | |
339 | u"info_dict": { | |
340 | u"upload_date": u"20070518", | |
341 | u"title": u"Maps - It Will Find You", | |
342 | u"description": u"Music video by Maps performing It Will Find You.", | |
343 | u"uploader": u"MuteUSA", | |
344 | u"uploader_id": u"MuteUSA" | |
2eb88d95 | 345 | } |
0e853ca4 PH |
346 | }, |
347 | { | |
348 | u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", | |
349 | u"file": u"UxxajLWwzqY.mp4", | |
350 | u"note": u"Test generic use_cipher_signature video (#897)", | |
351 | u"info_dict": { | |
352 | u"upload_date": u"20120506", | |
353 | u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", | |
c7bf7366 | 354 | u"description": u"md5:3e2666e0a55044490499ea45fe9037b7", |
45ed795c | 355 | u"uploader": u"Icona Pop", |
0e853ca4 | 356 | u"uploader_id": u"IconaPop" |
2eb88d95 | 357 | } |
c108eb73 JMF |
358 | }, |
359 | { | |
360 | u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", | |
361 | u"file": u"07FYdnEawAQ.mp4", | |
362 | u"note": u"Test VEVO video with age protection (#956)", | |
363 | u"info_dict": { | |
364 | u"upload_date": u"20130703", | |
365 | u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", | |
366 | u"description": u"md5:64249768eec3bc4276236606ea996373", | |
367 | u"uploader": u"justintimberlakeVEVO", | |
368 | u"uploader_id": u"justintimberlakeVEVO" | |
369 | } | |
370 | }, | |
1d043b93 JMF |
371 | { |
372 | u'url': u'https://www.youtube.com/watch?v=TGi3HqYrWHE', | |
373 | u'file': u'TGi3HqYrWHE.mp4', | |
374 | u'note': u'm3u8 video', | |
375 | u'info_dict': { | |
376 | u'title': u'Triathlon - Men - London 2012 Olympic Games', | |
377 | u'description': u'- Men - TR02 - Triathlon - 07 August 2012 - London 2012 Olympic Games', | |
378 | u'uploader': u'olympic', | |
379 | u'upload_date': u'20120807', | |
380 | u'uploader_id': u'olympic', | |
381 | }, | |
382 | u'params': { | |
383 | u'skip_download': True, | |
384 | }, | |
385 | }, | |
2eb88d95 PH |
386 | ] |
387 | ||
c5e8d7af PH |
388 | |
389 | @classmethod | |
390 | def suitable(cls, url): | |
391 | """Receives a URL and returns True if suitable for this IE.""" | |
e3ea4790 | 392 | if YoutubePlaylistIE.suitable(url): return False |
c5e8d7af PH |
393 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None |
394 | ||
c5e8d7af PH |
395 | def report_video_webpage_download(self, video_id): |
396 | """Report attempt to download video webpage.""" | |
397 | self.to_screen(u'%s: Downloading video webpage' % video_id) | |
398 | ||
399 | def report_video_info_webpage_download(self, video_id): | |
400 | """Report attempt to download video info webpage.""" | |
401 | self.to_screen(u'%s: Downloading video info webpage' % video_id) | |
402 | ||
c5e8d7af PH |
403 | def report_information_extraction(self, video_id): |
404 | """Report attempt to extract video information.""" | |
405 | self.to_screen(u'%s: Extracting video information' % video_id) | |
406 | ||
407 | def report_unavailable_format(self, video_id, format): | |
408 | """Report extracted video URL.""" | |
409 | self.to_screen(u'%s: Format %s not available' % (video_id, format)) | |
410 | ||
411 | def report_rtmp_download(self): | |
412 | """Indicate the download will use the RTMP protocol.""" | |
413 | self.to_screen(u'RTMP download detected') | |
414 | ||
98bcd283 | 415 | def _decrypt_signature(self, s): |
257a2501 | 416 | """Turn the encrypted s field into a working signature""" |
6b37f0be | 417 | |
444b1165 JMF |
418 | if len(s) == 92: |
419 | return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83] | |
420 | elif len(s) == 90: | |
421 | return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81] | |
8a9d86a2 | 422 | elif len(s) == 89: |
423 | return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1] | |
444b1165 | 424 | elif len(s) == 88: |
3e223834 | 425 | return s[7:28] + s[87] + s[29:45] + s[55] + s[46:55] + s[2] + s[56:87] + s[28] |
be547e1d | 426 | elif len(s) == 87: |
3a725669 | 427 | return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:] |
be547e1d | 428 | elif len(s) == 86: |
1cf911bc | 429 | return s[5:34] + s[0] + s[35:38] + s[3] + s[39:45] + s[38] + s[46:53] + s[73] + s[54:73] + s[85] + s[74:85] + s[53] |
be547e1d | 430 | elif len(s) == 85: |
aa8f2641 | 431 | return s[40] + s[82:43:-1] + s[22] + s[42:40:-1] + s[83] + s[39:22:-1] + s[0] + s[21:2:-1] |
be547e1d | 432 | elif len(s) == 84: |
23b00bc0 | 433 | return s[81:36:-1] + s[0] + s[35:2:-1] |
be547e1d | 434 | elif len(s) == 83: |
e1842025 | 435 | return s[81:64:-1] + s[82] + s[63:52:-1] + s[45] + s[51:45:-1] + s[1] + s[44:1:-1] + s[0] |
be547e1d | 436 | elif len(s) == 82: |
3669cdba | 437 | return s[1:19] + s[0] + s[20:68] + s[19] + s[69:82] |
be547e1d | 438 | elif len(s) == 81: |
aedd6bb9 | 439 | 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 |
440 | elif len(s) == 80: |
441 | return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80] | |
5c468ca8 JMF |
442 | elif len(s) == 79: |
443 | 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 |
444 | |
445 | else: | |
446 | raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) | |
c5e8d7af | 447 | |
75952c6e JMF |
448 | def _decrypt_signature_age_gate(self, s): |
449 | # The videos with age protection use another player, so the algorithms | |
450 | # can be different. | |
451 | if len(s) == 86: | |
452 | return s[2:63] + s[82] + s[64:82] + s[63] | |
453 | else: | |
454 | # Fallback to the other algortihms | |
b072a9de | 455 | return self._decrypt_signature(s) |
c5e8d7af | 456 | |
de7f3446 | 457 | def _get_available_subtitles(self, video_id): |
de7f3446 | 458 | try: |
7fad1c63 JMF |
459 | sub_list = self._download_webpage( |
460 | 'http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, | |
461 | video_id, note=False) | |
462 | except ExtractorError as err: | |
de7f3446 JMF |
463 | self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) |
464 | return {} | |
465 | lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
466 | ||
467 | sub_lang_list = {} | |
468 | for l in lang_list: | |
469 | lang = l[1] | |
470 | params = compat_urllib_parse.urlencode({ | |
471 | 'lang': lang, | |
472 | 'v': video_id, | |
473 | 'fmt': self._downloader.params.get('subtitlesformat'), | |
474 | }) | |
475 | url = u'http://www.youtube.com/api/timedtext?' + params | |
476 | sub_lang_list[lang] = url | |
477 | if not sub_lang_list: | |
478 | self._downloader.report_warning(u'video doesn\'t have subtitles') | |
479 | return {} | |
480 | return sub_lang_list | |
481 | ||
055e6f36 | 482 | def _get_available_automatic_caption(self, video_id, webpage): |
de7f3446 JMF |
483 | """We need the webpage for getting the captions url, pass it as an |
484 | argument to speed up the process.""" | |
de7f3446 JMF |
485 | sub_format = self._downloader.params.get('subtitlesformat') |
486 | self.to_screen(u'%s: Looking for automatic captions' % video_id) | |
487 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
055e6f36 | 488 | err_msg = u'Couldn\'t find automatic captions for %s' % video_id |
de7f3446 JMF |
489 | if mobj is None: |
490 | self._downloader.report_warning(err_msg) | |
491 | return {} | |
492 | player_config = json.loads(mobj.group(1)) | |
493 | try: | |
494 | args = player_config[u'args'] | |
495 | caption_url = args[u'ttsurl'] | |
496 | timestamp = args[u'timestamp'] | |
055e6f36 JMF |
497 | # We get the available subtitles |
498 | list_params = compat_urllib_parse.urlencode({ | |
499 | 'type': 'list', | |
500 | 'tlangs': 1, | |
501 | 'asrs': 1, | |
de7f3446 | 502 | }) |
055e6f36 JMF |
503 | list_url = caption_url + '&' + list_params |
504 | list_page = self._download_webpage(list_url, video_id) | |
505 | caption_list = xml.etree.ElementTree.fromstring(list_page.encode('utf-8')) | |
506 | original_lang = caption_list.find('track').attrib['lang_code'] | |
507 | ||
508 | sub_lang_list = {} | |
509 | for lang_node in caption_list.findall('target'): | |
510 | sub_lang = lang_node.attrib['lang_code'] | |
511 | params = compat_urllib_parse.urlencode({ | |
512 | 'lang': original_lang, | |
513 | 'tlang': sub_lang, | |
514 | 'fmt': sub_format, | |
515 | 'ts': timestamp, | |
516 | 'kind': 'asr', | |
517 | }) | |
518 | sub_lang_list[sub_lang] = caption_url + '&' + params | |
519 | return sub_lang_list | |
de7f3446 JMF |
520 | # An extractor error can be raise by the download process if there are |
521 | # no automatic captions but there are subtitles | |
522 | except (KeyError, ExtractorError): | |
523 | self._downloader.report_warning(err_msg) | |
524 | return {} | |
525 | ||
c5e8d7af PH |
526 | def _print_formats(self, formats): |
527 | print('Available formats:') | |
528 | for x in formats: | |
03cc7c20 JMF |
529 | print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'), |
530 | self._video_dimensions.get(x, '???'), | |
836a086c | 531 | ' ('+self._special_itags[x]+')' if x in self._special_itags else '')) |
c5e8d7af PH |
532 | |
533 | def _extract_id(self, url): | |
534 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
535 | if mobj is None: | |
536 | raise ExtractorError(u'Invalid URL: %s' % url) | |
537 | video_id = mobj.group(2) | |
538 | return video_id | |
539 | ||
1d043b93 JMF |
540 | def _get_video_url_list(self, url_map): |
541 | """ | |
542 | Transform a dictionary in the format {itag:url} to a list of (itag, url) | |
543 | with the requested formats. | |
544 | """ | |
545 | req_format = self._downloader.params.get('format', None) | |
546 | format_limit = self._downloader.params.get('format_limit', None) | |
547 | available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats | |
548 | if format_limit is not None and format_limit in available_formats: | |
549 | format_list = available_formats[available_formats.index(format_limit):] | |
550 | else: | |
551 | format_list = available_formats | |
552 | existing_formats = [x for x in format_list if x in url_map] | |
553 | if len(existing_formats) == 0: | |
554 | raise ExtractorError(u'no known formats available for video') | |
555 | if self._downloader.params.get('listformats', None): | |
556 | self._print_formats(existing_formats) | |
557 | return | |
558 | if req_format is None or req_format == 'best': | |
559 | video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality | |
560 | elif req_format == 'worst': | |
561 | video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality | |
562 | elif req_format in ('-1', 'all'): | |
563 | video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats | |
564 | else: | |
565 | # Specific formats. We pick the first in a slash-delimeted sequence. | |
bdc6b3fc AZ |
566 | # Format can be specified as itag or 'mp4' or 'flv' etc. We pick the highest quality |
567 | # available in the specified format. For example, | |
568 | # if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. | |
569 | # if '1/mp4/3/4' is requested and '1' and '5' (is a mp4) are available, we pick '1'. | |
570 | # if '1/mp4/3/4' is requested and '4' and '5' (is a mp4) are available, we pick '5'. | |
1d043b93 JMF |
571 | req_formats = req_format.split('/') |
572 | video_url_list = None | |
573 | for rf in req_formats: | |
574 | if rf in url_map: | |
575 | video_url_list = [(rf, url_map[rf])] | |
576 | break | |
bdc6b3fc AZ |
577 | if rf in self._video_formats_map: |
578 | for srf in self._video_formats_map[rf]: | |
579 | if srf in url_map: | |
580 | video_url_list = [(srf, url_map[srf])] | |
581 | break | |
582 | else: | |
583 | continue | |
584 | break | |
1d043b93 JMF |
585 | if video_url_list is None: |
586 | raise ExtractorError(u'requested format not available') | |
587 | return video_url_list | |
588 | ||
589 | def _extract_from_m3u8(self, manifest_url, video_id): | |
590 | url_map = {} | |
591 | def _get_urls(_manifest): | |
592 | lines = _manifest.split('\n') | |
593 | urls = filter(lambda l: l and not l.startswith('#'), | |
594 | lines) | |
595 | return urls | |
596 | manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') | |
597 | formats_urls = _get_urls(manifest) | |
598 | for format_url in formats_urls: | |
890f62e8 | 599 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') |
1d043b93 JMF |
600 | url_map[itag] = format_url |
601 | return url_map | |
602 | ||
c5e8d7af | 603 | def _real_extract(self, url): |
d7f44b5b PH |
604 | if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url): |
605 | self._downloader.report_warning(u'Did you forget to quote the URL? Remember that & is a meta-character in most shells, so you want to put the URL in quotes, like youtube-dl \'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\' (or simply youtube-dl BaW_jenozKc ).') | |
606 | ||
c5e8d7af PH |
607 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter |
608 | mobj = re.search(self._NEXT_URL_RE, url) | |
609 | if mobj: | |
610 | url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
611 | video_id = self._extract_id(url) | |
612 | ||
613 | # Get video webpage | |
614 | self.report_video_webpage_download(video_id) | |
615 | url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id | |
616 | request = compat_urllib_request.Request(url) | |
617 | try: | |
618 | video_webpage_bytes = compat_urllib_request.urlopen(request).read() | |
619 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
620 | raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err)) | |
621 | ||
622 | video_webpage = video_webpage_bytes.decode('utf-8', 'ignore') | |
623 | ||
624 | # Attempt to extract SWF player URL | |
625 | mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) | |
626 | if mobj is not None: | |
627 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
628 | else: | |
629 | player_url = None | |
630 | ||
631 | # Get video info | |
632 | self.report_video_info_webpage_download(video_id) | |
c108eb73 JMF |
633 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
634 | self.report_age_confirmation() | |
635 | age_gate = True | |
636 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
637 | # this can be viewed without login into Youtube | |
638 | data = compat_urllib_parse.urlencode({'video_id': video_id, | |
639 | 'el': 'embedded', | |
640 | 'gl': 'US', | |
641 | 'hl': 'en', | |
642 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
643 | 'asv': 3, | |
644 | 'sts':'1588', | |
645 | }) | |
646 | video_info_url = 'https://www.youtube.com/get_video_info?' + data | |
c5e8d7af PH |
647 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
648 | note=False, | |
649 | errnote='unable to download video info webpage') | |
650 | video_info = compat_parse_qs(video_info_webpage) | |
c108eb73 JMF |
651 | else: |
652 | age_gate = False | |
653 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
654 | video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
655 | % (video_id, el_type)) | |
656 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
657 | note=False, | |
658 | errnote='unable to download video info webpage') | |
659 | video_info = compat_parse_qs(video_info_webpage) | |
660 | if 'token' in video_info: | |
661 | break | |
c5e8d7af PH |
662 | if 'token' not in video_info: |
663 | if 'reason' in video_info: | |
9a82b238 | 664 | raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True) |
c5e8d7af PH |
665 | else: |
666 | raise ExtractorError(u'"token" parameter not in video info for unknown reason') | |
667 | ||
668 | # Check for "rental" videos | |
669 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
670 | raise ExtractorError(u'"rental" videos not supported') | |
671 | ||
672 | # Start extracting information | |
673 | self.report_information_extraction(video_id) | |
674 | ||
675 | # uploader | |
676 | if 'author' not in video_info: | |
677 | raise ExtractorError(u'Unable to extract uploader name') | |
678 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
679 | ||
680 | # uploader_id | |
681 | video_uploader_id = None | |
682 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
683 | if mobj is not None: | |
684 | video_uploader_id = mobj.group(1) | |
685 | else: | |
686 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
687 | ||
688 | # title | |
689 | if 'title' not in video_info: | |
690 | raise ExtractorError(u'Unable to extract video title') | |
691 | video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) | |
692 | ||
693 | # thumbnail image | |
7763b04e JMF |
694 | # We try first to get a high quality image: |
695 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
696 | video_webpage, re.DOTALL) | |
697 | if m_thumb is not None: | |
698 | video_thumbnail = m_thumb.group(1) | |
699 | elif 'thumbnail_url' not in video_info: | |
c5e8d7af PH |
700 | self._downloader.report_warning(u'unable to extract video thumbnail') |
701 | video_thumbnail = '' | |
702 | else: # don't panic if we can't find it | |
703 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
704 | ||
705 | # upload date | |
706 | upload_date = None | |
707 | mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL) | |
708 | if mobj is not None: | |
709 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
710 | upload_date = unified_strdate(upload_date) | |
711 | ||
712 | # description | |
713 | video_description = get_element_by_id("eow-description", video_webpage) | |
714 | if video_description: | |
715 | video_description = clean_html(video_description) | |
716 | else: | |
717 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
718 | if fd_mobj: | |
719 | video_description = unescapeHTML(fd_mobj.group(1)) | |
720 | else: | |
721 | video_description = u'' | |
722 | ||
723 | # subtitles | |
d82134c3 | 724 | video_subtitles = self.extract_subtitles(video_id, video_webpage) |
c5e8d7af | 725 | |
c5e8d7af | 726 | if self._downloader.params.get('listsubtitles', False): |
93d3a642 | 727 | self._list_available_subtitles(video_id) |
c5e8d7af PH |
728 | return |
729 | ||
730 | if 'length_seconds' not in video_info: | |
731 | self._downloader.report_warning(u'unable to extract video duration') | |
732 | video_duration = '' | |
733 | else: | |
734 | video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]) | |
735 | ||
c5e8d7af | 736 | # Decide which formats to download |
c5e8d7af PH |
737 | |
738 | try: | |
739 | mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) | |
50be92c1 PH |
740 | if not mobj: |
741 | raise ValueError('Could not find vevo ID') | |
c5e8d7af PH |
742 | info = json.loads(mobj.group(1)) |
743 | args = info['args'] | |
7ce7e394 JMF |
744 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
745 | # this signatures are encrypted | |
746 | m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map']) | |
747 | if m_s is not None: | |
748 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
c5e8d7af | 749 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
cde846b3 | 750 | m_s = re.search(r'[&,]s=', args.get('adaptive_fmts', u'')) |
b7a68384 | 751 | if m_s is not None: |
37b6d5f6 AZ |
752 | if 'url_encoded_fmt_stream_map' in video_info: |
753 | video_info['url_encoded_fmt_stream_map'][0] += ',' + args['adaptive_fmts'] | |
754 | else: | |
755 | video_info['url_encoded_fmt_stream_map'] = [args['adaptive_fmts']] | |
211fbc13 | 756 | elif 'adaptive_fmts' in video_info: |
37b6d5f6 AZ |
757 | if 'url_encoded_fmt_stream_map' in video_info: |
758 | video_info['url_encoded_fmt_stream_map'][0] += ',' + video_info['adaptive_fmts'][0] | |
759 | else: | |
760 | video_info['url_encoded_fmt_stream_map'] = video_info['adaptive_fmts'] | |
c5e8d7af PH |
761 | except ValueError: |
762 | pass | |
763 | ||
764 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
765 | self.report_rtmp_download() | |
766 | video_url_list = [(None, video_info['conn'][0])] | |
767 | elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1: | |
a7055eb9 JMF |
768 | if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]: |
769 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) | |
c5e8d7af PH |
770 | url_map = {} |
771 | for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): | |
772 | url_data = compat_parse_qs(url_data_str) | |
773 | if 'itag' in url_data and 'url' in url_data: | |
774 | url = url_data['url'][0] | |
775 | if 'sig' in url_data: | |
776 | url += '&signature=' + url_data['sig'][0] | |
777 | elif 's' in url_data: | |
769fda3c FV |
778 | if self._downloader.params.get('verbose'): |
779 | s = url_data['s'][0] | |
c108eb73 JMF |
780 | if age_gate: |
781 | player_version = self._search_regex(r'ad3-(.+?)\.swf', | |
ed27d356 JMF |
782 | video_info['ad3_module'][0] if 'ad3_module' in video_info else 'NOT FOUND', |
783 | 'flash player', fatal=False) | |
c108eb73 JMF |
784 | player = 'flash player %s' % player_version |
785 | else: | |
786 | player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage, | |
787 | 'html5 player', fatal=False) | |
5a76c651 JMF |
788 | parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.')) |
789 | self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % | |
790 | (len(s), parts_sizes, url_data['itag'][0], player)) | |
75952c6e JMF |
791 | encrypted_sig = url_data['s'][0] |
792 | if age_gate: | |
793 | signature = self._decrypt_signature_age_gate(encrypted_sig) | |
794 | else: | |
795 | signature = self._decrypt_signature(encrypted_sig) | |
c5e8d7af PH |
796 | url += '&signature=' + signature |
797 | if 'ratebypass' not in url: | |
798 | url += '&ratebypass=yes' | |
799 | url_map[url_data['itag'][0]] = url | |
1d043b93 JMF |
800 | video_url_list = self._get_video_url_list(url_map) |
801 | if not video_url_list: | |
c5e8d7af | 802 | return |
1d043b93 JMF |
803 | elif video_info.get('hlsvp'): |
804 | manifest_url = video_info['hlsvp'][0] | |
805 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
806 | video_url_list = self._get_video_url_list(url_map) | |
807 | if not video_url_list: | |
808 | return | |
809 | ||
c5e8d7af PH |
810 | else: |
811 | raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info') | |
812 | ||
813 | results = [] | |
814 | for format_param, video_real_url in video_url_list: | |
815 | # Extension | |
816 | video_extension = self._video_extensions.get(format_param, 'flv') | |
817 | ||
03cc7c20 JMF |
818 | video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension, |
819 | self._video_dimensions.get(format_param, '???'), | |
836a086c | 820 | ' ('+self._special_itags[format_param]+')' if format_param in self._special_itags else '') |
c5e8d7af PH |
821 | |
822 | results.append({ | |
823 | 'id': video_id, | |
824 | 'url': video_real_url, | |
825 | 'uploader': video_uploader, | |
826 | 'uploader_id': video_uploader_id, | |
827 | 'upload_date': upload_date, | |
828 | 'title': video_title, | |
829 | 'ext': video_extension, | |
830 | 'format': video_format, | |
831 | 'thumbnail': video_thumbnail, | |
832 | 'description': video_description, | |
833 | 'player_url': player_url, | |
834 | 'subtitles': video_subtitles, | |
835 | 'duration': video_duration | |
836 | }) | |
837 | return results | |
838 | ||
839 | class YoutubePlaylistIE(InfoExtractor): | |
0f818663 | 840 | IE_DESC = u'YouTube.com playlists' |
c5e8d7af PH |
841 | _VALID_URL = r"""(?: |
842 | (?:https?://)? | |
843 | (?:\w+\.)? | |
844 | youtube\.com/ | |
845 | (?: | |
846 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
847 | \? (?:.*?&)*? (?:p|a|list)= | |
848 | | p/ | |
849 | ) | |
c626a3d9 | 850 | ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,}) |
c5e8d7af PH |
851 | .* |
852 | | | |
c626a3d9 | 853 | ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,}) |
c5e8d7af PH |
854 | )""" |
855 | _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none' | |
856 | _MAX_RESULTS = 50 | |
857 | IE_NAME = u'youtube:playlist' | |
858 | ||
859 | @classmethod | |
860 | def suitable(cls, url): | |
861 | """Receives a URL and returns True if suitable for this IE.""" | |
862 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
863 | ||
864 | def _real_extract(self, url): | |
865 | # Extract playlist id | |
866 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
867 | if mobj is None: | |
868 | raise ExtractorError(u'Invalid URL: %s' % url) | |
869 | ||
870 | # Download playlist videos from API | |
871 | playlist_id = mobj.group(1) or mobj.group(2) | |
c5e8d7af PH |
872 | videos = [] |
873 | ||
755eb032 | 874 | for page_num in itertools.count(1): |
771822eb JMF |
875 | start_index = self._MAX_RESULTS * (page_num - 1) + 1 |
876 | if start_index >= 1000: | |
877 | self._downloader.report_warning(u'Max number of results reached') | |
878 | break | |
879 | url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index) | |
c5e8d7af PH |
880 | page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) |
881 | ||
882 | try: | |
883 | response = json.loads(page) | |
884 | except ValueError as err: | |
885 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
886 | ||
887 | if 'feed' not in response: | |
888 | raise ExtractorError(u'Got a malformed response from YouTube API') | |
889 | playlist_title = response['feed']['title']['$t'] | |
890 | if 'entry' not in response['feed']: | |
891 | # Number of videos is a multiple of self._MAX_RESULTS | |
892 | break | |
893 | ||
894 | for entry in response['feed']['entry']: | |
895 | index = entry['yt$position']['$t'] | |
c215217e JMF |
896 | if 'media$group' in entry and 'yt$videoid' in entry['media$group']: |
897 | videos.append(( | |
898 | index, | |
899 | 'https://www.youtube.com/watch?v=' + entry['media$group']['yt$videoid']['$t'] | |
900 | )) | |
c5e8d7af | 901 | |
c5e8d7af PH |
902 | videos = [v[1] for v in sorted(videos)] |
903 | ||
20c3893f | 904 | url_results = [self.url_result(vurl, 'Youtube') for vurl in videos] |
c5e8d7af PH |
905 | return [self.playlist_result(url_results, playlist_id, playlist_title)] |
906 | ||
907 | ||
908 | class YoutubeChannelIE(InfoExtractor): | |
0f818663 | 909 | IE_DESC = u'YouTube.com channels' |
c5e8d7af PH |
910 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
911 | _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en' | |
912 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' | |
252580c5 | 913 | _MORE_PAGES_URL = 'http://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' |
c5e8d7af PH |
914 | IE_NAME = u'youtube:channel' |
915 | ||
916 | def extract_videos_from_page(self, page): | |
917 | ids_in_page = [] | |
918 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
919 | if mobj.group(1) not in ids_in_page: | |
920 | ids_in_page.append(mobj.group(1)) | |
921 | return ids_in_page | |
922 | ||
923 | def _real_extract(self, url): | |
924 | # Extract channel id | |
925 | mobj = re.match(self._VALID_URL, url) | |
926 | if mobj is None: | |
927 | raise ExtractorError(u'Invalid URL: %s' % url) | |
928 | ||
929 | # Download channel page | |
930 | channel_id = mobj.group(1) | |
931 | video_ids = [] | |
932 | pagenum = 1 | |
933 | ||
934 | url = self._TEMPLATE_URL % (channel_id, pagenum) | |
935 | page = self._download_webpage(url, channel_id, | |
936 | u'Downloading page #%s' % pagenum) | |
937 | ||
938 | # Extract video identifiers | |
939 | ids_in_page = self.extract_videos_from_page(page) | |
940 | video_ids.extend(ids_in_page) | |
941 | ||
942 | # Download any subsequent channel pages using the json-based channel_ajax query | |
943 | if self._MORE_PAGES_INDICATOR in page: | |
755eb032 | 944 | for pagenum in itertools.count(1): |
c5e8d7af PH |
945 | url = self._MORE_PAGES_URL % (pagenum, channel_id) |
946 | page = self._download_webpage(url, channel_id, | |
947 | u'Downloading page #%s' % pagenum) | |
948 | ||
949 | page = json.loads(page) | |
950 | ||
951 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
952 | video_ids.extend(ids_in_page) | |
953 | ||
954 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
955 | break | |
956 | ||
957 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
958 | ||
959 | urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids] | |
20c3893f | 960 | url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls] |
c5e8d7af PH |
961 | return [self.playlist_result(url_entries, channel_id)] |
962 | ||
963 | ||
964 | class YoutubeUserIE(InfoExtractor): | |
0f818663 | 965 | IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' |
faab1d38 | 966 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?)|ytuser:)(?!feed/)([A-Za-z0-9_-]+)' |
c5e8d7af PH |
967 | _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' |
968 | _GDATA_PAGE_SIZE = 50 | |
fd9cf738 | 969 | _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' |
c5e8d7af PH |
970 | IE_NAME = u'youtube:user' |
971 | ||
e3ea4790 | 972 | @classmethod |
f4b05232 | 973 | def suitable(cls, url): |
e3ea4790 JMF |
974 | # Don't return True if the url can be extracted with other youtube |
975 | # extractor, the regex would is too permissive and it would match. | |
976 | other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) | |
977 | if any(ie.suitable(url) for ie in other_ies): return False | |
f4b05232 JMF |
978 | else: return super(YoutubeUserIE, cls).suitable(url) |
979 | ||
c5e8d7af PH |
980 | def _real_extract(self, url): |
981 | # Extract username | |
982 | mobj = re.match(self._VALID_URL, url) | |
983 | if mobj is None: | |
984 | raise ExtractorError(u'Invalid URL: %s' % url) | |
985 | ||
986 | username = mobj.group(1) | |
987 | ||
988 | # Download video ids using YouTube Data API. Result size per | |
989 | # query is limited (currently to 50 videos) so we need to query | |
990 | # page by page until there are no video ids - it means we got | |
991 | # all of them. | |
992 | ||
993 | video_ids = [] | |
c5e8d7af | 994 | |
755eb032 | 995 | for pagenum in itertools.count(0): |
c5e8d7af PH |
996 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 |
997 | ||
998 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
999 | page = self._download_webpage(gdata_url, username, | |
1000 | u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) | |
1001 | ||
fd9cf738 JMF |
1002 | try: |
1003 | response = json.loads(page) | |
1004 | except ValueError as err: | |
1005 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
1006 | ||
c5e8d7af PH |
1007 | # Extract video identifiers |
1008 | ids_in_page = [] | |
fd9cf738 JMF |
1009 | for entry in response['feed']['entry']: |
1010 | ids_in_page.append(entry['id']['$t'].split('/')[-1]) | |
c5e8d7af PH |
1011 | video_ids.extend(ids_in_page) |
1012 | ||
1013 | # A little optimization - if current page is not | |
1014 | # "full", ie. does not contain PAGE_SIZE video ids then | |
1015 | # we can assume that this page is the last one - there | |
1016 | # are no more ids on further pages - no need to query | |
1017 | # again. | |
1018 | ||
1019 | if len(ids_in_page) < self._GDATA_PAGE_SIZE: | |
1020 | break | |
1021 | ||
c5e8d7af | 1022 | urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] |
20c3893f | 1023 | url_results = [self.url_result(rurl, 'Youtube') for rurl in urls] |
c5e8d7af | 1024 | return [self.playlist_result(url_results, playlist_title = username)] |
b05654f0 PH |
1025 | |
1026 | class YoutubeSearchIE(SearchInfoExtractor): | |
0f818663 | 1027 | IE_DESC = u'YouTube.com searches' |
b05654f0 PH |
1028 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' |
1029 | _MAX_RESULTS = 1000 | |
1030 | IE_NAME = u'youtube:search' | |
1031 | _SEARCH_KEY = 'ytsearch' | |
1032 | ||
1033 | def report_download_page(self, query, pagenum): | |
1034 | """Report attempt to download search page with given number.""" | |
1035 | self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) | |
1036 | ||
1037 | def _get_n_results(self, query, n): | |
1038 | """Get a specified number of results for a query""" | |
1039 | ||
1040 | video_ids = [] | |
1041 | pagenum = 0 | |
1042 | limit = n | |
1043 | ||
1044 | while (50 * pagenum) < limit: | |
1045 | self.report_download_page(query, pagenum+1) | |
1046 | result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) | |
1047 | request = compat_urllib_request.Request(result_url) | |
1048 | try: | |
1049 | data = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
1050 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
1051 | raise ExtractorError(u'Unable to download API page: %s' % compat_str(err)) | |
1052 | api_response = json.loads(data)['data'] | |
1053 | ||
1054 | if not 'items' in api_response: | |
1055 | raise ExtractorError(u'[youtube] No video results') | |
1056 | ||
1057 | new_ids = list(video['id'] for video in api_response['items']) | |
1058 | video_ids += new_ids | |
1059 | ||
1060 | limit = min(n, api_response['totalItems']) | |
1061 | pagenum += 1 | |
1062 | ||
1063 | if len(video_ids) > n: | |
1064 | video_ids = video_ids[:n] | |
1065 | videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids] | |
1066 | return self.playlist_result(videos, query) | |
75dff0ee JMF |
1067 | |
1068 | ||
1069 | class YoutubeShowIE(InfoExtractor): | |
0f818663 | 1070 | IE_DESC = u'YouTube.com (multi-season) shows' |
75dff0ee JMF |
1071 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
1072 | IE_NAME = u'youtube:show' | |
1073 | ||
1074 | def _real_extract(self, url): | |
1075 | mobj = re.match(self._VALID_URL, url) | |
1076 | show_name = mobj.group(1) | |
1077 | webpage = self._download_webpage(url, show_name, u'Downloading show webpage') | |
1078 | # There's one playlist for each season of the show | |
1079 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
1080 | self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) | |
1081 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] | |
04cc9617 JMF |
1082 | |
1083 | ||
b2e8bc1b | 1084 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): |
d7ae0639 JMF |
1085 | """ |
1086 | Base class for extractors that fetch info from | |
1087 | http://www.youtube.com/feed_ajax | |
1088 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
1089 | """ | |
b2e8bc1b | 1090 | _LOGIN_REQUIRED = True |
04cc9617 | 1091 | _PAGING_STEP = 30 |
43ba5456 JMF |
1092 | # use action_load_personal_feed instead of action_load_system_feed |
1093 | _PERSONAL_FEED = False | |
04cc9617 | 1094 | |
d7ae0639 JMF |
1095 | @property |
1096 | def _FEED_TEMPLATE(self): | |
43ba5456 JMF |
1097 | action = 'action_load_system_feed' |
1098 | if self._PERSONAL_FEED: | |
1099 | action = 'action_load_personal_feed' | |
1100 | return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) | |
d7ae0639 JMF |
1101 | |
1102 | @property | |
1103 | def IE_NAME(self): | |
1104 | return u'youtube:%s' % self._FEED_NAME | |
04cc9617 | 1105 | |
81f0259b | 1106 | def _real_initialize(self): |
b2e8bc1b | 1107 | self._login() |
81f0259b | 1108 | |
04cc9617 JMF |
1109 | def _real_extract(self, url): |
1110 | feed_entries = [] | |
1111 | # The step argument is available only in 2.7 or higher | |
1112 | for i in itertools.count(0): | |
1113 | paging = i*self._PAGING_STEP | |
d7ae0639 JMF |
1114 | info = self._download_webpage(self._FEED_TEMPLATE % paging, |
1115 | u'%s feed' % self._FEED_NAME, | |
04cc9617 JMF |
1116 | u'Downloading page %s' % i) |
1117 | info = json.loads(info) | |
1118 | feed_html = info['feed_html'] | |
43ba5456 | 1119 | m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) |
04cc9617 JMF |
1120 | ids = orderedSet(m.group(1) for m in m_ids) |
1121 | feed_entries.extend(self.url_result(id, 'Youtube') for id in ids) | |
1122 | if info['paging'] is None: | |
1123 | break | |
d7ae0639 JMF |
1124 | return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) |
1125 | ||
1126 | class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): | |
1127 | IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' | |
1128 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
1129 | _FEED_NAME = 'subscriptions' | |
1130 | _PLAYLIST_TITLE = u'Youtube Subscriptions' | |
1131 | ||
1132 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): | |
1133 | IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' | |
1134 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' | |
1135 | _FEED_NAME = 'recommended' | |
1136 | _PLAYLIST_TITLE = u'Youtube Recommended videos' | |
c626a3d9 | 1137 | |
43ba5456 JMF |
1138 | class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): |
1139 | IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' | |
1140 | _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' | |
1141 | _FEED_NAME = 'watch_later' | |
1142 | _PLAYLIST_TITLE = u'Youtube Watch Later' | |
1143 | _PAGING_STEP = 100 | |
1144 | _PERSONAL_FEED = True | |
c626a3d9 JMF |
1145 | |
1146 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): | |
1147 | IE_NAME = u'youtube:favorites' | |
1148 | IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' | |
c7a7750d | 1149 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?' |
c626a3d9 JMF |
1150 | _LOGIN_REQUIRED = True |
1151 | ||
1152 | def _real_extract(self, url): | |
1153 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
1154 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id') | |
1155 | return self.url_result(playlist_id, 'YoutubePlaylist') |