]>
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 |
c5e8d7af | 8 | |
b05654f0 | 9 | from .common import InfoExtractor, SearchInfoExtractor |
c5e8d7af PH |
10 | from ..utils import ( |
11 | compat_http_client, | |
12 | compat_parse_qs, | |
13 | compat_urllib_error, | |
14 | compat_urllib_parse, | |
15 | compat_urllib_request, | |
16 | compat_str, | |
17 | ||
18 | clean_html, | |
19 | get_element_by_id, | |
20 | ExtractorError, | |
21 | unescapeHTML, | |
22 | unified_strdate, | |
04cc9617 | 23 | orderedSet, |
c5e8d7af PH |
24 | ) |
25 | ||
b2e8bc1b JMF |
26 | class YoutubeBaseInfoExtractor(InfoExtractor): |
27 | """Provide base functions for Youtube extractors""" | |
28 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
29 | _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' | |
30 | _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' | |
31 | _NETRC_MACHINE = 'youtube' | |
32 | # If True it will raise an error if no login info is provided | |
33 | _LOGIN_REQUIRED = False | |
34 | ||
35 | def report_lang(self): | |
36 | """Report attempt to set language.""" | |
37 | self.to_screen(u'Setting language') | |
38 | ||
39 | def _set_language(self): | |
40 | request = compat_urllib_request.Request(self._LANG_URL) | |
41 | try: | |
42 | self.report_lang() | |
43 | compat_urllib_request.urlopen(request).read() | |
44 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
45 | self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) | |
46 | return False | |
47 | return True | |
48 | ||
49 | def _login(self): | |
50 | (username, password) = self._get_login_info() | |
51 | # No authentication to be performed | |
52 | if username is None: | |
53 | if self._LOGIN_REQUIRED: | |
54 | raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) | |
55 | return False | |
56 | ||
57 | request = compat_urllib_request.Request(self._LOGIN_URL) | |
58 | try: | |
59 | login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
60 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
61 | self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) | |
62 | return False | |
63 | ||
64 | galx = None | |
65 | dsh = None | |
66 | match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page) | |
67 | if match: | |
68 | galx = match.group(1) | |
69 | match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page) | |
70 | if match: | |
71 | dsh = match.group(1) | |
c5e8d7af | 72 | |
b2e8bc1b JMF |
73 | # Log in |
74 | login_form_strs = { | |
75 | u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
76 | u'Email': username, | |
77 | u'GALX': galx, | |
78 | u'Passwd': password, | |
79 | u'PersistentCookie': u'yes', | |
80 | u'_utf8': u'霱', | |
81 | u'bgresponse': u'js_disabled', | |
82 | u'checkConnection': u'', | |
83 | u'checkedDomains': u'youtube', | |
84 | u'dnConn': u'', | |
85 | u'dsh': dsh, | |
86 | u'pstMsg': u'0', | |
87 | u'rmShown': u'1', | |
88 | u'secTok': u'', | |
89 | u'signIn': u'Sign in', | |
90 | u'timeStmp': u'', | |
91 | u'service': u'youtube', | |
92 | u'uilel': u'3', | |
93 | u'hl': u'en_US', | |
94 | } | |
95 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
96 | # chokes on unicode | |
97 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) | |
98 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
99 | request = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
100 | try: | |
101 | self.report_login() | |
102 | login_results = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
103 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
104 | self._downloader.report_warning(u'unable to log in: bad username or password') | |
105 | return False | |
106 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
107 | self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) | |
108 | return False | |
109 | return True | |
110 | ||
111 | def _confirm_age(self): | |
112 | age_form = { | |
113 | 'next_url': '/', | |
114 | 'action_confirm': 'Confirm', | |
115 | } | |
116 | request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) | |
117 | try: | |
118 | self.report_age_confirmation() | |
119 | compat_urllib_request.urlopen(request).read().decode('utf-8') | |
120 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
121 | raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) | |
122 | return True | |
123 | ||
124 | def _real_initialize(self): | |
125 | if self._downloader is None: | |
126 | return | |
127 | if not self._set_language(): | |
128 | return | |
129 | if not self._login(): | |
130 | return | |
131 | self._confirm_age() | |
132 | ||
133 | class YoutubeIE(YoutubeBaseInfoExtractor): | |
0f818663 | 134 | IE_DESC = u'YouTube.com' |
c5e8d7af PH |
135 | _VALID_URL = r"""^ |
136 | ( | |
137 | (?:https?://)? # http(s):// (optional) | |
138 | (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/| | |
139 | tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains | |
140 | (?:.*?\#/)? # handle anchor (#/) redirect urls | |
141 | (?: # the various things that can precede the ID: | |
142 | (?:(?:v|embed|e)/) # v/ or embed/ or e/ | |
143 | |(?: # or the v= param in all its forms | |
d741e55a | 144 | (?:(?:watch|movie)(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) |
c5e8d7af PH |
145 | (?:\?|\#!?) # the params delimiter ? or # or #! |
146 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
147 | v= | |
148 | ) | |
149 | )? # optional -> youtube.com/xxxx is OK | |
150 | )? # all until now is optional -> you can pass the naked ID | |
151 | ([0-9A-Za-z_-]+) # here is it! the YouTube video ID | |
152 | (?(1).+)? # if we found the ID, everything can follow | |
153 | $""" | |
c5e8d7af | 154 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' |
c5e8d7af | 155 | # Listed in order of quality |
1d043b93 JMF |
156 | _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13', |
157 | '95', '94', '93', '92', '132', '151', | |
939fbd26 JMF |
158 | # 3D |
159 | '85', '84', '102', '83', '101', '82', '100', | |
160 | # Dash video | |
161 | '138', '137', '248', '136', '247', '135', '246', | |
162 | '245', '244', '134', '243', '133', '242', '160', | |
163 | # Dash audio | |
164 | '141', '172', '140', '171', '139', | |
1d043b93 JMF |
165 | ] |
166 | _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13', | |
167 | '95', '94', '93', '92', '132', '151', | |
86fe61c8 | 168 | '85', '102', '84', '101', '83', '100', '82', |
939fbd26 JMF |
169 | # Dash video |
170 | '138', '248', '137', '247', '136', '246', '245', | |
171 | '244', '135', '243', '134', '242', '133', '160', | |
172 | # Dash audio | |
173 | '172', '141', '171', '140', '139', | |
1d043b93 | 174 | ] |
c5e8d7af PH |
175 | _video_extensions = { |
176 | '13': '3gp', | |
177 | '17': 'mp4', | |
178 | '18': 'mp4', | |
179 | '22': 'mp4', | |
180 | '37': 'mp4', | |
d69cf69a | 181 | '38': 'mp4', |
c5e8d7af PH |
182 | '43': 'webm', |
183 | '44': 'webm', | |
184 | '45': 'webm', | |
185 | '46': 'webm', | |
1d043b93 | 186 | |
86fe61c8 AZ |
187 | # 3d videos |
188 | '82': 'mp4', | |
189 | '83': 'mp4', | |
190 | '84': 'mp4', | |
191 | '85': 'mp4', | |
192 | '100': 'webm', | |
193 | '101': 'webm', | |
194 | '102': 'webm', | |
836a086c | 195 | |
1d043b93 JMF |
196 | # videos that use m3u8 |
197 | '92': 'mp4', | |
198 | '93': 'mp4', | |
199 | '94': 'mp4', | |
200 | '95': 'mp4', | |
201 | '96': 'mp4', | |
202 | '132': 'mp4', | |
203 | '151': 'mp4', | |
836a086c AZ |
204 | |
205 | # Dash mp4 | |
206 | '133': 'mp4', | |
207 | '134': 'mp4', | |
208 | '135': 'mp4', | |
209 | '136': 'mp4', | |
210 | '137': 'mp4', | |
211 | '138': 'mp4', | |
212 | '139': 'mp4', | |
213 | '140': 'mp4', | |
214 | '141': 'mp4', | |
215 | '160': 'mp4', | |
216 | ||
217 | # Dash webm | |
218 | '171': 'webm', | |
219 | '172': 'webm', | |
220 | '242': 'webm', | |
221 | '243': 'webm', | |
222 | '244': 'webm', | |
223 | '245': 'webm', | |
224 | '246': 'webm', | |
225 | '247': 'webm', | |
226 | '248': 'webm', | |
c5e8d7af PH |
227 | } |
228 | _video_dimensions = { | |
229 | '5': '240x400', | |
230 | '6': '???', | |
231 | '13': '???', | |
232 | '17': '144x176', | |
233 | '18': '360x640', | |
234 | '22': '720x1280', | |
235 | '34': '360x640', | |
236 | '35': '480x854', | |
237 | '37': '1080x1920', | |
238 | '38': '3072x4096', | |
239 | '43': '360x640', | |
240 | '44': '480x854', | |
241 | '45': '720x1280', | |
242 | '46': '1080x1920', | |
86fe61c8 AZ |
243 | '82': '360p', |
244 | '83': '480p', | |
245 | '84': '720p', | |
246 | '85': '1080p', | |
1d043b93 JMF |
247 | '92': '240p', |
248 | '93': '360p', | |
249 | '94': '480p', | |
250 | '95': '720p', | |
251 | '96': '1080p', | |
86fe61c8 AZ |
252 | '100': '360p', |
253 | '101': '480p', | |
836a086c | 254 | '102': '720p', |
1d043b93 JMF |
255 | '132': '240p', |
256 | '151': '72p', | |
836a086c AZ |
257 | '133': '240p', |
258 | '134': '360p', | |
259 | '135': '480p', | |
260 | '136': '720p', | |
261 | '137': '1080p', | |
262 | '138': '>1080p', | |
263 | '139': '48k', | |
264 | '140': '128k', | |
265 | '141': '256k', | |
266 | '160': '192p', | |
267 | '171': '128k', | |
268 | '172': '256k', | |
269 | '242': '240p', | |
270 | '243': '360p', | |
271 | '244': '480p', | |
272 | '245': '480p', | |
273 | '246': '480p', | |
274 | '247': '720p', | |
275 | '248': '1080p', | |
c5e8d7af | 276 | } |
836a086c AZ |
277 | _special_itags = { |
278 | '82': '3D', | |
279 | '83': '3D', | |
280 | '84': '3D', | |
281 | '85': '3D', | |
282 | '100': '3D', | |
283 | '101': '3D', | |
284 | '102': '3D', | |
285 | '133': 'DASH Video', | |
286 | '134': 'DASH Video', | |
287 | '135': 'DASH Video', | |
288 | '136': 'DASH Video', | |
289 | '137': 'DASH Video', | |
290 | '138': 'DASH Video', | |
291 | '139': 'DASH Audio', | |
292 | '140': 'DASH Audio', | |
293 | '141': 'DASH Audio', | |
294 | '160': 'DASH Video', | |
295 | '171': 'DASH Audio', | |
296 | '172': 'DASH Audio', | |
297 | '242': 'DASH Video', | |
298 | '243': 'DASH Video', | |
299 | '244': 'DASH Video', | |
300 | '245': 'DASH Video', | |
301 | '246': 'DASH Video', | |
302 | '247': 'DASH Video', | |
303 | '248': 'DASH Video', | |
c5e8d7af | 304 | } |
836a086c | 305 | |
c5e8d7af | 306 | IE_NAME = u'youtube' |
2eb88d95 PH |
307 | _TESTS = [ |
308 | { | |
0e853ca4 PH |
309 | u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", |
310 | u"file": u"BaW_jenozKc.mp4", | |
311 | u"info_dict": { | |
312 | u"title": u"youtube-dl test video \"'/\\ä↭𝕐", | |
313 | u"uploader": u"Philipp Hagemeister", | |
314 | u"uploader_id": u"phihag", | |
315 | u"upload_date": u"20121002", | |
316 | 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 | 317 | } |
0e853ca4 PH |
318 | }, |
319 | { | |
320 | u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U", | |
321 | u"file": u"1ltcDfZMA3U.flv", | |
322 | u"note": u"Test VEVO video (#897)", | |
323 | u"info_dict": { | |
324 | u"upload_date": u"20070518", | |
325 | u"title": u"Maps - It Will Find You", | |
326 | u"description": u"Music video by Maps performing It Will Find You.", | |
327 | u"uploader": u"MuteUSA", | |
328 | u"uploader_id": u"MuteUSA" | |
2eb88d95 | 329 | } |
0e853ca4 PH |
330 | }, |
331 | { | |
332 | u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", | |
333 | u"file": u"UxxajLWwzqY.mp4", | |
334 | u"note": u"Test generic use_cipher_signature video (#897)", | |
335 | u"info_dict": { | |
336 | u"upload_date": u"20120506", | |
337 | u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", | |
338 | u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c", | |
45ed795c | 339 | u"uploader": u"Icona Pop", |
0e853ca4 | 340 | u"uploader_id": u"IconaPop" |
2eb88d95 | 341 | } |
c108eb73 JMF |
342 | }, |
343 | { | |
344 | u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", | |
345 | u"file": u"07FYdnEawAQ.mp4", | |
346 | u"note": u"Test VEVO video with age protection (#956)", | |
347 | u"info_dict": { | |
348 | u"upload_date": u"20130703", | |
349 | u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", | |
350 | u"description": u"md5:64249768eec3bc4276236606ea996373", | |
351 | u"uploader": u"justintimberlakeVEVO", | |
352 | u"uploader_id": u"justintimberlakeVEVO" | |
353 | } | |
354 | }, | |
1d043b93 JMF |
355 | { |
356 | u'url': u'https://www.youtube.com/watch?v=TGi3HqYrWHE', | |
357 | u'file': u'TGi3HqYrWHE.mp4', | |
358 | u'note': u'm3u8 video', | |
359 | u'info_dict': { | |
360 | u'title': u'Triathlon - Men - London 2012 Olympic Games', | |
361 | u'description': u'- Men - TR02 - Triathlon - 07 August 2012 - London 2012 Olympic Games', | |
362 | u'uploader': u'olympic', | |
363 | u'upload_date': u'20120807', | |
364 | u'uploader_id': u'olympic', | |
365 | }, | |
366 | u'params': { | |
367 | u'skip_download': True, | |
368 | }, | |
369 | }, | |
2eb88d95 PH |
370 | ] |
371 | ||
c5e8d7af PH |
372 | |
373 | @classmethod | |
374 | def suitable(cls, url): | |
375 | """Receives a URL and returns True if suitable for this IE.""" | |
04cc9617 | 376 | if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False |
c5e8d7af PH |
377 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None |
378 | ||
c5e8d7af PH |
379 | def report_video_webpage_download(self, video_id): |
380 | """Report attempt to download video webpage.""" | |
381 | self.to_screen(u'%s: Downloading video webpage' % video_id) | |
382 | ||
383 | def report_video_info_webpage_download(self, video_id): | |
384 | """Report attempt to download video info webpage.""" | |
385 | self.to_screen(u'%s: Downloading video info webpage' % video_id) | |
386 | ||
387 | def report_video_subtitles_download(self, video_id): | |
388 | """Report attempt to download video info webpage.""" | |
389 | self.to_screen(u'%s: Checking available subtitles' % video_id) | |
390 | ||
391 | def report_video_subtitles_request(self, video_id, sub_lang, format): | |
392 | """Report attempt to download video info webpage.""" | |
393 | self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format)) | |
394 | ||
395 | def report_video_subtitles_available(self, video_id, sub_lang_list): | |
396 | """Report available subtitles.""" | |
397 | sub_lang = ",".join(list(sub_lang_list.keys())) | |
398 | self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang)) | |
399 | ||
400 | def report_information_extraction(self, video_id): | |
401 | """Report attempt to extract video information.""" | |
402 | self.to_screen(u'%s: Extracting video information' % video_id) | |
403 | ||
404 | def report_unavailable_format(self, video_id, format): | |
405 | """Report extracted video URL.""" | |
406 | self.to_screen(u'%s: Format %s not available' % (video_id, format)) | |
407 | ||
408 | def report_rtmp_download(self): | |
409 | """Indicate the download will use the RTMP protocol.""" | |
410 | self.to_screen(u'RTMP download detected') | |
411 | ||
98bcd283 | 412 | def _decrypt_signature(self, s): |
257a2501 | 413 | """Turn the encrypted s field into a working signature""" |
6b37f0be | 414 | |
444b1165 JMF |
415 | if len(s) == 92: |
416 | return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83] | |
417 | elif len(s) == 90: | |
418 | return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81] | |
8a9d86a2 | 419 | elif len(s) == 89: |
420 | return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1] | |
444b1165 | 421 | elif len(s) == 88: |
be547e1d PH |
422 | return s[48] + s[81:67:-1] + s[82] + s[66:62:-1] + s[85] + s[61:48:-1] + s[67] + s[47:12:-1] + s[3] + s[11:3:-1] + s[2] + s[12] |
423 | elif len(s) == 87: | |
3a725669 | 424 | return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:] |
be547e1d | 425 | elif len(s) == 86: |
30617051 | 426 | return s[5:40] + s[3] + s[41:48] + s[0] + s[49:86] |
be547e1d | 427 | elif len(s) == 85: |
f9c3c90c | 428 | return s[83:34:-1] + s[0] + s[33:27:-1] + s[3] + s[26:19:-1] + s[34] + s[18:3:-1] + s[27] |
be547e1d | 429 | elif len(s) == 84: |
f2aeefe2 | 430 | return s[5:40] + s[3] + s[41:48] + s[0] + s[49:84] |
be547e1d | 431 | elif len(s) == 83: |
e1842025 | 432 | 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 | 433 | elif len(s) == 82: |
3669cdba | 434 | return s[1:19] + s[0] + s[20:68] + s[19] + s[69:82] |
be547e1d | 435 | elif len(s) == 81: |
aedd6bb9 | 436 | 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 |
437 | elif len(s) == 80: |
438 | return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80] | |
5c468ca8 JMF |
439 | elif len(s) == 79: |
440 | 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 |
441 | |
442 | else: | |
443 | raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) | |
c5e8d7af | 444 | |
75952c6e JMF |
445 | def _decrypt_signature_age_gate(self, s): |
446 | # The videos with age protection use another player, so the algorithms | |
447 | # can be different. | |
448 | if len(s) == 86: | |
449 | return s[2:63] + s[82] + s[64:82] + s[63] | |
450 | else: | |
451 | # Fallback to the other algortihms | |
b072a9de | 452 | return self._decrypt_signature(s) |
75952c6e JMF |
453 | |
454 | ||
c5e8d7af PH |
455 | def _get_available_subtitles(self, video_id): |
456 | self.report_video_subtitles_download(video_id) | |
457 | request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id) | |
458 | try: | |
459 | sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
460 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
5d51a883 JMF |
461 | self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) |
462 | return {} | |
c5e8d7af PH |
463 | sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) |
464 | sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list) | |
465 | if not sub_lang_list: | |
5d51a883 JMF |
466 | self._downloader.report_warning(u'video doesn\'t have subtitles') |
467 | return {} | |
c5e8d7af PH |
468 | return sub_lang_list |
469 | ||
470 | def _list_available_subtitles(self, video_id): | |
471 | sub_lang_list = self._get_available_subtitles(video_id) | |
472 | self.report_video_subtitles_available(video_id, sub_lang_list) | |
473 | ||
474 | def _request_subtitle(self, sub_lang, sub_name, video_id, format): | |
475 | """ | |
5d51a883 | 476 | Return the subtitle as a string or None if they are not found |
c5e8d7af PH |
477 | """ |
478 | self.report_video_subtitles_request(video_id, sub_lang, format) | |
479 | params = compat_urllib_parse.urlencode({ | |
480 | 'lang': sub_lang, | |
481 | 'name': sub_name, | |
482 | 'v': video_id, | |
483 | 'fmt': format, | |
484 | }) | |
485 | url = 'http://www.youtube.com/api/timedtext?' + params | |
486 | try: | |
487 | sub = compat_urllib_request.urlopen(url).read().decode('utf-8') | |
488 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
5d51a883 JMF |
489 | self._downloader.report_warning(u'unable to download video subtitles for %s: %s' % (sub_lang, compat_str(err))) |
490 | return | |
c5e8d7af | 491 | if not sub: |
5d51a883 JMF |
492 | self._downloader.report_warning(u'Did not fetch video subtitles') |
493 | return | |
494 | return sub | |
c5e8d7af PH |
495 | |
496 | def _request_automatic_caption(self, video_id, webpage): | |
497 | """We need the webpage for getting the captions url, pass it as an | |
498 | argument to speed up the process.""" | |
aa6a10c4 | 499 | sub_lang = (self._downloader.params.get('subtitleslangs') or ['en'])[0] |
c5e8d7af PH |
500 | sub_format = self._downloader.params.get('subtitlesformat') |
501 | self.to_screen(u'%s: Looking for automatic captions' % video_id) | |
502 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
503 | err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang | |
504 | if mobj is None: | |
5d51a883 JMF |
505 | self._downloader.report_warning(err_msg) |
506 | return {} | |
c5e8d7af PH |
507 | player_config = json.loads(mobj.group(1)) |
508 | try: | |
509 | args = player_config[u'args'] | |
510 | caption_url = args[u'ttsurl'] | |
511 | timestamp = args[u'timestamp'] | |
512 | params = compat_urllib_parse.urlencode({ | |
513 | 'lang': 'en', | |
514 | 'tlang': sub_lang, | |
515 | 'fmt': sub_format, | |
516 | 'ts': timestamp, | |
517 | 'kind': 'asr', | |
518 | }) | |
519 | subtitles_url = caption_url + '&' + params | |
520 | sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions') | |
5d51a883 | 521 | return {sub_lang: sub} |
2f799533 JMF |
522 | # An extractor error can be raise by the download process if there are |
523 | # no automatic captions but there are subtitles | |
524 | except (KeyError, ExtractorError): | |
5d51a883 JMF |
525 | self._downloader.report_warning(err_msg) |
526 | return {} | |
88ae5991 JMF |
527 | |
528 | def _extract_subtitles(self, video_id): | |
c5e8d7af | 529 | """ |
5d51a883 JMF |
530 | Return a dictionary: {language: subtitles} or {} if the subtitles |
531 | couldn't be found | |
c5e8d7af | 532 | """ |
aa6a10c4 | 533 | available_subs_list = self._get_available_subtitles(video_id) |
c5e8d7af | 534 | sub_format = self._downloader.params.get('subtitlesformat') |
aa6a10c4 | 535 | if not available_subs_list: #There was some error, it didn't get the available subtitles |
5d51a883 | 536 | return {} |
df3df7fb | 537 | if self._downloader.params.get('allsubtitles', False): |
aa6a10c4 | 538 | sub_lang_list = available_subs_list |
df3df7fb | 539 | else: |
aa6a10c4 JMF |
540 | if self._downloader.params.get('subtitleslangs', False): |
541 | reqested_langs = self._downloader.params.get('subtitleslangs') | |
542 | elif 'en' in available_subs_list: | |
543 | reqested_langs = ['en'] | |
88ae5991 | 544 | else: |
aa6a10c4 JMF |
545 | reqested_langs = [list(available_subs_list.keys())[0]] |
546 | ||
547 | sub_lang_list = {} | |
548 | for sub_lang in reqested_langs: | |
549 | if not sub_lang in available_subs_list: | |
550 | self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang) | |
551 | continue | |
552 | sub_lang_list[sub_lang] = available_subs_list[sub_lang] | |
5d51a883 | 553 | subtitles = {} |
c5e8d7af PH |
554 | for sub_lang in sub_lang_list: |
555 | subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) | |
5d51a883 JMF |
556 | if subtitle: |
557 | subtitles[sub_lang] = subtitle | |
c5e8d7af PH |
558 | return subtitles |
559 | ||
560 | def _print_formats(self, formats): | |
561 | print('Available formats:') | |
562 | for x in formats: | |
03cc7c20 JMF |
563 | print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'), |
564 | self._video_dimensions.get(x, '???'), | |
836a086c | 565 | ' ('+self._special_itags[x]+')' if x in self._special_itags else '')) |
c5e8d7af | 566 | |
c5e8d7af PH |
567 | def _extract_id(self, url): |
568 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
569 | if mobj is None: | |
570 | raise ExtractorError(u'Invalid URL: %s' % url) | |
571 | video_id = mobj.group(2) | |
572 | return video_id | |
573 | ||
1d043b93 JMF |
574 | def _get_video_url_list(self, url_map): |
575 | """ | |
576 | Transform a dictionary in the format {itag:url} to a list of (itag, url) | |
577 | with the requested formats. | |
578 | """ | |
579 | req_format = self._downloader.params.get('format', None) | |
580 | format_limit = self._downloader.params.get('format_limit', None) | |
581 | available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats | |
582 | if format_limit is not None and format_limit in available_formats: | |
583 | format_list = available_formats[available_formats.index(format_limit):] | |
584 | else: | |
585 | format_list = available_formats | |
586 | existing_formats = [x for x in format_list if x in url_map] | |
587 | if len(existing_formats) == 0: | |
588 | raise ExtractorError(u'no known formats available for video') | |
589 | if self._downloader.params.get('listformats', None): | |
590 | self._print_formats(existing_formats) | |
591 | return | |
592 | if req_format is None or req_format == 'best': | |
593 | video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality | |
594 | elif req_format == 'worst': | |
595 | video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality | |
596 | elif req_format in ('-1', 'all'): | |
597 | video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats | |
598 | else: | |
599 | # Specific formats. We pick the first in a slash-delimeted sequence. | |
600 | # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. | |
601 | req_formats = req_format.split('/') | |
602 | video_url_list = None | |
603 | for rf in req_formats: | |
604 | if rf in url_map: | |
605 | video_url_list = [(rf, url_map[rf])] | |
606 | break | |
607 | if video_url_list is None: | |
608 | raise ExtractorError(u'requested format not available') | |
609 | return video_url_list | |
610 | ||
611 | def _extract_from_m3u8(self, manifest_url, video_id): | |
612 | url_map = {} | |
613 | def _get_urls(_manifest): | |
614 | lines = _manifest.split('\n') | |
615 | urls = filter(lambda l: l and not l.startswith('#'), | |
616 | lines) | |
617 | return urls | |
618 | manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') | |
619 | formats_urls = _get_urls(manifest) | |
620 | for format_url in formats_urls: | |
621 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') | |
622 | url_map[itag] = format_url | |
623 | return url_map | |
624 | ||
c5e8d7af | 625 | def _real_extract(self, url): |
d7f44b5b PH |
626 | if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url): |
627 | 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 ).') | |
628 | ||
c5e8d7af PH |
629 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter |
630 | mobj = re.search(self._NEXT_URL_RE, url) | |
631 | if mobj: | |
632 | url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
633 | video_id = self._extract_id(url) | |
634 | ||
635 | # Get video webpage | |
636 | self.report_video_webpage_download(video_id) | |
637 | url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id | |
638 | request = compat_urllib_request.Request(url) | |
639 | try: | |
640 | video_webpage_bytes = compat_urllib_request.urlopen(request).read() | |
641 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
642 | raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err)) | |
643 | ||
644 | video_webpage = video_webpage_bytes.decode('utf-8', 'ignore') | |
645 | ||
646 | # Attempt to extract SWF player URL | |
647 | mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) | |
648 | if mobj is not None: | |
649 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
650 | else: | |
651 | player_url = None | |
652 | ||
653 | # Get video info | |
654 | self.report_video_info_webpage_download(video_id) | |
c108eb73 JMF |
655 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
656 | self.report_age_confirmation() | |
657 | age_gate = True | |
658 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
659 | # this can be viewed without login into Youtube | |
660 | data = compat_urllib_parse.urlencode({'video_id': video_id, | |
661 | 'el': 'embedded', | |
662 | 'gl': 'US', | |
663 | 'hl': 'en', | |
664 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
665 | 'asv': 3, | |
666 | 'sts':'1588', | |
667 | }) | |
668 | video_info_url = 'https://www.youtube.com/get_video_info?' + data | |
c5e8d7af PH |
669 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
670 | note=False, | |
671 | errnote='unable to download video info webpage') | |
672 | video_info = compat_parse_qs(video_info_webpage) | |
c108eb73 JMF |
673 | else: |
674 | age_gate = False | |
675 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
676 | video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
677 | % (video_id, el_type)) | |
678 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
679 | note=False, | |
680 | errnote='unable to download video info webpage') | |
681 | video_info = compat_parse_qs(video_info_webpage) | |
682 | if 'token' in video_info: | |
683 | break | |
c5e8d7af PH |
684 | if 'token' not in video_info: |
685 | if 'reason' in video_info: | |
9a82b238 | 686 | raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True) |
c5e8d7af PH |
687 | else: |
688 | raise ExtractorError(u'"token" parameter not in video info for unknown reason') | |
689 | ||
690 | # Check for "rental" videos | |
691 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
692 | raise ExtractorError(u'"rental" videos not supported') | |
693 | ||
694 | # Start extracting information | |
695 | self.report_information_extraction(video_id) | |
696 | ||
697 | # uploader | |
698 | if 'author' not in video_info: | |
699 | raise ExtractorError(u'Unable to extract uploader name') | |
700 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
701 | ||
702 | # uploader_id | |
703 | video_uploader_id = None | |
704 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
705 | if mobj is not None: | |
706 | video_uploader_id = mobj.group(1) | |
707 | else: | |
708 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
709 | ||
710 | # title | |
711 | if 'title' not in video_info: | |
712 | raise ExtractorError(u'Unable to extract video title') | |
713 | video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) | |
714 | ||
715 | # thumbnail image | |
7763b04e JMF |
716 | # We try first to get a high quality image: |
717 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
718 | video_webpage, re.DOTALL) | |
719 | if m_thumb is not None: | |
720 | video_thumbnail = m_thumb.group(1) | |
721 | elif 'thumbnail_url' not in video_info: | |
c5e8d7af PH |
722 | self._downloader.report_warning(u'unable to extract video thumbnail') |
723 | video_thumbnail = '' | |
724 | else: # don't panic if we can't find it | |
725 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
726 | ||
727 | # upload date | |
728 | upload_date = None | |
729 | mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL) | |
730 | if mobj is not None: | |
731 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
732 | upload_date = unified_strdate(upload_date) | |
733 | ||
734 | # description | |
735 | video_description = get_element_by_id("eow-description", video_webpage) | |
736 | if video_description: | |
737 | video_description = clean_html(video_description) | |
738 | else: | |
739 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
740 | if fd_mobj: | |
741 | video_description = unescapeHTML(fd_mobj.group(1)) | |
742 | else: | |
743 | video_description = u'' | |
744 | ||
745 | # subtitles | |
746 | video_subtitles = None | |
747 | ||
88ae5991 JMF |
748 | if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False): |
749 | video_subtitles = self._extract_subtitles(video_id) | |
5d51a883 | 750 | elif self._downloader.params.get('writeautomaticsub', False): |
b004821f | 751 | video_subtitles = self._request_automatic_caption(video_id, video_webpage) |
c5e8d7af PH |
752 | |
753 | if self._downloader.params.get('listsubtitles', False): | |
93d3a642 | 754 | self._list_available_subtitles(video_id) |
c5e8d7af PH |
755 | return |
756 | ||
757 | if 'length_seconds' not in video_info: | |
758 | self._downloader.report_warning(u'unable to extract video duration') | |
759 | video_duration = '' | |
760 | else: | |
761 | video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]) | |
762 | ||
c5e8d7af | 763 | # Decide which formats to download |
c5e8d7af PH |
764 | |
765 | try: | |
766 | mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) | |
50be92c1 PH |
767 | if not mobj: |
768 | raise ValueError('Could not find vevo ID') | |
c5e8d7af PH |
769 | info = json.loads(mobj.group(1)) |
770 | args = info['args'] | |
7ce7e394 JMF |
771 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
772 | # this signatures are encrypted | |
773 | m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map']) | |
774 | if m_s is not None: | |
775 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
c5e8d7af | 776 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
cde846b3 | 777 | m_s = re.search(r'[&,]s=', args.get('adaptive_fmts', u'')) |
b7a68384 | 778 | if m_s is not None: |
37b6d5f6 AZ |
779 | if 'url_encoded_fmt_stream_map' in video_info: |
780 | video_info['url_encoded_fmt_stream_map'][0] += ',' + args['adaptive_fmts'] | |
781 | else: | |
782 | video_info['url_encoded_fmt_stream_map'] = [args['adaptive_fmts']] | |
211fbc13 | 783 | elif 'adaptive_fmts' in video_info: |
37b6d5f6 AZ |
784 | if 'url_encoded_fmt_stream_map' in video_info: |
785 | video_info['url_encoded_fmt_stream_map'][0] += ',' + video_info['adaptive_fmts'][0] | |
786 | else: | |
787 | video_info['url_encoded_fmt_stream_map'] = video_info['adaptive_fmts'] | |
c5e8d7af PH |
788 | except ValueError: |
789 | pass | |
790 | ||
791 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
792 | self.report_rtmp_download() | |
793 | video_url_list = [(None, video_info['conn'][0])] | |
794 | elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1: | |
a7055eb9 JMF |
795 | if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]: |
796 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) | |
c5e8d7af PH |
797 | url_map = {} |
798 | for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): | |
799 | url_data = compat_parse_qs(url_data_str) | |
800 | if 'itag' in url_data and 'url' in url_data: | |
801 | url = url_data['url'][0] | |
802 | if 'sig' in url_data: | |
803 | url += '&signature=' + url_data['sig'][0] | |
804 | elif 's' in url_data: | |
769fda3c FV |
805 | if self._downloader.params.get('verbose'): |
806 | s = url_data['s'][0] | |
c108eb73 JMF |
807 | if age_gate: |
808 | player_version = self._search_regex(r'ad3-(.+?)\.swf', | |
ed27d356 JMF |
809 | video_info['ad3_module'][0] if 'ad3_module' in video_info else 'NOT FOUND', |
810 | 'flash player', fatal=False) | |
c108eb73 JMF |
811 | player = 'flash player %s' % player_version |
812 | else: | |
813 | player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage, | |
814 | 'html5 player', fatal=False) | |
5a76c651 JMF |
815 | parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.')) |
816 | self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % | |
817 | (len(s), parts_sizes, url_data['itag'][0], player)) | |
75952c6e JMF |
818 | encrypted_sig = url_data['s'][0] |
819 | if age_gate: | |
820 | signature = self._decrypt_signature_age_gate(encrypted_sig) | |
821 | else: | |
822 | signature = self._decrypt_signature(encrypted_sig) | |
c5e8d7af PH |
823 | url += '&signature=' + signature |
824 | if 'ratebypass' not in url: | |
825 | url += '&ratebypass=yes' | |
826 | url_map[url_data['itag'][0]] = url | |
1d043b93 JMF |
827 | video_url_list = self._get_video_url_list(url_map) |
828 | if not video_url_list: | |
c5e8d7af | 829 | return |
1d043b93 JMF |
830 | elif video_info.get('hlsvp'): |
831 | manifest_url = video_info['hlsvp'][0] | |
832 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
833 | video_url_list = self._get_video_url_list(url_map) | |
834 | if not video_url_list: | |
835 | return | |
836 | ||
c5e8d7af PH |
837 | else: |
838 | raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info') | |
839 | ||
840 | results = [] | |
841 | for format_param, video_real_url in video_url_list: | |
842 | # Extension | |
843 | video_extension = self._video_extensions.get(format_param, 'flv') | |
844 | ||
03cc7c20 JMF |
845 | video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension, |
846 | self._video_dimensions.get(format_param, '???'), | |
836a086c | 847 | ' ('+self._special_itags[format_param]+')' if format_param in self._special_itags else '') |
c5e8d7af PH |
848 | |
849 | results.append({ | |
850 | 'id': video_id, | |
851 | 'url': video_real_url, | |
852 | 'uploader': video_uploader, | |
853 | 'uploader_id': video_uploader_id, | |
854 | 'upload_date': upload_date, | |
855 | 'title': video_title, | |
856 | 'ext': video_extension, | |
857 | 'format': video_format, | |
858 | 'thumbnail': video_thumbnail, | |
859 | 'description': video_description, | |
860 | 'player_url': player_url, | |
861 | 'subtitles': video_subtitles, | |
862 | 'duration': video_duration | |
863 | }) | |
864 | return results | |
865 | ||
866 | class YoutubePlaylistIE(InfoExtractor): | |
0f818663 | 867 | IE_DESC = u'YouTube.com playlists' |
c5e8d7af PH |
868 | _VALID_URL = r"""(?: |
869 | (?:https?://)? | |
870 | (?:\w+\.)? | |
871 | youtube\.com/ | |
872 | (?: | |
873 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
874 | \? (?:.*?&)*? (?:p|a|list)= | |
875 | | p/ | |
876 | ) | |
c626a3d9 | 877 | ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,}) |
c5e8d7af PH |
878 | .* |
879 | | | |
c626a3d9 | 880 | ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,}) |
c5e8d7af PH |
881 | )""" |
882 | _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none' | |
883 | _MAX_RESULTS = 50 | |
884 | IE_NAME = u'youtube:playlist' | |
885 | ||
886 | @classmethod | |
887 | def suitable(cls, url): | |
888 | """Receives a URL and returns True if suitable for this IE.""" | |
889 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
890 | ||
891 | def _real_extract(self, url): | |
892 | # Extract playlist id | |
893 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
894 | if mobj is None: | |
895 | raise ExtractorError(u'Invalid URL: %s' % url) | |
896 | ||
897 | # Download playlist videos from API | |
898 | playlist_id = mobj.group(1) or mobj.group(2) | |
c5e8d7af PH |
899 | videos = [] |
900 | ||
755eb032 | 901 | for page_num in itertools.count(1): |
771822eb JMF |
902 | start_index = self._MAX_RESULTS * (page_num - 1) + 1 |
903 | if start_index >= 1000: | |
904 | self._downloader.report_warning(u'Max number of results reached') | |
905 | break | |
906 | url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index) | |
c5e8d7af PH |
907 | page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) |
908 | ||
909 | try: | |
910 | response = json.loads(page) | |
911 | except ValueError as err: | |
912 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
913 | ||
914 | if 'feed' not in response: | |
915 | raise ExtractorError(u'Got a malformed response from YouTube API') | |
916 | playlist_title = response['feed']['title']['$t'] | |
917 | if 'entry' not in response['feed']: | |
918 | # Number of videos is a multiple of self._MAX_RESULTS | |
919 | break | |
920 | ||
921 | for entry in response['feed']['entry']: | |
922 | index = entry['yt$position']['$t'] | |
923 | if 'media$group' in entry and 'media$player' in entry['media$group']: | |
924 | videos.append((index, entry['media$group']['media$player']['url'])) | |
c5e8d7af PH |
925 | |
926 | videos = [v[1] for v in sorted(videos)] | |
927 | ||
20c3893f | 928 | url_results = [self.url_result(vurl, 'Youtube') for vurl in videos] |
c5e8d7af PH |
929 | return [self.playlist_result(url_results, playlist_id, playlist_title)] |
930 | ||
931 | ||
932 | class YoutubeChannelIE(InfoExtractor): | |
0f818663 | 933 | IE_DESC = u'YouTube.com channels' |
c5e8d7af PH |
934 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
935 | _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en' | |
936 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' | |
252580c5 | 937 | _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 |
938 | IE_NAME = u'youtube:channel' |
939 | ||
940 | def extract_videos_from_page(self, page): | |
941 | ids_in_page = [] | |
942 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
943 | if mobj.group(1) not in ids_in_page: | |
944 | ids_in_page.append(mobj.group(1)) | |
945 | return ids_in_page | |
946 | ||
947 | def _real_extract(self, url): | |
948 | # Extract channel id | |
949 | mobj = re.match(self._VALID_URL, url) | |
950 | if mobj is None: | |
951 | raise ExtractorError(u'Invalid URL: %s' % url) | |
952 | ||
953 | # Download channel page | |
954 | channel_id = mobj.group(1) | |
955 | video_ids = [] | |
956 | pagenum = 1 | |
957 | ||
958 | url = self._TEMPLATE_URL % (channel_id, pagenum) | |
959 | page = self._download_webpage(url, channel_id, | |
960 | u'Downloading page #%s' % pagenum) | |
961 | ||
962 | # Extract video identifiers | |
963 | ids_in_page = self.extract_videos_from_page(page) | |
964 | video_ids.extend(ids_in_page) | |
965 | ||
966 | # Download any subsequent channel pages using the json-based channel_ajax query | |
967 | if self._MORE_PAGES_INDICATOR in page: | |
755eb032 | 968 | for pagenum in itertools.count(1): |
c5e8d7af PH |
969 | url = self._MORE_PAGES_URL % (pagenum, channel_id) |
970 | page = self._download_webpage(url, channel_id, | |
971 | u'Downloading page #%s' % pagenum) | |
972 | ||
973 | page = json.loads(page) | |
974 | ||
975 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
976 | video_ids.extend(ids_in_page) | |
977 | ||
978 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
979 | break | |
980 | ||
981 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
982 | ||
983 | urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids] | |
20c3893f | 984 | url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls] |
c5e8d7af PH |
985 | return [self.playlist_result(url_entries, channel_id)] |
986 | ||
987 | ||
988 | class YoutubeUserIE(InfoExtractor): | |
0f818663 | 989 | IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' |
c5e8d7af PH |
990 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)' |
991 | _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' | |
992 | _GDATA_PAGE_SIZE = 50 | |
993 | _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d' | |
994 | _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]' | |
995 | IE_NAME = u'youtube:user' | |
996 | ||
997 | def _real_extract(self, url): | |
998 | # Extract username | |
999 | mobj = re.match(self._VALID_URL, url) | |
1000 | if mobj is None: | |
1001 | raise ExtractorError(u'Invalid URL: %s' % url) | |
1002 | ||
1003 | username = mobj.group(1) | |
1004 | ||
1005 | # Download video ids using YouTube Data API. Result size per | |
1006 | # query is limited (currently to 50 videos) so we need to query | |
1007 | # page by page until there are no video ids - it means we got | |
1008 | # all of them. | |
1009 | ||
1010 | video_ids = [] | |
c5e8d7af | 1011 | |
755eb032 | 1012 | for pagenum in itertools.count(0): |
c5e8d7af PH |
1013 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 |
1014 | ||
1015 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
1016 | page = self._download_webpage(gdata_url, username, | |
1017 | u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) | |
1018 | ||
1019 | # Extract video identifiers | |
1020 | ids_in_page = [] | |
1021 | ||
1022 | for mobj in re.finditer(self._VIDEO_INDICATOR, page): | |
1023 | if mobj.group(1) not in ids_in_page: | |
1024 | ids_in_page.append(mobj.group(1)) | |
1025 | ||
1026 | video_ids.extend(ids_in_page) | |
1027 | ||
1028 | # A little optimization - if current page is not | |
1029 | # "full", ie. does not contain PAGE_SIZE video ids then | |
1030 | # we can assume that this page is the last one - there | |
1031 | # are no more ids on further pages - no need to query | |
1032 | # again. | |
1033 | ||
1034 | if len(ids_in_page) < self._GDATA_PAGE_SIZE: | |
1035 | break | |
1036 | ||
c5e8d7af | 1037 | urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] |
20c3893f | 1038 | url_results = [self.url_result(rurl, 'Youtube') for rurl in urls] |
c5e8d7af | 1039 | return [self.playlist_result(url_results, playlist_title = username)] |
b05654f0 PH |
1040 | |
1041 | class YoutubeSearchIE(SearchInfoExtractor): | |
0f818663 | 1042 | IE_DESC = u'YouTube.com searches' |
b05654f0 PH |
1043 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' |
1044 | _MAX_RESULTS = 1000 | |
1045 | IE_NAME = u'youtube:search' | |
1046 | _SEARCH_KEY = 'ytsearch' | |
1047 | ||
1048 | def report_download_page(self, query, pagenum): | |
1049 | """Report attempt to download search page with given number.""" | |
1050 | self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) | |
1051 | ||
1052 | def _get_n_results(self, query, n): | |
1053 | """Get a specified number of results for a query""" | |
1054 | ||
1055 | video_ids = [] | |
1056 | pagenum = 0 | |
1057 | limit = n | |
1058 | ||
1059 | while (50 * pagenum) < limit: | |
1060 | self.report_download_page(query, pagenum+1) | |
1061 | result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) | |
1062 | request = compat_urllib_request.Request(result_url) | |
1063 | try: | |
1064 | data = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
1065 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
1066 | raise ExtractorError(u'Unable to download API page: %s' % compat_str(err)) | |
1067 | api_response = json.loads(data)['data'] | |
1068 | ||
1069 | if not 'items' in api_response: | |
1070 | raise ExtractorError(u'[youtube] No video results') | |
1071 | ||
1072 | new_ids = list(video['id'] for video in api_response['items']) | |
1073 | video_ids += new_ids | |
1074 | ||
1075 | limit = min(n, api_response['totalItems']) | |
1076 | pagenum += 1 | |
1077 | ||
1078 | if len(video_ids) > n: | |
1079 | video_ids = video_ids[:n] | |
1080 | videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids] | |
1081 | return self.playlist_result(videos, query) | |
75dff0ee JMF |
1082 | |
1083 | ||
1084 | class YoutubeShowIE(InfoExtractor): | |
0f818663 | 1085 | IE_DESC = u'YouTube.com (multi-season) shows' |
75dff0ee JMF |
1086 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
1087 | IE_NAME = u'youtube:show' | |
1088 | ||
1089 | def _real_extract(self, url): | |
1090 | mobj = re.match(self._VALID_URL, url) | |
1091 | show_name = mobj.group(1) | |
1092 | webpage = self._download_webpage(url, show_name, u'Downloading show webpage') | |
1093 | # There's one playlist for each season of the show | |
1094 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
1095 | self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) | |
1096 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] | |
04cc9617 JMF |
1097 | |
1098 | ||
b2e8bc1b | 1099 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): |
d7ae0639 JMF |
1100 | """ |
1101 | Base class for extractors that fetch info from | |
1102 | http://www.youtube.com/feed_ajax | |
1103 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
1104 | """ | |
b2e8bc1b | 1105 | _LOGIN_REQUIRED = True |
04cc9617 | 1106 | _PAGING_STEP = 30 |
43ba5456 JMF |
1107 | # use action_load_personal_feed instead of action_load_system_feed |
1108 | _PERSONAL_FEED = False | |
04cc9617 | 1109 | |
d7ae0639 JMF |
1110 | @property |
1111 | def _FEED_TEMPLATE(self): | |
43ba5456 JMF |
1112 | action = 'action_load_system_feed' |
1113 | if self._PERSONAL_FEED: | |
1114 | action = 'action_load_personal_feed' | |
1115 | return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) | |
d7ae0639 JMF |
1116 | |
1117 | @property | |
1118 | def IE_NAME(self): | |
1119 | return u'youtube:%s' % self._FEED_NAME | |
1120 | ||
81f0259b | 1121 | def _real_initialize(self): |
b2e8bc1b | 1122 | self._login() |
81f0259b | 1123 | |
04cc9617 JMF |
1124 | def _real_extract(self, url): |
1125 | feed_entries = [] | |
1126 | # The step argument is available only in 2.7 or higher | |
1127 | for i in itertools.count(0): | |
1128 | paging = i*self._PAGING_STEP | |
d7ae0639 JMF |
1129 | info = self._download_webpage(self._FEED_TEMPLATE % paging, |
1130 | u'%s feed' % self._FEED_NAME, | |
04cc9617 JMF |
1131 | u'Downloading page %s' % i) |
1132 | info = json.loads(info) | |
1133 | feed_html = info['feed_html'] | |
43ba5456 | 1134 | m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) |
04cc9617 JMF |
1135 | ids = orderedSet(m.group(1) for m in m_ids) |
1136 | feed_entries.extend(self.url_result(id, 'Youtube') for id in ids) | |
1137 | if info['paging'] is None: | |
1138 | break | |
d7ae0639 JMF |
1139 | return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) |
1140 | ||
1141 | class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): | |
1142 | IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' | |
1143 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
1144 | _FEED_NAME = 'subscriptions' | |
1145 | _PLAYLIST_TITLE = u'Youtube Subscriptions' | |
1146 | ||
1147 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): | |
1148 | IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' | |
1149 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' | |
1150 | _FEED_NAME = 'recommended' | |
1151 | _PLAYLIST_TITLE = u'Youtube Recommended videos' | |
c626a3d9 | 1152 | |
43ba5456 JMF |
1153 | class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): |
1154 | IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' | |
1155 | _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' | |
1156 | _FEED_NAME = 'watch_later' | |
1157 | _PLAYLIST_TITLE = u'Youtube Watch Later' | |
1158 | _PAGING_STEP = 100 | |
1159 | _PERSONAL_FEED = True | |
c626a3d9 JMF |
1160 | |
1161 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): | |
1162 | IE_NAME = u'youtube:favorites' | |
1163 | IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' | |
1164 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:o?rites)?' | |
1165 | _LOGIN_REQUIRED = True | |
1166 | ||
1167 | def _real_extract(self, url): | |
1168 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
1169 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id') | |
1170 | return self.url_result(playlist_id, 'YoutubePlaylist') |