]>
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 | ||
26 | ||
27 | class YoutubeIE(InfoExtractor): | |
0f818663 | 28 | IE_DESC = u'YouTube.com' |
c5e8d7af PH |
29 | _VALID_URL = r"""^ |
30 | ( | |
31 | (?:https?://)? # http(s):// (optional) | |
32 | (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/| | |
33 | tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains | |
34 | (?:.*?\#/)? # handle anchor (#/) redirect urls | |
35 | (?: # the various things that can precede the ID: | |
36 | (?:(?:v|embed|e)/) # v/ or embed/ or e/ | |
37 | |(?: # or the v= param in all its forms | |
bcd6e4bd | 38 | (?:watch|movie(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) |
c5e8d7af PH |
39 | (?:\?|\#!?) # the params delimiter ? or # or #! |
40 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
41 | v= | |
42 | ) | |
43 | )? # optional -> youtube.com/xxxx is OK | |
44 | )? # all until now is optional -> you can pass the naked ID | |
45 | ([0-9A-Za-z_-]+) # here is it! the YouTube video ID | |
46 | (?(1).+)? # if we found the ID, everything can follow | |
47 | $""" | |
48 | _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' | |
49 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
50 | _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' | |
51 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' | |
52 | _NETRC_MACHINE = 'youtube' | |
53 | # Listed in order of quality | |
54 | _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13'] | |
55 | _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13'] | |
56 | _video_extensions = { | |
57 | '13': '3gp', | |
58 | '17': 'mp4', | |
59 | '18': 'mp4', | |
60 | '22': 'mp4', | |
61 | '37': 'mp4', | |
d69cf69a | 62 | '38': 'mp4', |
c5e8d7af PH |
63 | '43': 'webm', |
64 | '44': 'webm', | |
65 | '45': 'webm', | |
66 | '46': 'webm', | |
67 | } | |
68 | _video_dimensions = { | |
69 | '5': '240x400', | |
70 | '6': '???', | |
71 | '13': '???', | |
72 | '17': '144x176', | |
73 | '18': '360x640', | |
74 | '22': '720x1280', | |
75 | '34': '360x640', | |
76 | '35': '480x854', | |
77 | '37': '1080x1920', | |
78 | '38': '3072x4096', | |
79 | '43': '360x640', | |
80 | '44': '480x854', | |
81 | '45': '720x1280', | |
82 | '46': '1080x1920', | |
83 | } | |
84 | IE_NAME = u'youtube' | |
2eb88d95 PH |
85 | _TESTS = [ |
86 | { | |
0e853ca4 PH |
87 | u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", |
88 | u"file": u"BaW_jenozKc.mp4", | |
89 | u"info_dict": { | |
90 | u"title": u"youtube-dl test video \"'/\\ä↭𝕐", | |
91 | u"uploader": u"Philipp Hagemeister", | |
92 | u"uploader_id": u"phihag", | |
93 | u"upload_date": u"20121002", | |
94 | 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 | 95 | } |
0e853ca4 PH |
96 | }, |
97 | { | |
98 | u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U", | |
99 | u"file": u"1ltcDfZMA3U.flv", | |
100 | u"note": u"Test VEVO video (#897)", | |
101 | u"info_dict": { | |
102 | u"upload_date": u"20070518", | |
103 | u"title": u"Maps - It Will Find You", | |
104 | u"description": u"Music video by Maps performing It Will Find You.", | |
105 | u"uploader": u"MuteUSA", | |
106 | u"uploader_id": u"MuteUSA" | |
2eb88d95 | 107 | } |
0e853ca4 PH |
108 | }, |
109 | { | |
110 | u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", | |
111 | u"file": u"UxxajLWwzqY.mp4", | |
112 | u"note": u"Test generic use_cipher_signature video (#897)", | |
113 | u"info_dict": { | |
114 | u"upload_date": u"20120506", | |
115 | u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", | |
116 | u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c", | |
117 | u"uploader": u"IconaPop", | |
118 | u"uploader_id": u"IconaPop" | |
2eb88d95 | 119 | } |
c108eb73 JMF |
120 | }, |
121 | { | |
122 | u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", | |
123 | u"file": u"07FYdnEawAQ.mp4", | |
124 | u"note": u"Test VEVO video with age protection (#956)", | |
125 | u"info_dict": { | |
126 | u"upload_date": u"20130703", | |
127 | u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", | |
128 | u"description": u"md5:64249768eec3bc4276236606ea996373", | |
129 | u"uploader": u"justintimberlakeVEVO", | |
130 | u"uploader_id": u"justintimberlakeVEVO" | |
131 | } | |
132 | }, | |
2eb88d95 PH |
133 | ] |
134 | ||
c5e8d7af PH |
135 | |
136 | @classmethod | |
137 | def suitable(cls, url): | |
138 | """Receives a URL and returns True if suitable for this IE.""" | |
04cc9617 | 139 | if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False |
c5e8d7af PH |
140 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None |
141 | ||
142 | def report_lang(self): | |
143 | """Report attempt to set language.""" | |
144 | self.to_screen(u'Setting language') | |
145 | ||
c5e8d7af PH |
146 | def report_video_webpage_download(self, video_id): |
147 | """Report attempt to download video webpage.""" | |
148 | self.to_screen(u'%s: Downloading video webpage' % video_id) | |
149 | ||
150 | def report_video_info_webpage_download(self, video_id): | |
151 | """Report attempt to download video info webpage.""" | |
152 | self.to_screen(u'%s: Downloading video info webpage' % video_id) | |
153 | ||
154 | def report_video_subtitles_download(self, video_id): | |
155 | """Report attempt to download video info webpage.""" | |
156 | self.to_screen(u'%s: Checking available subtitles' % video_id) | |
157 | ||
158 | def report_video_subtitles_request(self, video_id, sub_lang, format): | |
159 | """Report attempt to download video info webpage.""" | |
160 | self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format)) | |
161 | ||
162 | def report_video_subtitles_available(self, video_id, sub_lang_list): | |
163 | """Report available subtitles.""" | |
164 | sub_lang = ",".join(list(sub_lang_list.keys())) | |
165 | self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang)) | |
166 | ||
167 | def report_information_extraction(self, video_id): | |
168 | """Report attempt to extract video information.""" | |
169 | self.to_screen(u'%s: Extracting video information' % video_id) | |
170 | ||
171 | def report_unavailable_format(self, video_id, format): | |
172 | """Report extracted video URL.""" | |
173 | self.to_screen(u'%s: Format %s not available' % (video_id, format)) | |
174 | ||
175 | def report_rtmp_download(self): | |
176 | """Indicate the download will use the RTMP protocol.""" | |
177 | self.to_screen(u'RTMP download detected') | |
178 | ||
98bcd283 | 179 | def _decrypt_signature(self, s): |
257a2501 | 180 | """Turn the encrypted s field into a working signature""" |
6b37f0be | 181 | |
be547e1d PH |
182 | if len(s) == 88: |
183 | 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] | |
184 | elif len(s) == 87: | |
185 | return s[62] + s[82:62:-1] + s[83] + s[61:52:-1] + s[0] + s[51:2:-1] | |
186 | elif len(s) == 86: | |
187 | return s[2:63] + s[82] + s[64:82] + s[63] | |
188 | elif len(s) == 85: | |
189 | return s[76] + s[82:76:-1] + s[83] + s[75:60:-1] + s[0] + s[59:50:-1] + s[1] + s[49:2:-1] | |
190 | elif len(s) == 84: | |
191 | return s[83:36:-1] + s[2] + s[35:26:-1] + s[3] + s[25:3:-1] + s[26] | |
192 | elif len(s) == 83: | |
193 | return s[6] + s[3:6] + s[33] + s[7:24] + s[0] + s[25:33] + s[53] + s[34:53] + s[24] + s[54:] | |
194 | elif len(s) == 82: | |
195 | return s[36] + s[79:67:-1] + s[81] + s[66:40:-1] + s[33] + s[39:36:-1] + s[40] + s[35] + s[0] + s[67] + s[32:0:-1] + s[34] | |
196 | elif len(s) == 81: | |
197 | return s[6] + s[3:6] + s[33] + s[7:24] + s[0] + s[25:33] + s[2] + s[34:53] + s[24] + s[54:81] | |
88d4111c PH |
198 | elif len(s) == 92: |
199 | return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]; | |
be547e1d PH |
200 | |
201 | else: | |
202 | raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) | |
c5e8d7af PH |
203 | |
204 | def _get_available_subtitles(self, video_id): | |
205 | self.report_video_subtitles_download(video_id) | |
206 | request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id) | |
207 | try: | |
208 | sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
209 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
210 | return (u'unable to download video subtitles: %s' % compat_str(err), None) | |
211 | sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
212 | sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list) | |
213 | if not sub_lang_list: | |
214 | return (u'video doesn\'t have subtitles', None) | |
215 | return sub_lang_list | |
216 | ||
217 | def _list_available_subtitles(self, video_id): | |
218 | sub_lang_list = self._get_available_subtitles(video_id) | |
219 | self.report_video_subtitles_available(video_id, sub_lang_list) | |
220 | ||
221 | def _request_subtitle(self, sub_lang, sub_name, video_id, format): | |
222 | """ | |
223 | Return tuple: | |
224 | (error_message, sub_lang, sub) | |
225 | """ | |
226 | self.report_video_subtitles_request(video_id, sub_lang, format) | |
227 | params = compat_urllib_parse.urlencode({ | |
228 | 'lang': sub_lang, | |
229 | 'name': sub_name, | |
230 | 'v': video_id, | |
231 | 'fmt': format, | |
232 | }) | |
233 | url = 'http://www.youtube.com/api/timedtext?' + params | |
234 | try: | |
235 | sub = compat_urllib_request.urlopen(url).read().decode('utf-8') | |
236 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
237 | return (u'unable to download video subtitles: %s' % compat_str(err), None, None) | |
238 | if not sub: | |
239 | return (u'Did not fetch video subtitles', None, None) | |
240 | return (None, sub_lang, sub) | |
241 | ||
242 | def _request_automatic_caption(self, video_id, webpage): | |
243 | """We need the webpage for getting the captions url, pass it as an | |
244 | argument to speed up the process.""" | |
245 | sub_lang = self._downloader.params.get('subtitleslang') or 'en' | |
246 | sub_format = self._downloader.params.get('subtitlesformat') | |
247 | self.to_screen(u'%s: Looking for automatic captions' % video_id) | |
248 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
249 | err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang | |
250 | if mobj is None: | |
251 | return [(err_msg, None, None)] | |
252 | player_config = json.loads(mobj.group(1)) | |
253 | try: | |
254 | args = player_config[u'args'] | |
255 | caption_url = args[u'ttsurl'] | |
256 | timestamp = args[u'timestamp'] | |
257 | params = compat_urllib_parse.urlencode({ | |
258 | 'lang': 'en', | |
259 | 'tlang': sub_lang, | |
260 | 'fmt': sub_format, | |
261 | 'ts': timestamp, | |
262 | 'kind': 'asr', | |
263 | }) | |
264 | subtitles_url = caption_url + '&' + params | |
265 | sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions') | |
266 | return [(None, sub_lang, sub)] | |
267 | except KeyError: | |
268 | return [(err_msg, None, None)] | |
269 | ||
270 | def _extract_subtitle(self, video_id): | |
271 | """ | |
272 | Return a list with a tuple: | |
273 | [(error_message, sub_lang, sub)] | |
274 | """ | |
275 | sub_lang_list = self._get_available_subtitles(video_id) | |
276 | sub_format = self._downloader.params.get('subtitlesformat') | |
277 | if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles | |
278 | return [(sub_lang_list[0], None, None)] | |
279 | if self._downloader.params.get('subtitleslang', False): | |
280 | sub_lang = self._downloader.params.get('subtitleslang') | |
281 | elif 'en' in sub_lang_list: | |
282 | sub_lang = 'en' | |
283 | else: | |
284 | sub_lang = list(sub_lang_list.keys())[0] | |
285 | if not sub_lang in sub_lang_list: | |
286 | return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)] | |
287 | ||
288 | subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) | |
289 | return [subtitle] | |
290 | ||
291 | def _extract_all_subtitles(self, video_id): | |
292 | sub_lang_list = self._get_available_subtitles(video_id) | |
293 | sub_format = self._downloader.params.get('subtitlesformat') | |
294 | if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles | |
295 | return [(sub_lang_list[0], None, None)] | |
296 | subtitles = [] | |
297 | for sub_lang in sub_lang_list: | |
298 | subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) | |
299 | subtitles.append(subtitle) | |
300 | return subtitles | |
301 | ||
302 | def _print_formats(self, formats): | |
303 | print('Available formats:') | |
304 | for x in formats: | |
305 | print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???'))) | |
306 | ||
307 | def _real_initialize(self): | |
308 | if self._downloader is None: | |
309 | return | |
310 | ||
c5e8d7af PH |
311 | # Set language |
312 | request = compat_urllib_request.Request(self._LANG_URL) | |
313 | try: | |
314 | self.report_lang() | |
315 | compat_urllib_request.urlopen(request).read() | |
316 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
317 | self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) | |
318 | return | |
319 | ||
fefcb5d3 JMF |
320 | (username, password) = self._get_login_info() |
321 | ||
c5e8d7af PH |
322 | # No authentication to be performed |
323 | if username is None: | |
324 | return | |
325 | ||
326 | request = compat_urllib_request.Request(self._LOGIN_URL) | |
327 | try: | |
328 | login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
329 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
330 | self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) | |
331 | return | |
332 | ||
333 | galx = None | |
334 | dsh = None | |
335 | match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page) | |
336 | if match: | |
337 | galx = match.group(1) | |
338 | ||
339 | match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page) | |
340 | if match: | |
341 | dsh = match.group(1) | |
342 | ||
343 | # Log in | |
344 | login_form_strs = { | |
345 | u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
346 | u'Email': username, | |
347 | u'GALX': galx, | |
348 | u'Passwd': password, | |
349 | u'PersistentCookie': u'yes', | |
350 | u'_utf8': u'霱', | |
351 | u'bgresponse': u'js_disabled', | |
352 | u'checkConnection': u'', | |
353 | u'checkedDomains': u'youtube', | |
354 | u'dnConn': u'', | |
355 | u'dsh': dsh, | |
356 | u'pstMsg': u'0', | |
357 | u'rmShown': u'1', | |
358 | u'secTok': u'', | |
359 | u'signIn': u'Sign in', | |
360 | u'timeStmp': u'', | |
361 | u'service': u'youtube', | |
362 | u'uilel': u'3', | |
363 | u'hl': u'en_US', | |
364 | } | |
365 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
366 | # chokes on unicode | |
367 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) | |
368 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
369 | request = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
370 | try: | |
371 | self.report_login() | |
372 | login_results = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
373 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
374 | self._downloader.report_warning(u'unable to log in: bad username or password') | |
375 | return | |
376 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
377 | self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) | |
378 | return | |
379 | ||
380 | # Confirm age | |
381 | age_form = { | |
382 | 'next_url': '/', | |
383 | 'action_confirm': 'Confirm', | |
384 | } | |
385 | request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) | |
386 | try: | |
387 | self.report_age_confirmation() | |
93d3a642 | 388 | compat_urllib_request.urlopen(request).read().decode('utf-8') |
c5e8d7af PH |
389 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
390 | raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) | |
391 | ||
392 | def _extract_id(self, url): | |
393 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
394 | if mobj is None: | |
395 | raise ExtractorError(u'Invalid URL: %s' % url) | |
396 | video_id = mobj.group(2) | |
397 | return video_id | |
398 | ||
399 | def _real_extract(self, url): | |
d7f44b5b PH |
400 | if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url): |
401 | 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 ).') | |
402 | ||
c5e8d7af PH |
403 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter |
404 | mobj = re.search(self._NEXT_URL_RE, url) | |
405 | if mobj: | |
406 | url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
407 | video_id = self._extract_id(url) | |
408 | ||
409 | # Get video webpage | |
410 | self.report_video_webpage_download(video_id) | |
411 | url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id | |
412 | request = compat_urllib_request.Request(url) | |
413 | try: | |
414 | video_webpage_bytes = compat_urllib_request.urlopen(request).read() | |
415 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
416 | raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err)) | |
417 | ||
418 | video_webpage = video_webpage_bytes.decode('utf-8', 'ignore') | |
419 | ||
420 | # Attempt to extract SWF player URL | |
421 | mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) | |
422 | if mobj is not None: | |
423 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
424 | else: | |
425 | player_url = None | |
426 | ||
427 | # Get video info | |
428 | self.report_video_info_webpage_download(video_id) | |
c108eb73 JMF |
429 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
430 | self.report_age_confirmation() | |
431 | age_gate = True | |
432 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
433 | # this can be viewed without login into Youtube | |
434 | data = compat_urllib_parse.urlencode({'video_id': video_id, | |
435 | 'el': 'embedded', | |
436 | 'gl': 'US', | |
437 | 'hl': 'en', | |
438 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
439 | 'asv': 3, | |
440 | 'sts':'1588', | |
441 | }) | |
442 | video_info_url = 'https://www.youtube.com/get_video_info?' + data | |
c5e8d7af PH |
443 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
444 | note=False, | |
445 | errnote='unable to download video info webpage') | |
446 | video_info = compat_parse_qs(video_info_webpage) | |
c108eb73 JMF |
447 | else: |
448 | age_gate = False | |
449 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
450 | video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
451 | % (video_id, el_type)) | |
452 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
453 | note=False, | |
454 | errnote='unable to download video info webpage') | |
455 | video_info = compat_parse_qs(video_info_webpage) | |
456 | if 'token' in video_info: | |
457 | break | |
c5e8d7af PH |
458 | if 'token' not in video_info: |
459 | if 'reason' in video_info: | |
9a82b238 | 460 | raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True) |
c5e8d7af PH |
461 | else: |
462 | raise ExtractorError(u'"token" parameter not in video info for unknown reason') | |
463 | ||
464 | # Check for "rental" videos | |
465 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
466 | raise ExtractorError(u'"rental" videos not supported') | |
467 | ||
468 | # Start extracting information | |
469 | self.report_information_extraction(video_id) | |
470 | ||
471 | # uploader | |
472 | if 'author' not in video_info: | |
473 | raise ExtractorError(u'Unable to extract uploader name') | |
474 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
475 | ||
476 | # uploader_id | |
477 | video_uploader_id = None | |
478 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
479 | if mobj is not None: | |
480 | video_uploader_id = mobj.group(1) | |
481 | else: | |
482 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
483 | ||
484 | # title | |
485 | if 'title' not in video_info: | |
486 | raise ExtractorError(u'Unable to extract video title') | |
487 | video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) | |
488 | ||
489 | # thumbnail image | |
7763b04e JMF |
490 | # We try first to get a high quality image: |
491 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
492 | video_webpage, re.DOTALL) | |
493 | if m_thumb is not None: | |
494 | video_thumbnail = m_thumb.group(1) | |
495 | elif 'thumbnail_url' not in video_info: | |
c5e8d7af PH |
496 | self._downloader.report_warning(u'unable to extract video thumbnail') |
497 | video_thumbnail = '' | |
498 | else: # don't panic if we can't find it | |
499 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
500 | ||
501 | # upload date | |
502 | upload_date = None | |
503 | mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL) | |
504 | if mobj is not None: | |
505 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
506 | upload_date = unified_strdate(upload_date) | |
507 | ||
508 | # description | |
509 | video_description = get_element_by_id("eow-description", video_webpage) | |
510 | if video_description: | |
511 | video_description = clean_html(video_description) | |
512 | else: | |
513 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
514 | if fd_mobj: | |
515 | video_description = unescapeHTML(fd_mobj.group(1)) | |
516 | else: | |
517 | video_description = u'' | |
518 | ||
519 | # subtitles | |
520 | video_subtitles = None | |
521 | ||
522 | if self._downloader.params.get('writesubtitles', False): | |
523 | video_subtitles = self._extract_subtitle(video_id) | |
524 | if video_subtitles: | |
525 | (sub_error, sub_lang, sub) = video_subtitles[0] | |
526 | if sub_error: | |
b004821f JMF |
527 | self._downloader.report_warning(sub_error) |
528 | ||
529 | if self._downloader.params.get('writeautomaticsub', False): | |
530 | video_subtitles = self._request_automatic_caption(video_id, video_webpage) | |
531 | (sub_error, sub_lang, sub) = video_subtitles[0] | |
532 | if sub_error: | |
533 | self._downloader.report_warning(sub_error) | |
c5e8d7af PH |
534 | |
535 | if self._downloader.params.get('allsubtitles', False): | |
536 | video_subtitles = self._extract_all_subtitles(video_id) | |
537 | for video_subtitle in video_subtitles: | |
538 | (sub_error, sub_lang, sub) = video_subtitle | |
539 | if sub_error: | |
540 | self._downloader.report_warning(sub_error) | |
541 | ||
542 | if self._downloader.params.get('listsubtitles', False): | |
93d3a642 | 543 | self._list_available_subtitles(video_id) |
c5e8d7af PH |
544 | return |
545 | ||
546 | if 'length_seconds' not in video_info: | |
547 | self._downloader.report_warning(u'unable to extract video duration') | |
548 | video_duration = '' | |
549 | else: | |
550 | video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]) | |
551 | ||
c5e8d7af PH |
552 | # Decide which formats to download |
553 | req_format = self._downloader.params.get('format', None) | |
554 | ||
555 | try: | |
556 | mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) | |
50be92c1 PH |
557 | if not mobj: |
558 | raise ValueError('Could not find vevo ID') | |
c5e8d7af PH |
559 | info = json.loads(mobj.group(1)) |
560 | args = info['args'] | |
7ce7e394 JMF |
561 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
562 | # this signatures are encrypted | |
563 | m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map']) | |
564 | if m_s is not None: | |
565 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
c5e8d7af PH |
566 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
567 | except ValueError: | |
568 | pass | |
569 | ||
570 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
571 | self.report_rtmp_download() | |
572 | video_url_list = [(None, video_info['conn'][0])] | |
573 | elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1: | |
a7055eb9 JMF |
574 | if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]: |
575 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) | |
c5e8d7af PH |
576 | url_map = {} |
577 | for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): | |
578 | url_data = compat_parse_qs(url_data_str) | |
579 | if 'itag' in url_data and 'url' in url_data: | |
580 | url = url_data['url'][0] | |
581 | if 'sig' in url_data: | |
582 | url += '&signature=' + url_data['sig'][0] | |
583 | elif 's' in url_data: | |
769fda3c FV |
584 | if self._downloader.params.get('verbose'): |
585 | s = url_data['s'][0] | |
c108eb73 JMF |
586 | if age_gate: |
587 | player_version = self._search_regex(r'ad3-(.+?)\.swf', | |
588 | video_info['ad3_module'][0], 'flash player', | |
589 | fatal=False) | |
590 | player = 'flash player %s' % player_version | |
591 | else: | |
592 | player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage, | |
593 | 'html5 player', fatal=False) | |
594 | self.to_screen('encrypted signature length %d (%d.%d), itag %s, %s' % | |
769fda3c | 595 | (len(s), len(s.split('.')[0]), len(s.split('.')[1]), url_data['itag'][0], player)) |
c5e8d7af PH |
596 | signature = self._decrypt_signature(url_data['s'][0]) |
597 | url += '&signature=' + signature | |
598 | if 'ratebypass' not in url: | |
599 | url += '&ratebypass=yes' | |
600 | url_map[url_data['itag'][0]] = url | |
601 | ||
602 | format_limit = self._downloader.params.get('format_limit', None) | |
603 | available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats | |
604 | if format_limit is not None and format_limit in available_formats: | |
605 | format_list = available_formats[available_formats.index(format_limit):] | |
606 | else: | |
607 | format_list = available_formats | |
608 | existing_formats = [x for x in format_list if x in url_map] | |
609 | if len(existing_formats) == 0: | |
610 | raise ExtractorError(u'no known formats available for video') | |
611 | if self._downloader.params.get('listformats', None): | |
612 | self._print_formats(existing_formats) | |
613 | return | |
614 | if req_format is None or req_format == 'best': | |
615 | video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality | |
616 | elif req_format == 'worst': | |
d828f3a5 | 617 | video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality |
c5e8d7af PH |
618 | elif req_format in ('-1', 'all'): |
619 | video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats | |
620 | else: | |
621 | # Specific formats. We pick the first in a slash-delimeted sequence. | |
622 | # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. | |
623 | req_formats = req_format.split('/') | |
624 | video_url_list = None | |
625 | for rf in req_formats: | |
626 | if rf in url_map: | |
627 | video_url_list = [(rf, url_map[rf])] | |
628 | break | |
629 | if video_url_list is None: | |
630 | raise ExtractorError(u'requested format not available') | |
631 | else: | |
632 | raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info') | |
633 | ||
634 | results = [] | |
635 | for format_param, video_real_url in video_url_list: | |
636 | # Extension | |
637 | video_extension = self._video_extensions.get(format_param, 'flv') | |
638 | ||
639 | video_format = '{0} - {1}'.format(format_param if format_param else video_extension, | |
640 | self._video_dimensions.get(format_param, '???')) | |
641 | ||
642 | results.append({ | |
643 | 'id': video_id, | |
644 | 'url': video_real_url, | |
645 | 'uploader': video_uploader, | |
646 | 'uploader_id': video_uploader_id, | |
647 | 'upload_date': upload_date, | |
648 | 'title': video_title, | |
649 | 'ext': video_extension, | |
650 | 'format': video_format, | |
651 | 'thumbnail': video_thumbnail, | |
652 | 'description': video_description, | |
653 | 'player_url': player_url, | |
654 | 'subtitles': video_subtitles, | |
655 | 'duration': video_duration | |
656 | }) | |
657 | return results | |
658 | ||
659 | class YoutubePlaylistIE(InfoExtractor): | |
0f818663 | 660 | IE_DESC = u'YouTube.com playlists' |
c5e8d7af PH |
661 | _VALID_URL = r"""(?: |
662 | (?:https?://)? | |
663 | (?:\w+\.)? | |
664 | youtube\.com/ | |
665 | (?: | |
666 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
667 | \? (?:.*?&)*? (?:p|a|list)= | |
668 | | p/ | |
669 | ) | |
670 | ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,}) | |
671 | .* | |
672 | | | |
673 | ((?:PL|EC|UU)[0-9A-Za-z-_]{10,}) | |
674 | )""" | |
675 | _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none' | |
676 | _MAX_RESULTS = 50 | |
677 | IE_NAME = u'youtube:playlist' | |
678 | ||
679 | @classmethod | |
680 | def suitable(cls, url): | |
681 | """Receives a URL and returns True if suitable for this IE.""" | |
682 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
683 | ||
684 | def _real_extract(self, url): | |
685 | # Extract playlist id | |
686 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
687 | if mobj is None: | |
688 | raise ExtractorError(u'Invalid URL: %s' % url) | |
689 | ||
690 | # Download playlist videos from API | |
691 | playlist_id = mobj.group(1) or mobj.group(2) | |
692 | page_num = 1 | |
693 | videos = [] | |
694 | ||
695 | while True: | |
696 | url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1) | |
697 | page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) | |
698 | ||
699 | try: | |
700 | response = json.loads(page) | |
701 | except ValueError as err: | |
702 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
703 | ||
704 | if 'feed' not in response: | |
705 | raise ExtractorError(u'Got a malformed response from YouTube API') | |
706 | playlist_title = response['feed']['title']['$t'] | |
707 | if 'entry' not in response['feed']: | |
708 | # Number of videos is a multiple of self._MAX_RESULTS | |
709 | break | |
710 | ||
711 | for entry in response['feed']['entry']: | |
712 | index = entry['yt$position']['$t'] | |
713 | if 'media$group' in entry and 'media$player' in entry['media$group']: | |
714 | videos.append((index, entry['media$group']['media$player']['url'])) | |
715 | ||
716 | if len(response['feed']['entry']) < self._MAX_RESULTS: | |
717 | break | |
718 | page_num += 1 | |
719 | ||
720 | videos = [v[1] for v in sorted(videos)] | |
721 | ||
20c3893f | 722 | url_results = [self.url_result(vurl, 'Youtube') for vurl in videos] |
c5e8d7af PH |
723 | return [self.playlist_result(url_results, playlist_id, playlist_title)] |
724 | ||
725 | ||
726 | class YoutubeChannelIE(InfoExtractor): | |
0f818663 | 727 | IE_DESC = u'YouTube.com channels' |
c5e8d7af PH |
728 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
729 | _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en' | |
730 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' | |
731 | _MORE_PAGES_URL = 'http://www.youtube.com/channel_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' | |
732 | IE_NAME = u'youtube:channel' | |
733 | ||
734 | def extract_videos_from_page(self, page): | |
735 | ids_in_page = [] | |
736 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
737 | if mobj.group(1) not in ids_in_page: | |
738 | ids_in_page.append(mobj.group(1)) | |
739 | return ids_in_page | |
740 | ||
741 | def _real_extract(self, url): | |
742 | # Extract channel id | |
743 | mobj = re.match(self._VALID_URL, url) | |
744 | if mobj is None: | |
745 | raise ExtractorError(u'Invalid URL: %s' % url) | |
746 | ||
747 | # Download channel page | |
748 | channel_id = mobj.group(1) | |
749 | video_ids = [] | |
750 | pagenum = 1 | |
751 | ||
752 | url = self._TEMPLATE_URL % (channel_id, pagenum) | |
753 | page = self._download_webpage(url, channel_id, | |
754 | u'Downloading page #%s' % pagenum) | |
755 | ||
756 | # Extract video identifiers | |
757 | ids_in_page = self.extract_videos_from_page(page) | |
758 | video_ids.extend(ids_in_page) | |
759 | ||
760 | # Download any subsequent channel pages using the json-based channel_ajax query | |
761 | if self._MORE_PAGES_INDICATOR in page: | |
762 | while True: | |
763 | pagenum = pagenum + 1 | |
764 | ||
765 | url = self._MORE_PAGES_URL % (pagenum, channel_id) | |
766 | page = self._download_webpage(url, channel_id, | |
767 | u'Downloading page #%s' % pagenum) | |
768 | ||
769 | page = json.loads(page) | |
770 | ||
771 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
772 | video_ids.extend(ids_in_page) | |
773 | ||
774 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
775 | break | |
776 | ||
777 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
778 | ||
779 | urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids] | |
20c3893f | 780 | url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls] |
c5e8d7af PH |
781 | return [self.playlist_result(url_entries, channel_id)] |
782 | ||
783 | ||
784 | class YoutubeUserIE(InfoExtractor): | |
0f818663 | 785 | IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' |
c5e8d7af PH |
786 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)' |
787 | _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' | |
788 | _GDATA_PAGE_SIZE = 50 | |
789 | _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d' | |
790 | _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]' | |
791 | IE_NAME = u'youtube:user' | |
792 | ||
793 | def _real_extract(self, url): | |
794 | # Extract username | |
795 | mobj = re.match(self._VALID_URL, url) | |
796 | if mobj is None: | |
797 | raise ExtractorError(u'Invalid URL: %s' % url) | |
798 | ||
799 | username = mobj.group(1) | |
800 | ||
801 | # Download video ids using YouTube Data API. Result size per | |
802 | # query is limited (currently to 50 videos) so we need to query | |
803 | # page by page until there are no video ids - it means we got | |
804 | # all of them. | |
805 | ||
806 | video_ids = [] | |
807 | pagenum = 0 | |
808 | ||
809 | while True: | |
810 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 | |
811 | ||
812 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
813 | page = self._download_webpage(gdata_url, username, | |
814 | u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) | |
815 | ||
816 | # Extract video identifiers | |
817 | ids_in_page = [] | |
818 | ||
819 | for mobj in re.finditer(self._VIDEO_INDICATOR, page): | |
820 | if mobj.group(1) not in ids_in_page: | |
821 | ids_in_page.append(mobj.group(1)) | |
822 | ||
823 | video_ids.extend(ids_in_page) | |
824 | ||
825 | # A little optimization - if current page is not | |
826 | # "full", ie. does not contain PAGE_SIZE video ids then | |
827 | # we can assume that this page is the last one - there | |
828 | # are no more ids on further pages - no need to query | |
829 | # again. | |
830 | ||
831 | if len(ids_in_page) < self._GDATA_PAGE_SIZE: | |
832 | break | |
833 | ||
834 | pagenum += 1 | |
835 | ||
836 | urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] | |
20c3893f | 837 | url_results = [self.url_result(rurl, 'Youtube') for rurl in urls] |
c5e8d7af | 838 | return [self.playlist_result(url_results, playlist_title = username)] |
b05654f0 PH |
839 | |
840 | class YoutubeSearchIE(SearchInfoExtractor): | |
0f818663 | 841 | IE_DESC = u'YouTube.com searches' |
b05654f0 PH |
842 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' |
843 | _MAX_RESULTS = 1000 | |
844 | IE_NAME = u'youtube:search' | |
845 | _SEARCH_KEY = 'ytsearch' | |
846 | ||
847 | def report_download_page(self, query, pagenum): | |
848 | """Report attempt to download search page with given number.""" | |
849 | self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) | |
850 | ||
851 | def _get_n_results(self, query, n): | |
852 | """Get a specified number of results for a query""" | |
853 | ||
854 | video_ids = [] | |
855 | pagenum = 0 | |
856 | limit = n | |
857 | ||
858 | while (50 * pagenum) < limit: | |
859 | self.report_download_page(query, pagenum+1) | |
860 | result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) | |
861 | request = compat_urllib_request.Request(result_url) | |
862 | try: | |
863 | data = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
864 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
865 | raise ExtractorError(u'Unable to download API page: %s' % compat_str(err)) | |
866 | api_response = json.loads(data)['data'] | |
867 | ||
868 | if not 'items' in api_response: | |
869 | raise ExtractorError(u'[youtube] No video results') | |
870 | ||
871 | new_ids = list(video['id'] for video in api_response['items']) | |
872 | video_ids += new_ids | |
873 | ||
874 | limit = min(n, api_response['totalItems']) | |
875 | pagenum += 1 | |
876 | ||
877 | if len(video_ids) > n: | |
878 | video_ids = video_ids[:n] | |
879 | videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids] | |
880 | return self.playlist_result(videos, query) | |
75dff0ee JMF |
881 | |
882 | ||
883 | class YoutubeShowIE(InfoExtractor): | |
0f818663 | 884 | IE_DESC = u'YouTube.com (multi-season) shows' |
75dff0ee JMF |
885 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
886 | IE_NAME = u'youtube:show' | |
887 | ||
888 | def _real_extract(self, url): | |
889 | mobj = re.match(self._VALID_URL, url) | |
890 | show_name = mobj.group(1) | |
891 | webpage = self._download_webpage(url, show_name, u'Downloading show webpage') | |
892 | # There's one playlist for each season of the show | |
893 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
894 | self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) | |
895 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] | |
04cc9617 JMF |
896 | |
897 | ||
898 | class YoutubeSubscriptionsIE(YoutubeIE): | |
899 | """It's a subclass of YoutubeIE because we need to login""" | |
897f36d1 PH |
900 | IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' |
901 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
04cc9617 JMF |
902 | IE_NAME = u'youtube:subscriptions' |
903 | _FEED_TEMPLATE = 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=subscriptions&paging=%s' | |
904 | _PAGING_STEP = 30 | |
905 | ||
897f36d1 | 906 | # Overwrite YoutubeIE properties we don't want |
04cc9617 | 907 | _TESTS = [] |
04cc9617 JMF |
908 | @classmethod |
909 | def suitable(cls, url): | |
910 | return re.match(cls._VALID_URL, url) is not None | |
911 | ||
81f0259b JMF |
912 | def _real_initialize(self): |
913 | (username, password) = self._get_login_info() | |
914 | if username is None: | |
915 | raise ExtractorError(u'No login info available, needed for downloading the Youtube subscriptions.', expected=True) | |
916 | super(YoutubeSubscriptionsIE, self)._real_initialize() | |
917 | ||
04cc9617 JMF |
918 | def _real_extract(self, url): |
919 | feed_entries = [] | |
920 | # The step argument is available only in 2.7 or higher | |
921 | for i in itertools.count(0): | |
922 | paging = i*self._PAGING_STEP | |
923 | info = self._download_webpage(self._FEED_TEMPLATE % paging, 'feed', | |
924 | u'Downloading page %s' % i) | |
925 | info = json.loads(info) | |
926 | feed_html = info['feed_html'] | |
927 | m_ids = re.finditer(r'"/watch\?v=(.*?)"', feed_html) | |
928 | ids = orderedSet(m.group(1) for m in m_ids) | |
929 | feed_entries.extend(self.url_result(id, 'Youtube') for id in ids) | |
930 | if info['paging'] is None: | |
931 | break | |
932 | return self.playlist_result(feed_entries, playlist_title='Youtube Subscriptions') |