]>
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 | |
444b1165 JMF |
182 | if len(s) == 92: |
183 | return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83] | |
184 | elif len(s) == 90: | |
185 | return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81] | |
186 | elif len(s) == 88: | |
be547e1d PH |
187 | 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] |
188 | elif len(s) == 87: | |
189 | return s[62] + s[82:62:-1] + s[83] + s[61:52:-1] + s[0] + s[51:2:-1] | |
190 | elif len(s) == 86: | |
191 | return s[2:63] + s[82] + s[64:82] + s[63] | |
192 | elif len(s) == 85: | |
a4358cba | 193 | return s[2:8] + s[0] + s[9:21] + s[65] + s[22:65] + s[84] + s[66:82] + s[21] |
be547e1d PH |
194 | elif len(s) == 84: |
195 | return s[83:36:-1] + s[2] + s[35:26:-1] + s[3] + s[25:3:-1] + s[26] | |
196 | elif len(s) == 83: | |
197 | 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:] | |
198 | elif len(s) == 82: | |
199 | 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] | |
200 | elif len(s) == 81: | |
201 | 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] | |
202 | ||
203 | else: | |
204 | raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) | |
c5e8d7af PH |
205 | |
206 | def _get_available_subtitles(self, video_id): | |
207 | self.report_video_subtitles_download(video_id) | |
208 | request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id) | |
209 | try: | |
210 | sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
211 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
5d51a883 JMF |
212 | self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) |
213 | return {} | |
c5e8d7af PH |
214 | sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) |
215 | sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list) | |
216 | if not sub_lang_list: | |
5d51a883 JMF |
217 | self._downloader.report_warning(u'video doesn\'t have subtitles') |
218 | return {} | |
c5e8d7af PH |
219 | return sub_lang_list |
220 | ||
221 | def _list_available_subtitles(self, video_id): | |
222 | sub_lang_list = self._get_available_subtitles(video_id) | |
223 | self.report_video_subtitles_available(video_id, sub_lang_list) | |
224 | ||
225 | def _request_subtitle(self, sub_lang, sub_name, video_id, format): | |
226 | """ | |
5d51a883 | 227 | Return the subtitle as a string or None if they are not found |
c5e8d7af PH |
228 | """ |
229 | self.report_video_subtitles_request(video_id, sub_lang, format) | |
230 | params = compat_urllib_parse.urlencode({ | |
231 | 'lang': sub_lang, | |
232 | 'name': sub_name, | |
233 | 'v': video_id, | |
234 | 'fmt': format, | |
235 | }) | |
236 | url = 'http://www.youtube.com/api/timedtext?' + params | |
237 | try: | |
238 | sub = compat_urllib_request.urlopen(url).read().decode('utf-8') | |
239 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
5d51a883 JMF |
240 | self._downloader.report_warning(u'unable to download video subtitles for %s: %s' % (sub_lang, compat_str(err))) |
241 | return | |
c5e8d7af | 242 | if not sub: |
5d51a883 JMF |
243 | self._downloader.report_warning(u'Did not fetch video subtitles') |
244 | return | |
245 | return sub | |
c5e8d7af PH |
246 | |
247 | def _request_automatic_caption(self, video_id, webpage): | |
248 | """We need the webpage for getting the captions url, pass it as an | |
249 | argument to speed up the process.""" | |
250 | sub_lang = self._downloader.params.get('subtitleslang') or 'en' | |
251 | sub_format = self._downloader.params.get('subtitlesformat') | |
252 | self.to_screen(u'%s: Looking for automatic captions' % video_id) | |
253 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
254 | err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang | |
255 | if mobj is None: | |
5d51a883 JMF |
256 | self._downloader.report_warning(err_msg) |
257 | return {} | |
c5e8d7af PH |
258 | player_config = json.loads(mobj.group(1)) |
259 | try: | |
260 | args = player_config[u'args'] | |
261 | caption_url = args[u'ttsurl'] | |
262 | timestamp = args[u'timestamp'] | |
263 | params = compat_urllib_parse.urlencode({ | |
264 | 'lang': 'en', | |
265 | 'tlang': sub_lang, | |
266 | 'fmt': sub_format, | |
267 | 'ts': timestamp, | |
268 | 'kind': 'asr', | |
269 | }) | |
270 | subtitles_url = caption_url + '&' + params | |
271 | sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions') | |
5d51a883 | 272 | return {sub_lang: sub} |
c5e8d7af | 273 | except KeyError: |
5d51a883 JMF |
274 | self._downloader.report_warning(err_msg) |
275 | return {} | |
88ae5991 JMF |
276 | |
277 | def _extract_subtitles(self, video_id): | |
c5e8d7af | 278 | """ |
5d51a883 JMF |
279 | Return a dictionary: {language: subtitles} or {} if the subtitles |
280 | couldn't be found | |
c5e8d7af PH |
281 | """ |
282 | sub_lang_list = self._get_available_subtitles(video_id) | |
283 | sub_format = self._downloader.params.get('subtitlesformat') | |
5d51a883 JMF |
284 | if not sub_lang_list: #There was some error, it didn't get the available subtitles |
285 | return {} | |
88ae5991 JMF |
286 | if self._downloader.params.get('writesubtitles', False): |
287 | if self._downloader.params.get('subtitleslang', False): | |
288 | sub_lang = self._downloader.params.get('subtitleslang') | |
289 | elif 'en' in sub_lang_list: | |
290 | sub_lang = 'en' | |
291 | else: | |
292 | sub_lang = list(sub_lang_list.keys())[0] | |
293 | if not sub_lang in sub_lang_list: | |
294 | self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang) | |
295 | return {} | |
296 | sub_lang_list = {sub_lang: sub_lang_list[sub_lang]} | |
5d51a883 | 297 | subtitles = {} |
c5e8d7af PH |
298 | for sub_lang in sub_lang_list: |
299 | subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) | |
5d51a883 JMF |
300 | if subtitle: |
301 | subtitles[sub_lang] = subtitle | |
c5e8d7af PH |
302 | return subtitles |
303 | ||
304 | def _print_formats(self, formats): | |
305 | print('Available formats:') | |
306 | for x in formats: | |
307 | print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???'))) | |
308 | ||
309 | def _real_initialize(self): | |
310 | if self._downloader is None: | |
311 | return | |
312 | ||
c5e8d7af PH |
313 | # Set language |
314 | request = compat_urllib_request.Request(self._LANG_URL) | |
315 | try: | |
316 | self.report_lang() | |
317 | compat_urllib_request.urlopen(request).read() | |
318 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
319 | self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) | |
320 | return | |
321 | ||
fefcb5d3 JMF |
322 | (username, password) = self._get_login_info() |
323 | ||
c5e8d7af PH |
324 | # No authentication to be performed |
325 | if username is None: | |
326 | return | |
327 | ||
328 | request = compat_urllib_request.Request(self._LOGIN_URL) | |
329 | try: | |
330 | login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
331 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
332 | self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) | |
333 | return | |
334 | ||
335 | galx = None | |
336 | dsh = None | |
337 | match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page) | |
338 | if match: | |
339 | galx = match.group(1) | |
340 | ||
341 | match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page) | |
342 | if match: | |
343 | dsh = match.group(1) | |
344 | ||
345 | # Log in | |
346 | login_form_strs = { | |
347 | u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
348 | u'Email': username, | |
349 | u'GALX': galx, | |
350 | u'Passwd': password, | |
351 | u'PersistentCookie': u'yes', | |
352 | u'_utf8': u'霱', | |
353 | u'bgresponse': u'js_disabled', | |
354 | u'checkConnection': u'', | |
355 | u'checkedDomains': u'youtube', | |
356 | u'dnConn': u'', | |
357 | u'dsh': dsh, | |
358 | u'pstMsg': u'0', | |
359 | u'rmShown': u'1', | |
360 | u'secTok': u'', | |
361 | u'signIn': u'Sign in', | |
362 | u'timeStmp': u'', | |
363 | u'service': u'youtube', | |
364 | u'uilel': u'3', | |
365 | u'hl': u'en_US', | |
366 | } | |
367 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
368 | # chokes on unicode | |
369 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) | |
370 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
371 | request = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
372 | try: | |
373 | self.report_login() | |
374 | login_results = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
375 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
376 | self._downloader.report_warning(u'unable to log in: bad username or password') | |
377 | return | |
378 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
379 | self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) | |
380 | return | |
381 | ||
382 | # Confirm age | |
383 | age_form = { | |
384 | 'next_url': '/', | |
385 | 'action_confirm': 'Confirm', | |
386 | } | |
387 | request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) | |
388 | try: | |
389 | self.report_age_confirmation() | |
93d3a642 | 390 | compat_urllib_request.urlopen(request).read().decode('utf-8') |
c5e8d7af PH |
391 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
392 | raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) | |
393 | ||
394 | def _extract_id(self, url): | |
395 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
396 | if mobj is None: | |
397 | raise ExtractorError(u'Invalid URL: %s' % url) | |
398 | video_id = mobj.group(2) | |
399 | return video_id | |
400 | ||
401 | def _real_extract(self, url): | |
d7f44b5b PH |
402 | if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url): |
403 | 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 ).') | |
404 | ||
c5e8d7af PH |
405 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter |
406 | mobj = re.search(self._NEXT_URL_RE, url) | |
407 | if mobj: | |
408 | url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
409 | video_id = self._extract_id(url) | |
410 | ||
411 | # Get video webpage | |
412 | self.report_video_webpage_download(video_id) | |
413 | url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id | |
414 | request = compat_urllib_request.Request(url) | |
415 | try: | |
416 | video_webpage_bytes = compat_urllib_request.urlopen(request).read() | |
417 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
418 | raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err)) | |
419 | ||
420 | video_webpage = video_webpage_bytes.decode('utf-8', 'ignore') | |
421 | ||
422 | # Attempt to extract SWF player URL | |
423 | mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) | |
424 | if mobj is not None: | |
425 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
426 | else: | |
427 | player_url = None | |
428 | ||
429 | # Get video info | |
430 | self.report_video_info_webpage_download(video_id) | |
c108eb73 JMF |
431 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
432 | self.report_age_confirmation() | |
433 | age_gate = True | |
434 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
435 | # this can be viewed without login into Youtube | |
436 | data = compat_urllib_parse.urlencode({'video_id': video_id, | |
437 | 'el': 'embedded', | |
438 | 'gl': 'US', | |
439 | 'hl': 'en', | |
440 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
441 | 'asv': 3, | |
442 | 'sts':'1588', | |
443 | }) | |
444 | video_info_url = 'https://www.youtube.com/get_video_info?' + data | |
c5e8d7af PH |
445 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
446 | note=False, | |
447 | errnote='unable to download video info webpage') | |
448 | video_info = compat_parse_qs(video_info_webpage) | |
c108eb73 JMF |
449 | else: |
450 | age_gate = False | |
451 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
452 | video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
453 | % (video_id, el_type)) | |
454 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
455 | note=False, | |
456 | errnote='unable to download video info webpage') | |
457 | video_info = compat_parse_qs(video_info_webpage) | |
458 | if 'token' in video_info: | |
459 | break | |
c5e8d7af PH |
460 | if 'token' not in video_info: |
461 | if 'reason' in video_info: | |
9a82b238 | 462 | raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True) |
c5e8d7af PH |
463 | else: |
464 | raise ExtractorError(u'"token" parameter not in video info for unknown reason') | |
465 | ||
466 | # Check for "rental" videos | |
467 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
468 | raise ExtractorError(u'"rental" videos not supported') | |
469 | ||
470 | # Start extracting information | |
471 | self.report_information_extraction(video_id) | |
472 | ||
473 | # uploader | |
474 | if 'author' not in video_info: | |
475 | raise ExtractorError(u'Unable to extract uploader name') | |
476 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
477 | ||
478 | # uploader_id | |
479 | video_uploader_id = None | |
480 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
481 | if mobj is not None: | |
482 | video_uploader_id = mobj.group(1) | |
483 | else: | |
484 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
485 | ||
486 | # title | |
487 | if 'title' not in video_info: | |
488 | raise ExtractorError(u'Unable to extract video title') | |
489 | video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) | |
490 | ||
491 | # thumbnail image | |
7763b04e JMF |
492 | # We try first to get a high quality image: |
493 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
494 | video_webpage, re.DOTALL) | |
495 | if m_thumb is not None: | |
496 | video_thumbnail = m_thumb.group(1) | |
497 | elif 'thumbnail_url' not in video_info: | |
c5e8d7af PH |
498 | self._downloader.report_warning(u'unable to extract video thumbnail') |
499 | video_thumbnail = '' | |
500 | else: # don't panic if we can't find it | |
501 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
502 | ||
503 | # upload date | |
504 | upload_date = None | |
505 | mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL) | |
506 | if mobj is not None: | |
507 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
508 | upload_date = unified_strdate(upload_date) | |
509 | ||
510 | # description | |
511 | video_description = get_element_by_id("eow-description", video_webpage) | |
512 | if video_description: | |
513 | video_description = clean_html(video_description) | |
514 | else: | |
515 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
516 | if fd_mobj: | |
517 | video_description = unescapeHTML(fd_mobj.group(1)) | |
518 | else: | |
519 | video_description = u'' | |
520 | ||
521 | # subtitles | |
522 | video_subtitles = None | |
523 | ||
88ae5991 JMF |
524 | if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False): |
525 | video_subtitles = self._extract_subtitles(video_id) | |
5d51a883 | 526 | elif self._downloader.params.get('writeautomaticsub', False): |
b004821f | 527 | video_subtitles = self._request_automatic_caption(video_id, video_webpage) |
c5e8d7af | 528 | |
c5e8d7af | 529 | if self._downloader.params.get('listsubtitles', False): |
93d3a642 | 530 | self._list_available_subtitles(video_id) |
c5e8d7af PH |
531 | return |
532 | ||
533 | if 'length_seconds' not in video_info: | |
534 | self._downloader.report_warning(u'unable to extract video duration') | |
535 | video_duration = '' | |
536 | else: | |
537 | video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]) | |
538 | ||
c5e8d7af PH |
539 | # Decide which formats to download |
540 | req_format = self._downloader.params.get('format', None) | |
541 | ||
542 | try: | |
543 | mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) | |
50be92c1 PH |
544 | if not mobj: |
545 | raise ValueError('Could not find vevo ID') | |
c5e8d7af PH |
546 | info = json.loads(mobj.group(1)) |
547 | args = info['args'] | |
7ce7e394 JMF |
548 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
549 | # this signatures are encrypted | |
550 | m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map']) | |
551 | if m_s is not None: | |
552 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
c5e8d7af PH |
553 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
554 | except ValueError: | |
555 | pass | |
556 | ||
557 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
558 | self.report_rtmp_download() | |
559 | video_url_list = [(None, video_info['conn'][0])] | |
560 | elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1: | |
a7055eb9 JMF |
561 | if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]: |
562 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) | |
c5e8d7af PH |
563 | url_map = {} |
564 | for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): | |
565 | url_data = compat_parse_qs(url_data_str) | |
566 | if 'itag' in url_data and 'url' in url_data: | |
567 | url = url_data['url'][0] | |
568 | if 'sig' in url_data: | |
569 | url += '&signature=' + url_data['sig'][0] | |
570 | elif 's' in url_data: | |
769fda3c FV |
571 | if self._downloader.params.get('verbose'): |
572 | s = url_data['s'][0] | |
c108eb73 JMF |
573 | if age_gate: |
574 | player_version = self._search_regex(r'ad3-(.+?)\.swf', | |
575 | video_info['ad3_module'][0], 'flash player', | |
576 | fatal=False) | |
577 | player = 'flash player %s' % player_version | |
578 | else: | |
579 | player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage, | |
580 | 'html5 player', fatal=False) | |
5a76c651 JMF |
581 | parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.')) |
582 | self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % | |
583 | (len(s), parts_sizes, url_data['itag'][0], player)) | |
c5e8d7af PH |
584 | signature = self._decrypt_signature(url_data['s'][0]) |
585 | url += '&signature=' + signature | |
586 | if 'ratebypass' not in url: | |
587 | url += '&ratebypass=yes' | |
588 | url_map[url_data['itag'][0]] = url | |
589 | ||
590 | format_limit = self._downloader.params.get('format_limit', None) | |
591 | available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats | |
592 | if format_limit is not None and format_limit in available_formats: | |
593 | format_list = available_formats[available_formats.index(format_limit):] | |
594 | else: | |
595 | format_list = available_formats | |
596 | existing_formats = [x for x in format_list if x in url_map] | |
597 | if len(existing_formats) == 0: | |
598 | raise ExtractorError(u'no known formats available for video') | |
599 | if self._downloader.params.get('listformats', None): | |
600 | self._print_formats(existing_formats) | |
601 | return | |
602 | if req_format is None or req_format == 'best': | |
603 | video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality | |
604 | elif req_format == 'worst': | |
d828f3a5 | 605 | video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality |
c5e8d7af PH |
606 | elif req_format in ('-1', 'all'): |
607 | video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats | |
608 | else: | |
609 | # Specific formats. We pick the first in a slash-delimeted sequence. | |
610 | # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. | |
611 | req_formats = req_format.split('/') | |
612 | video_url_list = None | |
613 | for rf in req_formats: | |
614 | if rf in url_map: | |
615 | video_url_list = [(rf, url_map[rf])] | |
616 | break | |
617 | if video_url_list is None: | |
618 | raise ExtractorError(u'requested format not available') | |
619 | else: | |
620 | raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info') | |
621 | ||
622 | results = [] | |
623 | for format_param, video_real_url in video_url_list: | |
624 | # Extension | |
625 | video_extension = self._video_extensions.get(format_param, 'flv') | |
626 | ||
627 | video_format = '{0} - {1}'.format(format_param if format_param else video_extension, | |
628 | self._video_dimensions.get(format_param, '???')) | |
629 | ||
630 | results.append({ | |
631 | 'id': video_id, | |
632 | 'url': video_real_url, | |
633 | 'uploader': video_uploader, | |
634 | 'uploader_id': video_uploader_id, | |
635 | 'upload_date': upload_date, | |
636 | 'title': video_title, | |
637 | 'ext': video_extension, | |
638 | 'format': video_format, | |
639 | 'thumbnail': video_thumbnail, | |
640 | 'description': video_description, | |
641 | 'player_url': player_url, | |
642 | 'subtitles': video_subtitles, | |
643 | 'duration': video_duration | |
644 | }) | |
645 | return results | |
646 | ||
647 | class YoutubePlaylistIE(InfoExtractor): | |
0f818663 | 648 | IE_DESC = u'YouTube.com playlists' |
c5e8d7af PH |
649 | _VALID_URL = r"""(?: |
650 | (?:https?://)? | |
651 | (?:\w+\.)? | |
652 | youtube\.com/ | |
653 | (?: | |
654 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
655 | \? (?:.*?&)*? (?:p|a|list)= | |
656 | | p/ | |
657 | ) | |
658 | ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,}) | |
659 | .* | |
660 | | | |
661 | ((?:PL|EC|UU)[0-9A-Za-z-_]{10,}) | |
662 | )""" | |
663 | _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none' | |
664 | _MAX_RESULTS = 50 | |
665 | IE_NAME = u'youtube:playlist' | |
666 | ||
667 | @classmethod | |
668 | def suitable(cls, url): | |
669 | """Receives a URL and returns True if suitable for this IE.""" | |
670 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
671 | ||
672 | def _real_extract(self, url): | |
673 | # Extract playlist id | |
674 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
675 | if mobj is None: | |
676 | raise ExtractorError(u'Invalid URL: %s' % url) | |
677 | ||
678 | # Download playlist videos from API | |
679 | playlist_id = mobj.group(1) or mobj.group(2) | |
680 | page_num = 1 | |
681 | videos = [] | |
682 | ||
683 | while True: | |
684 | url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1) | |
685 | page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) | |
686 | ||
687 | try: | |
688 | response = json.loads(page) | |
689 | except ValueError as err: | |
690 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
691 | ||
692 | if 'feed' not in response: | |
693 | raise ExtractorError(u'Got a malformed response from YouTube API') | |
694 | playlist_title = response['feed']['title']['$t'] | |
695 | if 'entry' not in response['feed']: | |
696 | # Number of videos is a multiple of self._MAX_RESULTS | |
697 | break | |
698 | ||
699 | for entry in response['feed']['entry']: | |
700 | index = entry['yt$position']['$t'] | |
701 | if 'media$group' in entry and 'media$player' in entry['media$group']: | |
702 | videos.append((index, entry['media$group']['media$player']['url'])) | |
703 | ||
704 | if len(response['feed']['entry']) < self._MAX_RESULTS: | |
705 | break | |
706 | page_num += 1 | |
707 | ||
708 | videos = [v[1] for v in sorted(videos)] | |
709 | ||
20c3893f | 710 | url_results = [self.url_result(vurl, 'Youtube') for vurl in videos] |
c5e8d7af PH |
711 | return [self.playlist_result(url_results, playlist_id, playlist_title)] |
712 | ||
713 | ||
714 | class YoutubeChannelIE(InfoExtractor): | |
0f818663 | 715 | IE_DESC = u'YouTube.com channels' |
c5e8d7af PH |
716 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
717 | _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en' | |
718 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' | |
719 | _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' | |
720 | IE_NAME = u'youtube:channel' | |
721 | ||
722 | def extract_videos_from_page(self, page): | |
723 | ids_in_page = [] | |
724 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
725 | if mobj.group(1) not in ids_in_page: | |
726 | ids_in_page.append(mobj.group(1)) | |
727 | return ids_in_page | |
728 | ||
729 | def _real_extract(self, url): | |
730 | # Extract channel id | |
731 | mobj = re.match(self._VALID_URL, url) | |
732 | if mobj is None: | |
733 | raise ExtractorError(u'Invalid URL: %s' % url) | |
734 | ||
735 | # Download channel page | |
736 | channel_id = mobj.group(1) | |
737 | video_ids = [] | |
738 | pagenum = 1 | |
739 | ||
740 | url = self._TEMPLATE_URL % (channel_id, pagenum) | |
741 | page = self._download_webpage(url, channel_id, | |
742 | u'Downloading page #%s' % pagenum) | |
743 | ||
744 | # Extract video identifiers | |
745 | ids_in_page = self.extract_videos_from_page(page) | |
746 | video_ids.extend(ids_in_page) | |
747 | ||
748 | # Download any subsequent channel pages using the json-based channel_ajax query | |
749 | if self._MORE_PAGES_INDICATOR in page: | |
750 | while True: | |
751 | pagenum = pagenum + 1 | |
752 | ||
753 | url = self._MORE_PAGES_URL % (pagenum, channel_id) | |
754 | page = self._download_webpage(url, channel_id, | |
755 | u'Downloading page #%s' % pagenum) | |
756 | ||
757 | page = json.loads(page) | |
758 | ||
759 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
760 | video_ids.extend(ids_in_page) | |
761 | ||
762 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
763 | break | |
764 | ||
765 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
766 | ||
767 | urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids] | |
20c3893f | 768 | url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls] |
c5e8d7af PH |
769 | return [self.playlist_result(url_entries, channel_id)] |
770 | ||
771 | ||
772 | class YoutubeUserIE(InfoExtractor): | |
0f818663 | 773 | IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' |
c5e8d7af PH |
774 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)' |
775 | _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' | |
776 | _GDATA_PAGE_SIZE = 50 | |
777 | _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d' | |
778 | _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]' | |
779 | IE_NAME = u'youtube:user' | |
780 | ||
781 | def _real_extract(self, url): | |
782 | # Extract username | |
783 | mobj = re.match(self._VALID_URL, url) | |
784 | if mobj is None: | |
785 | raise ExtractorError(u'Invalid URL: %s' % url) | |
786 | ||
787 | username = mobj.group(1) | |
788 | ||
789 | # Download video ids using YouTube Data API. Result size per | |
790 | # query is limited (currently to 50 videos) so we need to query | |
791 | # page by page until there are no video ids - it means we got | |
792 | # all of them. | |
793 | ||
794 | video_ids = [] | |
795 | pagenum = 0 | |
796 | ||
797 | while True: | |
798 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 | |
799 | ||
800 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
801 | page = self._download_webpage(gdata_url, username, | |
802 | u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) | |
803 | ||
804 | # Extract video identifiers | |
805 | ids_in_page = [] | |
806 | ||
807 | for mobj in re.finditer(self._VIDEO_INDICATOR, page): | |
808 | if mobj.group(1) not in ids_in_page: | |
809 | ids_in_page.append(mobj.group(1)) | |
810 | ||
811 | video_ids.extend(ids_in_page) | |
812 | ||
813 | # A little optimization - if current page is not | |
814 | # "full", ie. does not contain PAGE_SIZE video ids then | |
815 | # we can assume that this page is the last one - there | |
816 | # are no more ids on further pages - no need to query | |
817 | # again. | |
818 | ||
819 | if len(ids_in_page) < self._GDATA_PAGE_SIZE: | |
820 | break | |
821 | ||
822 | pagenum += 1 | |
823 | ||
824 | urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] | |
20c3893f | 825 | url_results = [self.url_result(rurl, 'Youtube') for rurl in urls] |
c5e8d7af | 826 | return [self.playlist_result(url_results, playlist_title = username)] |
b05654f0 PH |
827 | |
828 | class YoutubeSearchIE(SearchInfoExtractor): | |
0f818663 | 829 | IE_DESC = u'YouTube.com searches' |
b05654f0 PH |
830 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' |
831 | _MAX_RESULTS = 1000 | |
832 | IE_NAME = u'youtube:search' | |
833 | _SEARCH_KEY = 'ytsearch' | |
834 | ||
835 | def report_download_page(self, query, pagenum): | |
836 | """Report attempt to download search page with given number.""" | |
837 | self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) | |
838 | ||
839 | def _get_n_results(self, query, n): | |
840 | """Get a specified number of results for a query""" | |
841 | ||
842 | video_ids = [] | |
843 | pagenum = 0 | |
844 | limit = n | |
845 | ||
846 | while (50 * pagenum) < limit: | |
847 | self.report_download_page(query, pagenum+1) | |
848 | result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) | |
849 | request = compat_urllib_request.Request(result_url) | |
850 | try: | |
851 | data = compat_urllib_request.urlopen(request).read().decode('utf-8') | |
852 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
853 | raise ExtractorError(u'Unable to download API page: %s' % compat_str(err)) | |
854 | api_response = json.loads(data)['data'] | |
855 | ||
856 | if not 'items' in api_response: | |
857 | raise ExtractorError(u'[youtube] No video results') | |
858 | ||
859 | new_ids = list(video['id'] for video in api_response['items']) | |
860 | video_ids += new_ids | |
861 | ||
862 | limit = min(n, api_response['totalItems']) | |
863 | pagenum += 1 | |
864 | ||
865 | if len(video_ids) > n: | |
866 | video_ids = video_ids[:n] | |
867 | videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids] | |
868 | return self.playlist_result(videos, query) | |
75dff0ee JMF |
869 | |
870 | ||
871 | class YoutubeShowIE(InfoExtractor): | |
0f818663 | 872 | IE_DESC = u'YouTube.com (multi-season) shows' |
75dff0ee JMF |
873 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
874 | IE_NAME = u'youtube:show' | |
875 | ||
876 | def _real_extract(self, url): | |
877 | mobj = re.match(self._VALID_URL, url) | |
878 | show_name = mobj.group(1) | |
879 | webpage = self._download_webpage(url, show_name, u'Downloading show webpage') | |
880 | # There's one playlist for each season of the show | |
881 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
882 | self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) | |
883 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] | |
04cc9617 JMF |
884 | |
885 | ||
886 | class YoutubeSubscriptionsIE(YoutubeIE): | |
887 | """It's a subclass of YoutubeIE because we need to login""" | |
897f36d1 PH |
888 | IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' |
889 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
04cc9617 JMF |
890 | IE_NAME = u'youtube:subscriptions' |
891 | _FEED_TEMPLATE = 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=subscriptions&paging=%s' | |
892 | _PAGING_STEP = 30 | |
893 | ||
897f36d1 | 894 | # Overwrite YoutubeIE properties we don't want |
04cc9617 | 895 | _TESTS = [] |
04cc9617 JMF |
896 | @classmethod |
897 | def suitable(cls, url): | |
898 | return re.match(cls._VALID_URL, url) is not None | |
899 | ||
81f0259b JMF |
900 | def _real_initialize(self): |
901 | (username, password) = self._get_login_info() | |
902 | if username is None: | |
903 | raise ExtractorError(u'No login info available, needed for downloading the Youtube subscriptions.', expected=True) | |
904 | super(YoutubeSubscriptionsIE, self)._real_initialize() | |
905 | ||
04cc9617 JMF |
906 | def _real_extract(self, url): |
907 | feed_entries = [] | |
908 | # The step argument is available only in 2.7 or higher | |
909 | for i in itertools.count(0): | |
910 | paging = i*self._PAGING_STEP | |
911 | info = self._download_webpage(self._FEED_TEMPLATE % paging, 'feed', | |
912 | u'Downloading page %s' % i) | |
913 | info = json.loads(info) | |
914 | feed_html = info['feed_html'] | |
915 | m_ids = re.finditer(r'"/watch\?v=(.*?)"', feed_html) | |
916 | ids = orderedSet(m.group(1) for m in m_ids) | |
917 | feed_entries.extend(self.url_result(id, 'Youtube') for id in ids) | |
918 | if info['paging'] is None: | |
919 | break | |
920 | return self.playlist_result(feed_entries, playlist_title='Youtube Subscriptions') |