8 from .common import InfoExtractor, SearchInfoExtractor
14 compat_urllib_request,
25 class YoutubeIE(InfoExtractor):
26 """Information extractor for youtube.com."""
30 (?:https?://)? # http(s):// (optional)
31 (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/|
32 tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains
33 (?:.*?\#/)? # handle anchor (#/) redirect urls
34 (?: # the various things that can precede the ID:
35 (?:(?:v|embed|e)/) # v/ or embed/ or e/
36 |(?: # or the v= param in all its forms
37 (?:watch(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
38 (?:\?|\#!?) # the params delimiter ? or # or #!
39 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
42 )? # optional -> youtube.com/xxxx is OK
43 )? # all until now is optional -> you can pass the naked ID
44 ([0-9A-Za-z_-]+) # here is it! the YouTube video ID
45 (?(1).+)? # if we found the ID, everything can follow
47 _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
48 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
49 _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
50 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
51 _NETRC_MACHINE = 'youtube'
52 # Listed in order of quality
53 _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13']
54 _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13']
86 def suitable(cls, url):
87 """Receives a URL and returns True if suitable for this IE."""
88 if YoutubePlaylistIE.suitable(url): return False
89 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
91 def report_lang(self):
92 """Report attempt to set language."""
93 self.to_screen(u'Setting language')
95 def report_login(self):
96 """Report attempt to log in."""
97 self.to_screen(u'Logging in')
99 def report_video_webpage_download(self, video_id):
100 """Report attempt to download video webpage."""
101 self.to_screen(u'%s: Downloading video webpage' % video_id)
103 def report_video_info_webpage_download(self, video_id):
104 """Report attempt to download video info webpage."""
105 self.to_screen(u'%s: Downloading video info webpage' % video_id)
107 def report_video_subtitles_download(self, video_id):
108 """Report attempt to download video info webpage."""
109 self.to_screen(u'%s: Checking available subtitles' % video_id)
111 def report_video_subtitles_request(self, video_id, sub_lang, format):
112 """Report attempt to download video info webpage."""
113 self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
115 def report_video_subtitles_available(self, video_id, sub_lang_list):
116 """Report available subtitles."""
117 sub_lang = ",".join(list(sub_lang_list.keys()))
118 self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
120 def report_information_extraction(self, video_id):
121 """Report attempt to extract video information."""
122 self.to_screen(u'%s: Extracting video information' % video_id)
124 def report_unavailable_format(self, video_id, format):
125 """Report extracted video URL."""
126 self.to_screen(u'%s: Format %s not available' % (video_id, format))
128 def report_rtmp_download(self):
129 """Indicate the download will use the RTMP protocol."""
130 self.to_screen(u'RTMP download detected')
133 def _decrypt_signature(s):
134 """Decrypt the key the two subkeys must have a length of 43"""
136 if len(a) != 43 or len(b) != 43:
137 raise ExtractorError(u'Unable to decrypt signature, subkeys lengths not valid')
138 b = ''.join([b[:8],a[0],b[9:18],b[-4],b[19:39], b[18]])[0:40]
140 s_dec = '.'.join((a,b))[::-1]
143 def _get_available_subtitles(self, video_id):
144 self.report_video_subtitles_download(video_id)
145 request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
147 sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
148 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
149 return (u'unable to download video subtitles: %s' % compat_str(err), None)
150 sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
151 sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
152 if not sub_lang_list:
153 return (u'video doesn\'t have subtitles', None)
156 def _list_available_subtitles(self, video_id):
157 sub_lang_list = self._get_available_subtitles(video_id)
158 self.report_video_subtitles_available(video_id, sub_lang_list)
160 def _request_subtitle(self, sub_lang, sub_name, video_id, format):
163 (error_message, sub_lang, sub)
165 self.report_video_subtitles_request(video_id, sub_lang, format)
166 params = compat_urllib_parse.urlencode({
172 url = 'http://www.youtube.com/api/timedtext?' + params
174 sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
175 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
176 return (u'unable to download video subtitles: %s' % compat_str(err), None, None)
178 return (u'Did not fetch video subtitles', None, None)
179 return (None, sub_lang, sub)
181 def _request_automatic_caption(self, video_id, webpage):
182 """We need the webpage for getting the captions url, pass it as an
183 argument to speed up the process."""
184 sub_lang = self._downloader.params.get('subtitleslang') or 'en'
185 sub_format = self._downloader.params.get('subtitlesformat')
186 self.to_screen(u'%s: Looking for automatic captions' % video_id)
187 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
188 err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
190 return [(err_msg, None, None)]
191 player_config = json.loads(mobj.group(1))
193 args = player_config[u'args']
194 caption_url = args[u'ttsurl']
195 timestamp = args[u'timestamp']
196 params = compat_urllib_parse.urlencode({
203 subtitles_url = caption_url + '&' + params
204 sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
205 return [(None, sub_lang, sub)]
207 return [(err_msg, None, None)]
209 def _extract_subtitle(self, video_id):
211 Return a list with a tuple:
212 [(error_message, sub_lang, sub)]
214 sub_lang_list = self._get_available_subtitles(video_id)
215 sub_format = self._downloader.params.get('subtitlesformat')
216 if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
217 return [(sub_lang_list[0], None, None)]
218 if self._downloader.params.get('subtitleslang', False):
219 sub_lang = self._downloader.params.get('subtitleslang')
220 elif 'en' in sub_lang_list:
223 sub_lang = list(sub_lang_list.keys())[0]
224 if not sub_lang in sub_lang_list:
225 return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)]
227 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
230 def _extract_all_subtitles(self, video_id):
231 sub_lang_list = self._get_available_subtitles(video_id)
232 sub_format = self._downloader.params.get('subtitlesformat')
233 if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
234 return [(sub_lang_list[0], None, None)]
236 for sub_lang in sub_lang_list:
237 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
238 subtitles.append(subtitle)
241 def _print_formats(self, formats):
242 print('Available formats:')
244 print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???')))
246 def _real_initialize(self):
247 if self._downloader is None:
252 downloader_params = self._downloader.params
254 # Attempt to use provided username and password or .netrc data
255 if downloader_params.get('username', None) is not None:
256 username = downloader_params['username']
257 password = downloader_params['password']
258 elif downloader_params.get('usenetrc', False):
260 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
265 raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
266 except (IOError, netrc.NetrcParseError) as err:
267 self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
271 request = compat_urllib_request.Request(self._LANG_URL)
274 compat_urllib_request.urlopen(request).read()
275 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
276 self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
279 # No authentication to be performed
283 request = compat_urllib_request.Request(self._LOGIN_URL)
285 login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
286 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
287 self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
292 match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
294 galx = match.group(1)
296 match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
302 u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
306 u'PersistentCookie': u'yes',
308 u'bgresponse': u'js_disabled',
309 u'checkConnection': u'',
310 u'checkedDomains': u'youtube',
316 u'signIn': u'Sign in',
318 u'service': u'youtube',
322 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
324 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
325 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
326 request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
329 login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
330 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
331 self._downloader.report_warning(u'unable to log in: bad username or password')
333 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
334 self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
340 'action_confirm': 'Confirm',
342 request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
344 self.report_age_confirmation()
345 compat_urllib_request.urlopen(request).read().decode('utf-8')
346 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
347 raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
349 def _extract_id(self, url):
350 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
352 raise ExtractorError(u'Invalid URL: %s' % url)
353 video_id = mobj.group(2)
356 def _real_extract(self, url):
357 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
358 mobj = re.search(self._NEXT_URL_RE, url)
360 url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
361 video_id = self._extract_id(url)
364 self.report_video_webpage_download(video_id)
365 url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
366 request = compat_urllib_request.Request(url)
368 video_webpage_bytes = compat_urllib_request.urlopen(request).read()
369 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
370 raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
372 video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
374 # Attempt to extract SWF player URL
375 mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
377 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
382 self.report_video_info_webpage_download(video_id)
383 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
384 video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
385 % (video_id, el_type))
386 video_info_webpage = self._download_webpage(video_info_url, video_id,
388 errnote='unable to download video info webpage')
389 video_info = compat_parse_qs(video_info_webpage)
390 if 'token' in video_info:
392 if 'token' not in video_info:
393 if 'reason' in video_info:
394 raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0])
396 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
398 # Check for "rental" videos
399 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
400 raise ExtractorError(u'"rental" videos not supported')
402 # Start extracting information
403 self.report_information_extraction(video_id)
406 if 'author' not in video_info:
407 raise ExtractorError(u'Unable to extract uploader name')
408 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
411 video_uploader_id = None
412 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
414 video_uploader_id = mobj.group(1)
416 self._downloader.report_warning(u'unable to extract uploader nickname')
419 if 'title' not in video_info:
420 raise ExtractorError(u'Unable to extract video title')
421 video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
424 if 'thumbnail_url' not in video_info:
425 self._downloader.report_warning(u'unable to extract video thumbnail')
427 else: # don't panic if we can't find it
428 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
432 mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
434 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
435 upload_date = unified_strdate(upload_date)
438 video_description = get_element_by_id("eow-description", video_webpage)
439 if video_description:
440 video_description = clean_html(video_description)
442 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
444 video_description = unescapeHTML(fd_mobj.group(1))
446 video_description = u''
449 video_subtitles = None
451 if self._downloader.params.get('writesubtitles', False):
452 video_subtitles = self._extract_subtitle(video_id)
454 (sub_error, sub_lang, sub) = video_subtitles[0]
456 # We try with the automatic captions
457 video_subtitles = self._request_automatic_caption(video_id, video_webpage)
458 (sub_error_auto, sub_lang, sub) = video_subtitles[0]
462 # We report the original error
463 self._downloader.report_warning(sub_error)
465 if self._downloader.params.get('allsubtitles', False):
466 video_subtitles = self._extract_all_subtitles(video_id)
467 for video_subtitle in video_subtitles:
468 (sub_error, sub_lang, sub) = video_subtitle
470 self._downloader.report_warning(sub_error)
472 if self._downloader.params.get('listsubtitles', False):
473 self._list_available_subtitles(video_id)
476 if 'length_seconds' not in video_info:
477 self._downloader.report_warning(u'unable to extract video duration')
480 video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
482 # Decide which formats to download
483 req_format = self._downloader.params.get('format', None)
486 mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
487 info = json.loads(mobj.group(1))
489 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
490 # this signatures are encrypted
491 m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
493 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
494 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
498 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
499 self.report_rtmp_download()
500 video_url_list = [(None, video_info['conn'][0])]
501 elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
503 for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
504 url_data = compat_parse_qs(url_data_str)
505 if 'itag' in url_data and 'url' in url_data:
506 url = url_data['url'][0]
507 if 'sig' in url_data:
508 url += '&signature=' + url_data['sig'][0]
509 elif 's' in url_data:
510 signature = self._decrypt_signature(url_data['s'][0])
511 url += '&signature=' + signature
512 if 'ratebypass' not in url:
513 url += '&ratebypass=yes'
514 url_map[url_data['itag'][0]] = url
516 format_limit = self._downloader.params.get('format_limit', None)
517 available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
518 if format_limit is not None and format_limit in available_formats:
519 format_list = available_formats[available_formats.index(format_limit):]
521 format_list = available_formats
522 existing_formats = [x for x in format_list if x in url_map]
523 if len(existing_formats) == 0:
524 raise ExtractorError(u'no known formats available for video')
525 if self._downloader.params.get('listformats', None):
526 self._print_formats(existing_formats)
528 if req_format is None or req_format == 'best':
529 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
530 elif req_format == 'worst':
531 video_url_list = [(existing_formats[len(existing_formats)-1], url_map[existing_formats[len(existing_formats)-1]])] # worst quality
532 elif req_format in ('-1', 'all'):
533 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
535 # Specific formats. We pick the first in a slash-delimeted sequence.
536 # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
537 req_formats = req_format.split('/')
538 video_url_list = None
539 for rf in req_formats:
541 video_url_list = [(rf, url_map[rf])]
543 if video_url_list is None:
544 raise ExtractorError(u'requested format not available')
546 raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
549 for format_param, video_real_url in video_url_list:
551 video_extension = self._video_extensions.get(format_param, 'flv')
553 video_format = '{0} - {1}'.format(format_param if format_param else video_extension,
554 self._video_dimensions.get(format_param, '???'))
558 'url': video_real_url,
559 'uploader': video_uploader,
560 'uploader_id': video_uploader_id,
561 'upload_date': upload_date,
562 'title': video_title,
563 'ext': video_extension,
564 'format': video_format,
565 'thumbnail': video_thumbnail,
566 'description': video_description,
567 'player_url': player_url,
568 'subtitles': video_subtitles,
569 'duration': video_duration
573 class YoutubePlaylistIE(InfoExtractor):
574 """Information Extractor for YouTube playlists."""
581 (?:course|view_play_list|my_playlists|artist|playlist|watch)
582 \? (?:.*?&)*? (?:p|a|list)=
585 ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,})
588 ((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
590 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
592 IE_NAME = u'youtube:playlist'
595 def suitable(cls, url):
596 """Receives a URL and returns True if suitable for this IE."""
597 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
599 def _real_extract(self, url):
600 # Extract playlist id
601 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
603 raise ExtractorError(u'Invalid URL: %s' % url)
605 # Download playlist videos from API
606 playlist_id = mobj.group(1) or mobj.group(2)
611 url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1)
612 page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
615 response = json.loads(page)
616 except ValueError as err:
617 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
619 if 'feed' not in response:
620 raise ExtractorError(u'Got a malformed response from YouTube API')
621 playlist_title = response['feed']['title']['$t']
622 if 'entry' not in response['feed']:
623 # Number of videos is a multiple of self._MAX_RESULTS
626 for entry in response['feed']['entry']:
627 index = entry['yt$position']['$t']
628 if 'media$group' in entry and 'media$player' in entry['media$group']:
629 videos.append((index, entry['media$group']['media$player']['url']))
631 if len(response['feed']['entry']) < self._MAX_RESULTS:
635 videos = [v[1] for v in sorted(videos)]
637 url_results = [self.url_result(url, 'Youtube') for url in videos]
638 return [self.playlist_result(url_results, playlist_id, playlist_title)]
641 class YoutubeChannelIE(InfoExtractor):
642 """Information Extractor for YouTube channels."""
644 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
645 _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
646 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
647 _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'
648 IE_NAME = u'youtube:channel'
650 def extract_videos_from_page(self, page):
652 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
653 if mobj.group(1) not in ids_in_page:
654 ids_in_page.append(mobj.group(1))
657 def _real_extract(self, url):
659 mobj = re.match(self._VALID_URL, url)
661 raise ExtractorError(u'Invalid URL: %s' % url)
663 # Download channel page
664 channel_id = mobj.group(1)
668 url = self._TEMPLATE_URL % (channel_id, pagenum)
669 page = self._download_webpage(url, channel_id,
670 u'Downloading page #%s' % pagenum)
672 # Extract video identifiers
673 ids_in_page = self.extract_videos_from_page(page)
674 video_ids.extend(ids_in_page)
676 # Download any subsequent channel pages using the json-based channel_ajax query
677 if self._MORE_PAGES_INDICATOR in page:
679 pagenum = pagenum + 1
681 url = self._MORE_PAGES_URL % (pagenum, channel_id)
682 page = self._download_webpage(url, channel_id,
683 u'Downloading page #%s' % pagenum)
685 page = json.loads(page)
687 ids_in_page = self.extract_videos_from_page(page['content_html'])
688 video_ids.extend(ids_in_page)
690 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
693 self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
695 urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
696 url_entries = [self.url_result(url, 'Youtube') for url in urls]
697 return [self.playlist_result(url_entries, channel_id)]
700 class YoutubeUserIE(InfoExtractor):
701 """Information Extractor for YouTube users."""
703 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
704 _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
705 _GDATA_PAGE_SIZE = 50
706 _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
707 _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]'
708 IE_NAME = u'youtube:user'
710 def _real_extract(self, url):
712 mobj = re.match(self._VALID_URL, url)
714 raise ExtractorError(u'Invalid URL: %s' % url)
716 username = mobj.group(1)
718 # Download video ids using YouTube Data API. Result size per
719 # query is limited (currently to 50 videos) so we need to query
720 # page by page until there are no video ids - it means we got
727 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
729 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
730 page = self._download_webpage(gdata_url, username,
731 u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
733 # Extract video identifiers
736 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
737 if mobj.group(1) not in ids_in_page:
738 ids_in_page.append(mobj.group(1))
740 video_ids.extend(ids_in_page)
742 # A little optimization - if current page is not
743 # "full", ie. does not contain PAGE_SIZE video ids then
744 # we can assume that this page is the last one - there
745 # are no more ids on further pages - no need to query
748 if len(ids_in_page) < self._GDATA_PAGE_SIZE:
753 urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
754 url_results = [self.url_result(url, 'Youtube') for url in urls]
755 return [self.playlist_result(url_results, playlist_title = username)]
757 class YoutubeSearchIE(SearchInfoExtractor):
758 """Information Extractor for YouTube search queries."""
759 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
761 IE_NAME = u'youtube:search'
762 _SEARCH_KEY = 'ytsearch'
764 def report_download_page(self, query, pagenum):
765 """Report attempt to download search page with given number."""
766 self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
768 def _get_n_results(self, query, n):
769 """Get a specified number of results for a query"""
775 while (50 * pagenum) < limit:
776 self.report_download_page(query, pagenum+1)
777 result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
778 request = compat_urllib_request.Request(result_url)
780 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
781 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
782 raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
783 api_response = json.loads(data)['data']
785 if not 'items' in api_response:
786 raise ExtractorError(u'[youtube] No video results')
788 new_ids = list(video['id'] for video in api_response['items'])
791 limit = min(n, api_response['totalItems'])
794 if len(video_ids) > n:
795 video_ids = video_ids[:n]
796 videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
797 return self.playlist_result(videos, query)