11 from .common import InfoExtractor, SearchInfoExtractor
12 from .subtitles import SubtitlesInfoExtractor
13 from ..jsinterp import JSInterpreter
14 from ..swfinterp import SWFInterpreter
19 compat_urllib_request,
26 get_element_by_attribute,
37 class YoutubeBaseInfoExtractor(InfoExtractor):
38 """Provide base functions for Youtube extractors"""
39 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
40 _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
41 _AGE_URL = 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
42 _NETRC_MACHINE = 'youtube'
43 # If True it will raise an error if no login info is provided
44 _LOGIN_REQUIRED = False
46 def _set_language(self):
47 return bool(self._download_webpage(
49 note=u'Setting language', errnote='unable to set language',
53 (username, password) = self._get_login_info()
54 # No authentication to be performed
56 if self._LOGIN_REQUIRED:
57 raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
60 login_page = self._download_webpage(
61 self._LOGIN_URL, None,
62 note=u'Downloading login page',
63 errnote=u'unable to fetch login page', fatal=False)
64 if login_page is False:
67 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
68 login_page, u'Login GALX parameter')
72 u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
76 u'PersistentCookie': u'yes',
78 u'bgresponse': u'js_disabled',
79 u'checkConnection': u'',
80 u'checkedDomains': u'youtube',
85 u'signIn': u'Sign in',
87 u'service': u'youtube',
91 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
93 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
94 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
96 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
97 login_results = self._download_webpage(
99 note=u'Logging in', errnote=u'unable to log in', fatal=False)
100 if login_results is False:
102 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
103 self._downloader.report_warning(u'unable to log in: bad username or password')
107 def _confirm_age(self):
110 'action_confirm': 'Confirm',
112 req = compat_urllib_request.Request(self._AGE_URL,
113 compat_urllib_parse.urlencode(age_form).encode('ascii'))
115 self._download_webpage(
117 note=u'Confirming age', errnote=u'Unable to confirm age')
120 def _real_initialize(self):
121 if self._downloader is None:
123 if not self._set_language():
125 if not self._login():
130 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
131 IE_DESC = u'YouTube.com'
132 _VALID_URL = r"""(?x)^
134 (?:https?://|//)? # http(s):// or protocol-independent URL (optional)
135 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
136 (?:www\.)?deturl\.com/www\.youtube\.com/|
137 (?:www\.)?pwnyoutube\.com/|
138 (?:www\.)?yourepeat\.com/|
139 tube\.majestyc\.net/|
140 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
141 (?:.*?\#/)? # handle anchor (#/) redirect urls
142 (?: # the various things that can precede the ID:
143 (?:(?:v|embed|e)/) # v/ or embed/ or e/
144 |(?: # or the v= param in all its forms
145 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
146 (?:\?|\#!?) # the params delimiter ? or # or #!
147 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
151 |youtu\.be/ # just youtu.be/xxxx
152 |https?://(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
154 )? # all until now is optional -> you can pass the naked ID
155 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
156 (?(1).+)? # if we found the ID, everything can follow
158 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
160 '5': {'ext': 'flv', 'width': 400, 'height': 240},
161 '6': {'ext': 'flv', 'width': 450, 'height': 270},
162 '13': {'ext': '3gp'},
163 '17': {'ext': '3gp', 'width': 176, 'height': 144},
164 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
165 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
166 '34': {'ext': 'flv', 'width': 640, 'height': 360},
167 '35': {'ext': 'flv', 'width': 854, 'height': 480},
168 '36': {'ext': '3gp', 'width': 320, 'height': 240},
169 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
170 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
171 '43': {'ext': 'webm', 'width': 640, 'height': 360},
172 '44': {'ext': 'webm', 'width': 854, 'height': 480},
173 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
174 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
178 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
179 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
180 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
181 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
182 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
183 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
184 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
186 # Apple HTTP Live Streaming
187 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
188 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
189 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
190 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
191 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
192 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
193 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
196 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
197 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
198 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
199 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
200 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
201 '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
202 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
203 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
206 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
207 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
208 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
211 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
212 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
213 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
214 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
215 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
216 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
217 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
218 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
219 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
220 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
221 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
222 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
223 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
224 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
225 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
228 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 48, 'preference': -50},
229 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
232 '_rtmp': {'protocol': 'rtmp'},
238 u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc",
239 u"file": u"BaW_jenozKc.mp4",
241 u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
242 u"uploader": u"Philipp Hagemeister",
243 u"uploader_id": u"phihag",
244 u"upload_date": u"20121002",
245 u"description": u"test chars: \"'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .",
246 u"categories": [u'Science & Technology'],
250 u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY",
251 u"file": u"UxxajLWwzqY.mp4",
252 u"note": u"Test generic use_cipher_signature video (#897)",
254 u"upload_date": u"20120506",
255 u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
256 u"description": u"md5:fea86fda2d5a5784273df5c7cc994d9f",
257 u"uploader": u"Icona Pop",
258 u"uploader_id": u"IconaPop"
262 u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ",
263 u"file": u"07FYdnEawAQ.mp4",
264 u"note": u"Test VEVO video with age protection (#956)",
266 u"upload_date": u"20130703",
267 u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
268 u"description": u"md5:64249768eec3bc4276236606ea996373",
269 u"uploader": u"justintimberlakeVEVO",
270 u"uploader_id": u"justintimberlakeVEVO"
274 u"url": u"//www.YouTube.com/watch?v=yZIXLfi8CZQ",
275 u"file": u"yZIXLfi8CZQ.mp4",
276 u"note": u"Embed-only video (#1746)",
278 u"upload_date": u"20120608",
279 u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012",
280 u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7",
281 u"uploader": u"SET India",
282 u"uploader_id": u"setindia"
286 u"url": u"http://www.youtube.com/watch?v=a9LDPn-MO4I",
287 u"file": u"a9LDPn-MO4I.m4a",
288 u"note": u"256k DASH audio (format 141) via DASH manifest",
290 u"upload_date": "20121002",
291 u"uploader_id": "8KVIDEO",
292 u"description": "No description available.",
293 u"uploader": "8KVIDEO",
294 u"title": "UHDTV TEST 8K VIDEO.mp4"
297 u"youtube_include_dash_manifest": True,
301 # DASH manifest with encrypted signature
303 u'url': u'https://www.youtube.com/watch?v=IB3lcPjvWLA',
305 u'id': u'IB3lcPjvWLA',
307 u'title': u'Afrojack - The Spark ft. Spree Wilson',
308 u'description': u'md5:9717375db5a9a3992be4668bbf3bc0a8',
309 u'uploader': u'AfrojackVEVO',
310 u'uploader_id': u'AfrojackVEVO',
311 u'upload_date': u'20131011',
314 u'youtube_include_dash_manifest': True,
322 def suitable(cls, url):
323 """Receives a URL and returns True if suitable for this IE."""
324 if YoutubePlaylistIE.suitable(url): return False
325 return re.match(cls._VALID_URL, url) is not None
327 def __init__(self, *args, **kwargs):
328 super(YoutubeIE, self).__init__(*args, **kwargs)
329 self._player_cache = {}
331 def report_video_info_webpage_download(self, video_id):
332 """Report attempt to download video info webpage."""
333 self.to_screen(u'%s: Downloading video info webpage' % video_id)
335 def report_information_extraction(self, video_id):
336 """Report attempt to extract video information."""
337 self.to_screen(u'%s: Extracting video information' % video_id)
339 def report_unavailable_format(self, video_id, format):
340 """Report extracted video URL."""
341 self.to_screen(u'%s: Format %s not available' % (video_id, format))
343 def report_rtmp_download(self):
344 """Indicate the download will use the RTMP protocol."""
345 self.to_screen(u'RTMP download detected')
347 def _extract_signature_function(self, video_id, player_url, slen):
349 r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
352 raise ExtractorError('Cannot identify player %r' % player_url)
353 player_type = id_m.group('ext')
354 player_id = id_m.group('id')
356 # Read from filesystem cache
357 func_id = '%s_%s_%d' % (player_type, player_id, slen)
358 assert os.path.basename(func_id) == func_id
359 cache_dir = get_cachedir(self._downloader.params)
361 cache_enabled = cache_dir is not None
363 cache_fn = os.path.join(os.path.expanduser(cache_dir),
367 with io.open(cache_fn, 'r', encoding='utf-8') as cachef:
368 cache_spec = json.load(cachef)
369 return lambda s: u''.join(s[i] for i in cache_spec)
371 pass # No cache available
373 if player_type == 'js':
374 code = self._download_webpage(
375 player_url, video_id,
376 note=u'Downloading %s player %s' % (player_type, player_id),
377 errnote=u'Download of %s failed' % player_url)
378 res = self._parse_sig_js(code)
379 elif player_type == 'swf':
380 urlh = self._request_webpage(
381 player_url, video_id,
382 note=u'Downloading %s player %s' % (player_type, player_id),
383 errnote=u'Download of %s failed' % player_url)
385 res = self._parse_sig_swf(code)
387 assert False, 'Invalid player type %r' % player_type
391 test_string = u''.join(map(compat_chr, range(slen)))
392 cache_res = res(test_string)
393 cache_spec = [ord(c) for c in cache_res]
395 os.makedirs(os.path.dirname(cache_fn))
396 except OSError as ose:
397 if ose.errno != errno.EEXIST:
399 write_json_file(cache_spec, cache_fn)
401 tb = traceback.format_exc()
402 self._downloader.report_warning(
403 u'Writing cache to %r failed: %s' % (cache_fn, tb))
407 def _print_sig_code(self, func, slen):
408 def gen_sig_code(idxs):
409 def _genslice(start, end, step):
410 starts = u'' if start == 0 else str(start)
411 ends = (u':%d' % (end+step)) if end + step >= 0 else u':'
412 steps = u'' if step == 1 else (u':%d' % step)
413 return u's[%s%s%s]' % (starts, ends, steps)
416 start = '(Never used)' # Quelch pyflakes warnings - start will be
417 # set as soon as step is set
418 for i, prev in zip(idxs[1:], idxs[:-1]):
422 yield _genslice(start, prev, step)
425 if i - prev in [-1, 1]:
430 yield u's[%d]' % prev
434 yield _genslice(start, i, step)
436 test_string = u''.join(map(compat_chr, range(slen)))
437 cache_res = func(test_string)
438 cache_spec = [ord(c) for c in cache_res]
439 expr_code = u' + '.join(gen_sig_code(cache_spec))
440 code = u'if len(s) == %d:\n return %s\n' % (slen, expr_code)
441 self.to_screen(u'Extracted signature function:\n' + code)
443 def _parse_sig_js(self, jscode):
444 funcname = self._search_regex(
445 r'signature=([$a-zA-Z]+)', jscode,
446 u'Initial JS player signature function name')
448 jsi = JSInterpreter(jscode)
449 initial_function = jsi.extract_function(funcname)
450 return lambda s: initial_function([s])
452 def _parse_sig_swf(self, file_contents):
453 swfi = SWFInterpreter(file_contents)
454 TARGET_CLASSNAME = u'SignatureDecipher'
455 searched_class = swfi.extract_class(TARGET_CLASSNAME)
456 initial_function = swfi.extract_function(searched_class, u'decipher')
457 return lambda s: initial_function([s])
459 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
460 """Turn the encrypted s field into a working signature"""
462 if player_url is None:
463 raise ExtractorError(u'Cannot decrypt signature without player_url')
465 if player_url.startswith(u'//'):
466 player_url = u'https:' + player_url
468 player_id = (player_url, len(s))
469 if player_id not in self._player_cache:
470 func = self._extract_signature_function(
471 video_id, player_url, len(s)
473 self._player_cache[player_id] = func
474 func = self._player_cache[player_id]
475 if self._downloader.params.get('youtube_print_sig_code'):
476 self._print_sig_code(func, len(s))
478 except Exception as e:
479 tb = traceback.format_exc()
480 raise ExtractorError(
481 u'Automatic signature extraction failed: ' + tb, cause=e)
483 def _get_available_subtitles(self, video_id, webpage):
485 sub_list = self._download_webpage(
486 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
487 video_id, note=False)
488 except ExtractorError as err:
489 self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
491 lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
496 params = compat_urllib_parse.urlencode({
499 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
500 'name': unescapeHTML(l[0]).encode('utf-8'),
502 url = u'https://www.youtube.com/api/timedtext?' + params
503 sub_lang_list[lang] = url
504 if not sub_lang_list:
505 self._downloader.report_warning(u'video doesn\'t have subtitles')
509 def _get_available_automatic_caption(self, video_id, webpage):
510 """We need the webpage for getting the captions url, pass it as an
511 argument to speed up the process."""
512 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
513 self.to_screen(u'%s: Looking for automatic captions' % video_id)
514 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
515 err_msg = u'Couldn\'t find automatic captions for %s' % video_id
517 self._downloader.report_warning(err_msg)
519 player_config = json.loads(mobj.group(1))
521 args = player_config[u'args']
522 caption_url = args[u'ttsurl']
523 timestamp = args[u'timestamp']
524 # We get the available subtitles
525 list_params = compat_urllib_parse.urlencode({
530 list_url = caption_url + '&' + list_params
531 caption_list = self._download_xml(list_url, video_id)
532 original_lang_node = caption_list.find('track')
533 if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' :
534 self._downloader.report_warning(u'Video doesn\'t have automatic captions')
536 original_lang = original_lang_node.attrib['lang_code']
539 for lang_node in caption_list.findall('target'):
540 sub_lang = lang_node.attrib['lang_code']
541 params = compat_urllib_parse.urlencode({
542 'lang': original_lang,
548 sub_lang_list[sub_lang] = caption_url + '&' + params
550 # An extractor error can be raise by the download process if there are
551 # no automatic captions but there are subtitles
552 except (KeyError, ExtractorError):
553 self._downloader.report_warning(err_msg)
557 def extract_id(cls, url):
558 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
560 raise ExtractorError(u'Invalid URL: %s' % url)
561 video_id = mobj.group(2)
564 def _extract_from_m3u8(self, manifest_url, video_id):
566 def _get_urls(_manifest):
567 lines = _manifest.split('\n')
568 urls = filter(lambda l: l and not l.startswith('#'),
571 manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest')
572 formats_urls = _get_urls(manifest)
573 for format_url in formats_urls:
574 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
575 url_map[itag] = format_url
578 def _extract_annotations(self, video_id):
579 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
580 return self._download_webpage(url, video_id, note=u'Searching for annotations.', errnote=u'Unable to download video annotations.')
582 def _real_extract(self, url):
584 u'http' if self._downloader.params.get('prefer_insecure', False)
587 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
588 mobj = re.search(self._NEXT_URL_RE, url)
590 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
591 video_id = self.extract_id(url)
594 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
595 video_webpage = self._download_webpage(url, video_id)
597 # Attempt to extract SWF player URL
598 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
600 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
605 self.report_video_info_webpage_download(video_id)
606 if re.search(r'player-age-gate-content">', video_webpage) is not None:
607 self.report_age_confirmation()
609 # We simulate the access to the video from www.youtube.com/v/{video_id}
610 # this can be viewed without login into Youtube
611 data = compat_urllib_parse.urlencode({
612 'video_id': video_id,
613 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
616 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
617 video_info_webpage = self._download_webpage(video_info_url, video_id,
619 errnote='unable to download video info webpage')
620 video_info = compat_parse_qs(video_info_webpage)
623 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
624 video_info_url = (proto + '://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
625 % (video_id, el_type))
626 video_info_webpage = self._download_webpage(video_info_url, video_id,
628 errnote='unable to download video info webpage')
629 video_info = compat_parse_qs(video_info_webpage)
630 if 'token' in video_info:
632 if 'token' not in video_info:
633 if 'reason' in video_info:
634 raise ExtractorError(
635 u'YouTube said: %s' % video_info['reason'][0],
636 expected=True, video_id=video_id)
638 raise ExtractorError(
639 u'"token" parameter not in video info for unknown reason',
642 if 'view_count' in video_info:
643 view_count = int(video_info['view_count'][0])
647 # Check for "rental" videos
648 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
649 raise ExtractorError(u'"rental" videos not supported')
651 # Start extracting information
652 self.report_information_extraction(video_id)
655 if 'author' not in video_info:
656 raise ExtractorError(u'Unable to extract uploader name')
657 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
660 video_uploader_id = None
661 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
663 video_uploader_id = mobj.group(1)
665 self._downloader.report_warning(u'unable to extract uploader nickname')
668 if 'title' in video_info:
669 video_title = video_info['title'][0]
671 self._downloader.report_warning(u'Unable to extract video title')
675 # We try first to get a high quality image:
676 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
677 video_webpage, re.DOTALL)
678 if m_thumb is not None:
679 video_thumbnail = m_thumb.group(1)
680 elif 'thumbnail_url' not in video_info:
681 self._downloader.report_warning(u'unable to extract video thumbnail')
682 video_thumbnail = None
683 else: # don't panic if we can't find it
684 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
688 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
691 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
694 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
695 upload_date = unified_strdate(upload_date)
697 m_cat_container = get_element_by_id("eow-category", video_webpage)
699 category = self._html_search_regex(
700 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
702 video_categories = None if category is None else [category]
704 video_categories = None
707 video_description = get_element_by_id("eow-description", video_webpage)
708 if video_description:
709 video_description = re.sub(r'''(?x)
711 (?:[a-zA-Z-]+="[^"]+"\s+)*?
713 (?:[a-zA-Z-]+="[^"]+"\s+)*?
714 class="yt-uix-redirect-link"\s*>
717 ''', r'\1', video_description)
718 video_description = clean_html(video_description)
720 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
722 video_description = unescapeHTML(fd_mobj.group(1))
724 video_description = u''
726 def _extract_count(klass):
727 count = self._search_regex(
728 r'class="%s">([\d,]+)</span>' % re.escape(klass),
729 video_webpage, klass, default=None)
730 if count is not None:
731 return int(count.replace(',', ''))
733 like_count = _extract_count(u'likes-count')
734 dislike_count = _extract_count(u'dislikes-count')
737 video_subtitles = self.extract_subtitles(video_id, video_webpage)
739 if self._downloader.params.get('listsubtitles', False):
740 self._list_available_subtitles(video_id, video_webpage)
743 if 'length_seconds' not in video_info:
744 self._downloader.report_warning(u'unable to extract video duration')
745 video_duration = None
747 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
750 video_annotations = None
751 if self._downloader.params.get('writeannotations', False):
752 video_annotations = self._extract_annotations(video_id)
754 # Decide which formats to download
756 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
758 raise ValueError('Could not find vevo ID')
759 json_code = uppercase_escape(mobj.group(1))
760 ytplayer_config = json.loads(json_code)
761 args = ytplayer_config['args']
762 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
763 # this signatures are encrypted
764 if 'url_encoded_fmt_stream_map' not in args:
765 raise ValueError(u'No stream_map present') # caught below
766 re_signature = re.compile(r'[&,]s=')
767 m_s = re_signature.search(args['url_encoded_fmt_stream_map'])
769 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
770 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
771 m_s = re_signature.search(args.get('adaptive_fmts', u''))
773 if 'adaptive_fmts' in video_info:
774 video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts']
776 video_info['adaptive_fmts'] = [args['adaptive_fmts']]
780 def _map_to_format_list(urlmap):
782 for itag, video_real_url in urlmap.items():
785 'url': video_real_url,
786 'player_url': player_url,
788 if itag in self._formats:
789 dct.update(self._formats[itag])
793 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
794 self.report_rtmp_download()
796 'format_id': '_rtmp',
798 'url': video_info['conn'][0],
799 'player_url': player_url,
801 elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1:
802 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0]
803 if 'rtmpe%3Dyes' in encoded_url_map:
804 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
806 for url_data_str in encoded_url_map.split(','):
807 url_data = compat_parse_qs(url_data_str)
808 if 'itag' in url_data and 'url' in url_data:
809 url = url_data['url'][0]
810 if 'sig' in url_data:
811 url += '&signature=' + url_data['sig'][0]
812 elif 's' in url_data:
813 encrypted_sig = url_data['s'][0]
816 jsplayer_url_json = self._search_regex(
817 r'"assets":.+?"js":\s*("[^"]+")',
818 video_webpage, u'JS player URL')
819 player_url = json.loads(jsplayer_url_json)
820 if player_url is None:
821 player_url_json = self._search_regex(
822 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
823 video_webpage, u'age gate player URL')
824 player_url = json.loads(player_url_json)
826 if self._downloader.params.get('verbose'):
827 if player_url is None:
828 player_version = 'unknown'
829 player_desc = 'unknown'
831 if player_url.endswith('swf'):
832 player_version = self._search_regex(
833 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
834 u'flash player', fatal=False)
835 player_desc = 'flash player %s' % player_version
837 player_version = self._search_regex(
838 r'html5player-(.+?)\.js', video_webpage,
839 'html5 player', fatal=False)
840 player_desc = u'html5 player %s' % player_version
842 parts_sizes = u'.'.join(compat_str(len(part)) for part in encrypted_sig.split('.'))
843 self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
844 (len(encrypted_sig), parts_sizes, url_data['itag'][0], player_desc))
846 signature = self._decrypt_signature(
847 encrypted_sig, video_id, player_url, age_gate)
848 url += '&signature=' + signature
849 if 'ratebypass' not in url:
850 url += '&ratebypass=yes'
851 url_map[url_data['itag'][0]] = url
852 formats = _map_to_format_list(url_map)
853 elif video_info.get('hlsvp'):
854 manifest_url = video_info['hlsvp'][0]
855 url_map = self._extract_from_m3u8(manifest_url, video_id)
856 formats = _map_to_format_list(url_map)
858 raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
860 # Look for the DASH manifest
861 if (self._downloader.params.get('youtube_include_dash_manifest', False)):
863 # The DASH manifest used needs to be the one from the original video_webpage.
864 # The one found in get_video_info seems to be using different signatures.
865 # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
866 # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
867 # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
869 dash_manifest_url = video_info.get('dashmpd')[0]
871 dash_manifest_url = ytplayer_config['args']['dashmpd']
872 def decrypt_sig(mobj):
874 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
875 return '/signature/%s' % dec_s
876 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
877 dash_doc = self._download_xml(
878 dash_manifest_url, video_id,
879 note=u'Downloading DASH manifest',
880 errnote=u'Could not download DASH manifest')
881 for r in dash_doc.findall(u'.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
882 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
885 format_id = r.attrib['id']
886 video_url = url_el.text
887 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
889 'format_id': format_id,
891 'width': int_or_none(r.attrib.get('width')),
892 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
893 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
894 'filesize': filesize,
897 existing_format = next(
899 if fo['format_id'] == format_id)
900 except StopIteration:
901 f.update(self._formats.get(format_id, {}))
904 existing_format.update(f)
906 except (ExtractorError, KeyError) as e:
907 self.report_warning(u'Skipping DASH manifest: %s' % e, video_id)
909 self._sort_formats(formats)
913 'uploader': video_uploader,
914 'uploader_id': video_uploader_id,
915 'upload_date': upload_date,
916 'title': video_title,
917 'thumbnail': video_thumbnail,
918 'description': video_description,
919 'categories': video_categories,
920 'subtitles': video_subtitles,
921 'duration': video_duration,
922 'age_limit': 18 if age_gate else 0,
923 'annotations': video_annotations,
924 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
925 'view_count': view_count,
926 'like_count': like_count,
927 'dislike_count': dislike_count,
931 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
932 IE_DESC = u'YouTube.com playlists'
933 _VALID_URL = r"""(?x)(?:
938 (?:course|view_play_list|my_playlists|artist|playlist|watch)
939 \? (?:.*?&)*? (?:p|a|list)=
943 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
944 # Top tracks, they can also include dots
949 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
951 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
952 _MORE_PAGES_INDICATOR = r'data-link-type="next"'
953 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
954 IE_NAME = u'youtube:playlist'
956 def _real_initialize(self):
959 def _ids_to_results(self, ids):
960 return [self.url_result(vid_id, 'Youtube', video_id=vid_id)
963 def _extract_mix(self, playlist_id):
964 # The mixes are generated from a a single video
965 # the id of the playlist is just 'RD' + video_id
966 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
967 webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix')
968 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
969 title_span = (search_title('playlist-title') or
970 search_title('title long-title') or search_title('title'))
971 title = clean_html(title_span)
972 video_re = r'''(?x)data-video-username=".*?".*?
973 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id)
974 ids = orderedSet(re.findall(video_re, webpage, flags=re.DOTALL))
975 url_results = self._ids_to_results(ids)
977 return self.playlist_result(url_results, playlist_id, title)
979 def _real_extract(self, url):
980 # Extract playlist id
981 mobj = re.match(self._VALID_URL, url)
983 raise ExtractorError(u'Invalid URL: %s' % url)
984 playlist_id = mobj.group(1) or mobj.group(2)
986 # Check if it's a video-specific URL
987 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
988 if 'v' in query_dict:
989 video_id = query_dict['v'][0]
990 if self._downloader.params.get('noplaylist'):
991 self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id)
992 return self.url_result(video_id, 'Youtube', video_id=video_id)
994 self.to_screen(u'Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
996 if playlist_id.startswith('RD'):
997 # Mixes require a custom extraction process
998 return self._extract_mix(playlist_id)
999 if playlist_id.startswith('TL'):
1000 raise ExtractorError(u'For downloading YouTube.com top lists, use '
1001 u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)
1003 url = self._TEMPLATE_URL % playlist_id
1004 page = self._download_webpage(url, playlist_id)
1005 more_widget_html = content_html = page
1007 # Check if the playlist exists or is private
1008 if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
1009 raise ExtractorError(
1010 u'The playlist doesn\'t exist or is private, use --username or '
1011 '--netrc to access it.',
1014 # Extract the video ids from the playlist pages
1017 for page_num in itertools.count(1):
1018 matches = re.finditer(self._VIDEO_RE, content_html)
1019 # We remove the duplicates and the link with index 0
1020 # (it's not the first video of the playlist)
1021 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1024 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1028 more = self._download_json(
1029 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1030 'Downloading page #%s' % page_num,
1031 transform_source=uppercase_escape)
1032 content_html = more['content_html']
1033 more_widget_html = more['load_more_widget_html']
1035 playlist_title = self._html_search_regex(
1036 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1039 url_results = self._ids_to_results(ids)
1040 return self.playlist_result(url_results, playlist_id, playlist_title)
1043 class YoutubeTopListIE(YoutubePlaylistIE):
1044 IE_NAME = u'youtube:toplist'
1045 IE_DESC = (u'YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1046 u' (Example: "yttoplist:music:Top Tracks")')
1047 _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1049 def _real_extract(self, url):
1050 mobj = re.match(self._VALID_URL, url)
1051 channel = mobj.group('chann')
1052 title = mobj.group('title')
1053 query = compat_urllib_parse.urlencode({'title': title})
1054 playlist_re = 'href="([^"]+?%s.*?)"' % re.escape(query)
1055 channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title)
1056 link = self._html_search_regex(playlist_re, channel_page, u'list')
1057 url = compat_urlparse.urljoin('https://www.youtube.com/', link)
1059 video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1061 # sometimes the webpage doesn't contain the videos
1062 # retry until we get them
1063 for i in itertools.count(0):
1064 msg = u'Downloading Youtube mix'
1066 msg += ', retry #%d' % i
1067 webpage = self._download_webpage(url, title, msg)
1068 ids = orderedSet(re.findall(video_re, webpage))
1071 url_results = self._ids_to_results(ids)
1072 return self.playlist_result(url_results, playlist_title=title)
1075 class YoutubeChannelIE(InfoExtractor):
1076 IE_DESC = u'YouTube.com channels'
1077 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
1078 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
1079 _MORE_PAGES_URL = 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
1080 IE_NAME = u'youtube:channel'
1082 def extract_videos_from_page(self, page):
1084 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
1085 if mobj.group(1) not in ids_in_page:
1086 ids_in_page.append(mobj.group(1))
1089 def _real_extract(self, url):
1090 # Extract channel id
1091 mobj = re.match(self._VALID_URL, url)
1093 raise ExtractorError(u'Invalid URL: %s' % url)
1095 # Download channel page
1096 channel_id = mobj.group(1)
1098 url = 'https://www.youtube.com/channel/%s/videos' % channel_id
1099 channel_page = self._download_webpage(url, channel_id)
1100 autogenerated = re.search(r'''(?x)
1102 channel-header-autogenerated-label|
1103 yt-channel-title-autogenerated
1104 )[^"]*"''', channel_page) is not None
1107 # The videos are contained in a single page
1108 # the ajax pages can't be used, they are empty
1109 video_ids = self.extract_videos_from_page(channel_page)
1111 # Download all channel pages using the json-based channel_ajax query
1112 for pagenum in itertools.count(1):
1113 url = self._MORE_PAGES_URL % (pagenum, channel_id)
1114 page = self._download_json(
1115 url, channel_id, note=u'Downloading page #%s' % pagenum,
1116 transform_source=uppercase_escape)
1118 ids_in_page = self.extract_videos_from_page(page['content_html'])
1119 video_ids.extend(ids_in_page)
1121 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
1124 self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
1126 url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id)
1127 for video_id in video_ids]
1128 return self.playlist_result(url_entries, channel_id)
1131 class YoutubeUserIE(InfoExtractor):
1132 IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
1133 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
1134 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
1135 _GDATA_PAGE_SIZE = 50
1136 _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1137 IE_NAME = u'youtube:user'
1140 def suitable(cls, url):
1141 # Don't return True if the url can be extracted with other youtube
1142 # extractor, the regex would is too permissive and it would match.
1143 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1144 if any(ie.suitable(url) for ie in other_ies): return False
1145 else: return super(YoutubeUserIE, cls).suitable(url)
1147 def _real_extract(self, url):
1149 mobj = re.match(self._VALID_URL, url)
1151 raise ExtractorError(u'Invalid URL: %s' % url)
1153 username = mobj.group(1)
1155 # Download video ids using YouTube Data API. Result size per
1156 # query is limited (currently to 50 videos) so we need to query
1157 # page by page until there are no video ids - it means we got
1160 def download_page(pagenum):
1161 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1163 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1164 page = self._download_webpage(
1165 gdata_url, username,
1166 u'Downloading video ids from %d to %d' % (
1167 start_index, start_index + self._GDATA_PAGE_SIZE))
1170 response = json.loads(page)
1171 except ValueError as err:
1172 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
1173 if 'entry' not in response['feed']:
1176 # Extract video identifiers
1177 entries = response['feed']['entry']
1178 for entry in entries:
1179 title = entry['title']['$t']
1180 video_id = entry['id']['$t'].split('/')[-1]
1184 'ie_key': 'Youtube',
1188 url_results = PagedList(download_page, self._GDATA_PAGE_SIZE)
1190 return self.playlist_result(url_results, playlist_title=username)
1193 class YoutubeSearchIE(SearchInfoExtractor):
1194 IE_DESC = u'YouTube.com searches'
1195 _API_URL = u'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1197 IE_NAME = u'youtube:search'
1198 _SEARCH_KEY = 'ytsearch'
1200 def _get_n_results(self, query, n):
1201 """Get a specified number of results for a query"""
1208 while (PAGE_SIZE * pagenum) < limit:
1209 result_url = self._API_URL % (
1210 compat_urllib_parse.quote_plus(query.encode('utf-8')),
1211 (PAGE_SIZE * pagenum) + 1)
1212 data_json = self._download_webpage(
1213 result_url, video_id=u'query "%s"' % query,
1214 note=u'Downloading page %s' % (pagenum + 1),
1215 errnote=u'Unable to download API page')
1216 data = json.loads(data_json)
1217 api_response = data['data']
1219 if 'items' not in api_response:
1220 raise ExtractorError(
1221 u'[youtube] No video results', expected=True)
1223 new_ids = list(video['id'] for video in api_response['items'])
1224 video_ids += new_ids
1226 limit = min(n, api_response['totalItems'])
1229 if len(video_ids) > n:
1230 video_ids = video_ids[:n]
1231 videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
1232 for video_id in video_ids]
1233 return self.playlist_result(videos, query)
1236 class YoutubeSearchDateIE(YoutubeSearchIE):
1237 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1238 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1239 _SEARCH_KEY = 'ytsearchdate'
1240 IE_DESC = u'YouTube.com searches, newest videos first'
1243 class YoutubeSearchURLIE(InfoExtractor):
1244 IE_DESC = u'YouTube.com search URLs'
1245 IE_NAME = u'youtube:search_url'
1246 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1248 def _real_extract(self, url):
1249 mobj = re.match(self._VALID_URL, url)
1250 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1252 webpage = self._download_webpage(url, query)
1253 result_code = self._search_regex(
1254 r'(?s)<ol class="item-section"(.*?)</ol>', webpage, u'result HTML')
1256 part_codes = re.findall(
1257 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1259 for part_code in part_codes:
1260 part_title = self._html_search_regex(
1261 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1262 part_url_snippet = self._html_search_regex(
1263 r'(?s)href="([^"]+)"', part_code, 'item URL')
1264 part_url = compat_urlparse.urljoin(
1265 'https://www.youtube.com/', part_url_snippet)
1269 'title': part_title,
1273 '_type': 'playlist',
1279 class YoutubeShowIE(InfoExtractor):
1280 IE_DESC = u'YouTube.com (multi-season) shows'
1281 _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
1282 IE_NAME = u'youtube:show'
1284 def _real_extract(self, url):
1285 mobj = re.match(self._VALID_URL, url)
1286 show_name = mobj.group(1)
1287 webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
1288 # There's one playlist for each season of the show
1289 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1290 self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
1291 return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
1294 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1296 Base class for extractors that fetch info from
1297 http://www.youtube.com/feed_ajax
1298 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1300 _LOGIN_REQUIRED = True
1301 # use action_load_personal_feed instead of action_load_system_feed
1302 _PERSONAL_FEED = False
1305 def _FEED_TEMPLATE(self):
1306 action = 'action_load_system_feed'
1307 if self._PERSONAL_FEED:
1308 action = 'action_load_personal_feed'
1309 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1313 return u'youtube:%s' % self._FEED_NAME
1315 def _real_initialize(self):
1318 def _real_extract(self, url):
1321 for i in itertools.count(1):
1322 info = self._download_json(self._FEED_TEMPLATE % paging,
1323 u'%s feed' % self._FEED_NAME,
1324 u'Downloading page %s' % i)
1325 feed_html = info.get('feed_html') or info.get('content_html')
1326 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1327 ids = orderedSet(m.group(1) for m in m_ids)
1328 feed_entries.extend(
1329 self.url_result(video_id, 'Youtube', video_id=video_id)
1330 for video_id in ids)
1332 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1336 paging = mobj.group('paging')
1337 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1339 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1340 IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1341 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1342 _FEED_NAME = 'subscriptions'
1343 _PLAYLIST_TITLE = u'Youtube Subscriptions'
1345 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1346 IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1347 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1348 _FEED_NAME = 'recommended'
1349 _PLAYLIST_TITLE = u'Youtube Recommended videos'
1351 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1352 IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1353 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1354 _FEED_NAME = 'watch_later'
1355 _PLAYLIST_TITLE = u'Youtube Watch Later'
1356 _PERSONAL_FEED = True
1358 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1359 IE_DESC = u'Youtube watch history, "ythistory" keyword (requires authentication)'
1360 _VALID_URL = u'https?://www\.youtube\.com/feed/history|:ythistory'
1361 _FEED_NAME = 'history'
1362 _PERSONAL_FEED = True
1363 _PLAYLIST_TITLE = u'Youtube Watch History'
1365 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1366 IE_NAME = u'youtube:favorites'
1367 IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1368 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1369 _LOGIN_REQUIRED = True
1371 def _real_extract(self, url):
1372 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1373 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id')
1374 return self.url_result(playlist_id, 'YoutubePlaylist')
1377 class YoutubeTruncatedURLIE(InfoExtractor):
1378 IE_NAME = 'youtube:truncated_url'
1379 IE_DESC = False # Do not list
1380 _VALID_URL = r'''(?x)
1381 (?:https?://)?[^/]+/watch\?(?:
1383 annotation_id=annotation_[^&]+
1385 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1389 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1390 'only_matching': True,
1392 'url': 'http://www.youtube.com/watch?',
1393 'only_matching': True,
1396 def _real_extract(self, url):
1397 raise ExtractorError(
1398 u'Did you forget to quote the URL? Remember that & is a meta '
1399 u'character in most shells, so you want to put the URL in quotes, '
1401 u'"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1402 u' or simply youtube-dl BaW_jenozKc .',