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 _TWOFACTOR_URL
= 'https://accounts.google.com/SecondFactor'
41 _LANG_URL
= r
'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
42 _AGE_URL
= 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
43 _NETRC_MACHINE
= 'youtube'
44 # If True it will raise an error if no login info is provided
45 _LOGIN_REQUIRED
= False
47 def _set_language(self
):
48 return bool(self
._download
_webpage
(
50 note
=u
'Setting language', errnote
='unable to set language',
55 Attempt to log in to YouTube.
56 True is returned if successful or skipped.
57 False is returned if login failed.
59 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
61 (username
, password
) = self
._get
_login
_info
()
62 # No authentication to be performed
64 if self
._LOGIN
_REQUIRED
:
65 raise ExtractorError(u
'No login info available, needed for using %s.' % self
.IE_NAME
, expected
=True)
68 login_page
= self
._download
_webpage
(
69 self
._LOGIN
_URL
, None,
70 note
=u
'Downloading login page',
71 errnote
=u
'unable to fetch login page', fatal
=False)
72 if login_page
is False:
75 galx
= self
._search
_regex
(r
'(?s)<input.+?name="GALX".+?value="(.+?)"',
76 login_page
, u
'Login GALX parameter')
80 u
'continue': u
'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
85 u
'PersistentCookie': u
'yes',
87 u
'bgresponse': u
'js_disabled',
88 u
'checkConnection': u
'',
89 u
'checkedDomains': u
'youtube',
94 u
'signIn': u
'Sign in',
96 u
'service': u
'youtube',
101 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
103 login_form
= dict((k
.encode('utf-8'), v
.encode('utf-8')) for k
,v
in login_form_strs
.items())
104 login_data
= compat_urllib_parse
.urlencode(login_form
).encode('ascii')
106 req
= compat_urllib_request
.Request(self
._LOGIN
_URL
, login_data
)
107 login_results
= self
._download
_webpage
(
109 note
=u
'Logging in', errnote
=u
'unable to log in', fatal
=False)
110 if login_results
is False:
113 if re
.search(r
'id="errormsg_0_Passwd"', login_results
) is not None:
114 raise ExtractorError(u
'Please use your account password and a two-factor code instead of an application-specific password.', expected
=True)
117 # TODO add SMS and phone call support - these require making a request and then prompting the user
119 if re
.search(r
'(?i)<form[^>]* id="gaia_secondfactorform"', login_results
) is not None:
120 tfa_code
= self
._get
_tfa
_info
()
123 self
._downloader
.report_warning(u
'Two-factor authentication required. Provide it with --twofactor <code>')
124 self
._downloader
.report_warning(u
'(Note that only TOTP (Google Authenticator App) codes work at this time.)')
127 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
129 match
= re
.search(r
'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
131 self._downloader.report_warning(u'Failed to get secTok
- did the page structure change?
')
132 secTok = match.group(1)
133 match = re.search(r'id="timeStmp"\n\s
+value
=\'(.+)\'/>', login_results, re.M | re.U)
135 self._downloader.report_warning(u'Failed to get timeStmp
- did the page structure change?
')
136 timeStmp = match.group(1)
139 u'continue': u'https
://www
.youtube
.com
/signin?action_handle_signin
=true
&feature
=sign_in_button
&hl
=en_US
&nomobiletemp
=1',
141 u'smsUserPin
': tfa_code,
142 u'smsVerifyPin
': u'Verify
',
144 u'PersistentCookie
': u'yes
',
145 u'checkConnection
': u'',
146 u'checkedDomains
': u'youtube
',
149 u'timeStmp
': timeStmp,
150 u'service
': u'youtube
',
153 tfa_form = dict((k.encode('utf
-8'), v.encode('utf
-8')) for k,v in tfa_form_strs.items())
154 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii
')
156 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
157 tfa_results = self._download_webpage(
159 note=u'Submitting TFA code
', errnote=u'unable to submit tfa
', fatal=False)
161 if tfa_results is False:
164 if re.search(r'(?i
)<form
[^
>]* id="gaia_secondfactorform"', tfa_results) is not None:
165 self._downloader.report_warning(u'Two
-factor code expired
. Please
try again
, or use a one
-use backup code instead
.')
167 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', tfa_results) is not None:
168 self._downloader.report_warning(u'unable to log
in - did the page structure change?
')
170 if re.search(r'smsauth
-interstitial
-reviewsettings
', tfa_results) is not None:
171 self._downloader.report_warning(u'Your Google account has a security notice
. Please log
in on your web browser
, resolve the notice
, and try again
.')
174 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', login_results) is not None:
175 self._downloader.report_warning(u'unable to log
in: bad username
or password
')
179 def _confirm_age(self):
182 'action_confirm
': 'Confirm
',
184 req = compat_urllib_request.Request(self._AGE_URL,
185 compat_urllib_parse.urlencode(age_form).encode('ascii
'))
187 self._download_webpage(
189 note=u'Confirming age
', errnote=u'Unable to confirm age
')
192 def _real_initialize(self):
193 if self._downloader is None:
195 if not self._set_language():
197 if not self._login():
202 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
203 IE_DESC = u'YouTube
.com
'
204 _VALID_URL = r"""(?x)^
206 (?:https?://|//)? # http(s):// or protocol-independent URL (optional)
207 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
208 (?:www\.)?deturl\.com/www\.youtube\.com/|
209 (?:www\.)?pwnyoutube\.com/|
210 (?:www\.)?yourepeat\.com/|
211 tube\.majestyc\.net/|
212 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
213 (?:.*?\#/)? # handle anchor (#/) redirect urls
214 (?: # the various things that can precede the ID:
215 (?:(?:v|embed|e)/) # v/ or embed/ or e/
216 |(?: # or the v= param in all its forms
217 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
218 (?:\?|\#!?) # the params delimiter ? or # or #!
219 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
223 |youtu\.be/ # just youtu.be/xxxx
224 |https?://(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
226 )? # all until now is optional -> you can pass the naked ID
227 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
228 (?(1).+)? # if we found the ID, everything can follow
230 _NEXT_URL_RE = r'[\?&]next_url
=([^
&]+)'
232 '5': {'ext': 'flv', 'width': 400, 'height': 240},
233 '6': {'ext': 'flv', 'width': 450, 'height': 270},
234 '13': {'ext': '3gp'},
235 '17': {'ext': '3gp', 'width': 176, 'height': 144},
236 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
237 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
238 '34': {'ext': 'flv', 'width': 640, 'height': 360},
239 '35': {'ext': 'flv', 'width': 854, 'height': 480},
240 '36': {'ext': '3gp', 'width': 320, 'height': 240},
241 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
242 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
243 '43': {'ext': 'webm', 'width': 640, 'height': 360},
244 '44': {'ext': 'webm', 'width': 854, 'height': 480},
245 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
246 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
250 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
251 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
252 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
253 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
254 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
255 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
256 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
258 # Apple HTTP Live Streaming
259 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
260 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
261 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
262 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
263 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
264 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
265 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
268 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
269 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
270 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
271 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
272 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
273 '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
274 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
275 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
278 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
279 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
280 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
283 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
284 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
285 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
286 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
287 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
288 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
289 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
290 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
291 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
293 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
300 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
301 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
304 '_rtmp
': {'protocol': 'rtmp'},
310 u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc",
311 u"file": u"BaW_jenozKc.mp4",
313 u"title": u"youtube-dl test video \"'/\\ä↭𝕐
",
314 u"uploader
": u"Philipp Hagemeister
",
315 u"uploader_id
": u"phihag
",
316 u"upload_date
": u"20121002",
317 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 .",
318 u"categories": [u'Science
& Technology
'],
320 'dislike_count
': int,
324 u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY",
325 u"file": u"UxxajLWwzqY.mp4",
326 u"note": u"Test generic use_cipher_signature video (#897)",
328 u"upload_date": u"20120506",
329 u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
330 u"description": u"md5:fea86fda2d5a5784273df5c7cc994d9f",
331 u"uploader": u"Icona Pop",
332 u"uploader_id": u"IconaPop"
336 u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ",
337 u"file": u"07FYdnEawAQ.mp4",
338 u"note": u"Test VEVO video with age protection (#956)",
340 u"upload_date": u"20130703",
341 u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
342 u"description": u"md5:64249768eec3bc4276236606ea996373",
343 u"uploader": u"justintimberlakeVEVO",
344 u"uploader_id": u"justintimberlakeVEVO"
348 u"url": u"//www.YouTube.com/watch?v=yZIXLfi8CZQ",
349 u"file": u"yZIXLfi8CZQ.mp4",
350 u"note": u"Embed-only video (#1746)",
352 u"upload_date": u"20120608",
353 u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012",
354 u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7",
355 u"uploader": u"SET India",
356 u"uploader_id": u"setindia"
360 u"url": u"http://www.youtube.com/watch?v=a9LDPn-MO4I",
361 u"file": u"a9LDPn-MO4I.m4a",
362 u"note": u"256k DASH audio (format 141) via DASH manifest",
364 u"upload_date": "20121002",
365 u"uploader_id": "8KVIDEO",
366 u"description": "No description available.",
367 u"uploader": "8KVIDEO",
368 u"title": "UHDTV TEST 8K VIDEO.mp4"
371 u"youtube_include_dash_manifest": True,
375 # DASH manifest with encrypted signature
377 u'url
': u'https
://www
.youtube
.com
/watch?v
=IB3lcPjvWLA
',
379 u'id': u'IB3lcPjvWLA
',
381 u'title
': u'Afrojack
- The Spark ft
. Spree Wilson
',
382 u'description
': u'md5
:9717375db5a9a3992be4668bbf3bc0a8
',
383 u'uploader
': u'AfrojackVEVO
',
384 u'uploader_id
': u'AfrojackVEVO
',
385 u'upload_date
': u'20131011',
388 u'youtube_include_dash_manifest
': True,
396 def suitable(cls, url):
397 """Receives a URL and returns True if suitable for this IE."""
398 if YoutubePlaylistIE.suitable(url): return False
399 return re.match(cls._VALID_URL, url) is not None
401 def __init__(self, *args, **kwargs):
402 super(YoutubeIE, self).__init__(*args, **kwargs)
403 self._player_cache = {}
405 def report_video_info_webpage_download(self, video_id):
406 """Report attempt to download video info webpage."""
407 self.to_screen(u'%s: Downloading video info webpage
' % video_id)
409 def report_information_extraction(self, video_id):
410 """Report attempt to extract video information."""
411 self.to_screen(u'%s: Extracting video information
' % video_id)
413 def report_unavailable_format(self, video_id, format):
414 """Report extracted video URL."""
415 self.to_screen(u'%s: Format
%s not available
' % (video_id, format))
417 def report_rtmp_download(self):
418 """Indicate the download will use the RTMP protocol."""
419 self.to_screen(u'RTMP download detected
')
421 def _signature_cache_id(self, example_sig):
422 """ Return a string representation of a signature """
423 return u'.'.join(compat_str(len(part)) for part in example_sig.split('.'))
425 def _extract_signature_function(self, video_id, player_url, example_sig):
427 r'.*-(?P
<id>[a
-zA
-Z0
-9_-]+)(?
:/watch_as3|
/html5player
)?\
.(?P
<ext
>[a
-z
]+)$
',
430 raise ExtractorError('Cannot identify player
%r' % player_url)
431 player_type = id_m.group('ext
')
432 player_id = id_m.group('id')
434 # Read from filesystem cache
435 func_id = '%s_%s_%s' % (
436 player_type, player_id, self._signature_cache_id(example_sig))
437 assert os.path.basename(func_id) == func_id
438 cache_dir = get_cachedir(self._downloader.params)
440 cache_enabled = cache_dir is not None
442 cache_fn = os.path.join(os.path.expanduser(cache_dir),
446 with io.open(cache_fn, 'r
', encoding='utf
-8') as cachef:
447 cache_spec = json.load(cachef)
448 return lambda s: u''.join(s[i] for i in cache_spec)
450 pass # No cache available
453 file_size = os.path.getsize(cache_fn)
454 except (OSError, IOError) as oe:
456 self._downloader.report_warning(
457 u'Cache
%s failed (%s)' % (cache_fn, file_size))
459 if player_type == 'js
':
460 code = self._download_webpage(
461 player_url, video_id,
462 note=u'Downloading
%s player
%s' % (player_type, player_id),
463 errnote=u'Download of
%s failed
' % player_url)
464 res = self._parse_sig_js(code)
465 elif player_type == 'swf
':
466 urlh = self._request_webpage(
467 player_url, video_id,
468 note=u'Downloading
%s player
%s' % (player_type, player_id),
469 errnote=u'Download of
%s failed
' % player_url)
471 res = self._parse_sig_swf(code)
473 assert False, 'Invalid player
type %r' % player_type
477 test_string = u''.join(map(compat_chr, range(len(example_sig))))
478 cache_res = res(test_string)
479 cache_spec = [ord(c) for c in cache_res]
481 os.makedirs(os.path.dirname(cache_fn))
482 except OSError as ose:
483 if ose.errno != errno.EEXIST:
485 write_json_file(cache_spec, cache_fn)
487 tb = traceback.format_exc()
488 self._downloader.report_warning(
489 u'Writing cache to
%r failed
: %s' % (cache_fn, tb))
493 def _print_sig_code(self, func, example_sig):
494 def gen_sig_code(idxs):
495 def _genslice(start, end, step):
496 starts = u'' if start == 0 else str(start)
497 ends = (u':%d' % (end+step)) if end + step >= 0 else u':'
498 steps = u'' if step == 1 else (u':%d' % step)
499 return u's
[%s%s%s]' % (starts, ends, steps)
502 start = '(Never used
)' # Quelch pyflakes warnings - start will be
503 # set as soon as step is set
504 for i, prev in zip(idxs[1:], idxs[:-1]):
508 yield _genslice(start, prev, step)
511 if i - prev in [-1, 1]:
516 yield u's
[%d]' % prev
520 yield _genslice(start, i, step)
522 test_string = u''.join(map(compat_chr, range(len(example_sig))))
523 cache_res = func(test_string)
524 cache_spec = [ord(c) for c in cache_res]
525 expr_code = u' + '.join(gen_sig_code(cache_spec))
526 signature_id_tuple = '(%s)' % (
527 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
528 code = (u'if tuple(len(p
) for p
in s
.split(\'.\')) == %s:\n'
529 u' return %s\n') % (signature_id_tuple, expr_code)
530 self.to_screen(u'Extracted signature function
:\n' + code)
532 def _parse_sig_js(self, jscode):
533 funcname = self._search_regex(
534 r'signature
=([$a
-zA
-Z
]+)', jscode,
535 u'Initial JS player signature function name
')
537 jsi = JSInterpreter(jscode)
538 initial_function = jsi.extract_function(funcname)
539 return lambda s: initial_function([s])
541 def _parse_sig_swf(self, file_contents):
542 swfi = SWFInterpreter(file_contents)
543 TARGET_CLASSNAME = u'SignatureDecipher
'
544 searched_class = swfi.extract_class(TARGET_CLASSNAME)
545 initial_function = swfi.extract_function(searched_class, u'decipher
')
546 return lambda s: initial_function([s])
548 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
549 """Turn the encrypted s field into a working signature"""
551 if player_url is None:
552 raise ExtractorError(u'Cannot decrypt signature without player_url
')
554 if player_url.startswith(u'//'):
555 player_url = u'https
:' + player_url
557 player_id = (player_url, self._signature_cache_id(s))
558 if player_id not in self._player_cache:
559 func = self._extract_signature_function(
560 video_id, player_url, s
562 self._player_cache[player_id] = func
563 func = self._player_cache[player_id]
564 if self._downloader.params.get('youtube_print_sig_code
'):
565 self._print_sig_code(func, s)
567 except Exception as e:
568 tb = traceback.format_exc()
569 raise ExtractorError(
570 u'Signature extraction failed
: ' + tb, cause=e)
572 def _get_available_subtitles(self, video_id, webpage):
574 sub_list = self._download_webpage(
575 'https
://video
.google
.com
/timedtext?hl
=en
&type=list&v
=%s' % video_id,
576 video_id, note=False)
577 except ExtractorError as err:
578 self._downloader.report_warning(u'unable to download video subtitles
: %s' % compat_str(err))
580 lang_list = re.findall(r'name
="([^"]*)"[^>]+lang_code="([\w\
-]+)"', sub_list)
585 if lang in sub_lang_list:
587 params = compat_urllib_parse.urlencode({
590 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
591 'name': unescapeHTML(l[0]).encode('utf-8'),
593 url = u'https://www.youtube.com/api/timedtext?' + params
594 sub_lang_list[lang] = url
595 if not sub_lang_list:
596 self._downloader.report_warning(u'video doesn\'t have subtitles')
600 def _get_available_automatic_caption(self, video_id, webpage):
601 """We need the webpage for getting the captions url, pass it as an
602 argument to speed up the process."""
603 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
604 self.to_screen(u'%s: Looking for automatic captions' % video_id)
605 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
606 err_msg = u'Couldn\'t find automatic captions for %s' % video_id
608 self._downloader.report_warning(err_msg)
610 player_config = json.loads(mobj.group(1))
612 args = player_config[u'args']
613 caption_url = args[u'ttsurl']
614 timestamp = args[u'timestamp']
615 # We get the available subtitles
616 list_params = compat_urllib_parse.urlencode({
621 list_url = caption_url + '&' + list_params
622 caption_list = self._download_xml(list_url, video_id)
623 original_lang_node = caption_list.find('track')
624 if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' :
625 self._downloader.report_warning(u'Video doesn\'t have automatic captions')
627 original_lang = original_lang_node.attrib['lang_code']
630 for lang_node in caption_list.findall('target'):
631 sub_lang = lang_node.attrib['lang_code']
632 params = compat_urllib_parse.urlencode({
633 'lang': original_lang,
639 sub_lang_list[sub_lang] = caption_url + '&' + params
641 # An extractor error can be raise by the download process if there are
642 # no automatic captions but there are subtitles
643 except (KeyError, ExtractorError):
644 self._downloader.report_warning(err_msg)
648 def extract_id(cls, url):
649 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
651 raise ExtractorError(u'Invalid URL: %s' % url)
652 video_id = mobj.group(2)
655 def _extract_from_m3u8(self, manifest_url, video_id):
657 def _get_urls(_manifest):
658 lines = _manifest.split('\n')
659 urls = filter(lambda l: l and not l.startswith('#'),
662 manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest')
663 formats_urls = _get_urls(manifest)
664 for format_url in formats_urls:
665 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
666 url_map[itag] = format_url
669 def _extract_annotations(self, video_id):
670 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
671 return self._download_webpage(url, video_id, note=u'Searching for annotations.', errnote=u'Unable to download video annotations.')
673 def _real_extract(self, url):
675 u'http' if self._downloader.params.get('prefer_insecure', False)
678 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
679 mobj = re.search(self._NEXT_URL_RE, url)
681 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
682 video_id = self.extract_id(url)
685 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
686 video_webpage = self._download_webpage(url, video_id)
688 # Attempt to extract SWF player URL
689 mobj = re.search(r'swfConfig.*?"(https?
:\\/\\/.*?watch
.*?
-.*?\
.swf
)"', video_webpage)
691 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
696 self.report_video_info_webpage_download(video_id)
697 if re.search(r'player-age-gate-content">', video_webpage) is not None:
698 self.report_age_confirmation()
700 # We simulate the access to the video from www.youtube.com/v/{video_id}
701 # this can be viewed without login into Youtube
702 data = compat_urllib_parse.urlencode({
703 'video_id
': video_id,
704 'eurl
': 'https
://youtube
.googleapis
.com
/v
/' + video_id,
705 'sts
': self._search_regex(
706 r'"sts"\s
*:\s
*(\d
+)', video_webpage, 'sts
'),
708 video_info_url = proto + '://www
.youtube
.com
/get_video_info?
' + data
709 video_info_webpage = self._download_webpage(video_info_url, video_id,
711 errnote='unable to download video info webpage
')
712 video_info = compat_parse_qs(video_info_webpage)
715 for el_type in ['&el
=embedded
', '&el
=detailpage
', '&el
=vevo
', '']:
716 video_info_url = (proto + '://www
.youtube
.com
/get_video_info?
&video_id
=%s%s&ps
=default
&eurl
=&gl
=US
&hl
=en
'
717 % (video_id, el_type))
718 video_info_webpage = self._download_webpage(video_info_url, video_id,
720 errnote='unable to download video info webpage
')
721 video_info = compat_parse_qs(video_info_webpage)
722 if 'token
' in video_info:
724 if 'token
' not in video_info:
725 if 'reason
' in video_info:
726 raise ExtractorError(
727 u'YouTube said
: %s' % video_info['reason
'][0],
728 expected=True, video_id=video_id)
730 raise ExtractorError(
731 u'"token" parameter
not in video info
for unknown reason
',
734 if 'view_count
' in video_info:
735 view_count = int(video_info['view_count
'][0])
739 # Check for "rental" videos
740 if 'ypc_video_rental_bar_text
' in video_info and 'author
' not in video_info:
741 raise ExtractorError(u'"rental" videos
not supported
')
743 # Start extracting information
744 self.report_information_extraction(video_id)
747 if 'author
' not in video_info:
748 raise ExtractorError(u'Unable to extract uploader name
')
749 video_uploader = compat_urllib_parse.unquote_plus(video_info['author
'][0])
752 video_uploader_id = None
753 mobj = re.search(r'<link itemprop
="url" href
="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
755 video_uploader_id = mobj.group(1)
757 self._downloader.report_warning(u'unable to extract uploader nickname')
760 if 'title' in video_info:
761 video_title = video_info['title'][0]
763 self._downloader.report_warning(u'Unable to extract video title')
767 # We try first to get a high quality image:
768 m_thumb = re.search(r'<span itemprop="thumbnail
".*?href="(.*?
)">',
769 video_webpage, re.DOTALL)
770 if m_thumb is not None:
771 video_thumbnail = m_thumb.group(1)
772 elif 'thumbnail_url' not in video_info:
773 self._downloader.report_warning(u'unable to extract video thumbnail')
774 video_thumbnail = None
775 else: # don't panic if we can't find it
776 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
780 mobj = re.search(r'(?s)id="eow
-date
.*?
>(.*?
)</span
>', video_webpage)
783 r'(?s
)id="watch-uploader-info".*?
>.*?
(?
:Published|Uploaded|Streamed live
) on (.*?
)</strong
>',
786 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
787 upload_date = unified_strdate(upload_date)
789 m_cat_container = self._search_regex(
790 r'(?s
)<h4
[^
>]*>\s
*Category\s
*</h4
>\s
*<ul
[^
>]*>(.*?
)</ul
>',
791 video_webpage, 'categories
', fatal=False)
793 category = self._html_search_regex(
794 r'(?s
)<a
[^
<]+>(.*?
)</a
>', m_cat_container, 'category
',
796 video_categories = None if category is None else [category]
798 video_categories = None
801 video_description = get_element_by_id("eow-description", video_webpage)
802 if video_description:
803 video_description = re.sub(r'''(?x)
805 (?:[a-zA-Z-]+="[^"]+"\s+)*?
807 (?:[a-zA-Z-]+="[^"]+"\s+)*?
808 class="yt-uix-redirect-link"\s*>
811 ''', r'\
1', video_description)
812 video_description = clean_html(video_description)
814 fd_mobj = re.search(r'<meta name
="description" content
="([^"]+)"', video_webpage)
816 video_description = unescapeHTML(fd_mobj.group(1))
818 video_description = u''
820 def _extract_count(count_name):
821 count = self._search_regex(
822 r'id="watch
-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
823 video_webpage, count_name, default=None)
824 if count is not None:
825 return int(count.replace(',', ''))
827 like_count = _extract_count(u'like')
828 dislike_count = _extract_count(u'dislike')
831 video_subtitles = self.extract_subtitles(video_id, video_webpage)
833 if self._downloader.params.get('listsubtitles', False):
834 self._list_available_subtitles(video_id, video_webpage)
837 if 'length_seconds' not in video_info:
838 self._downloader.report_warning(u'unable to extract video duration')
839 video_duration = None
841 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
844 video_annotations = None
845 if self._downloader.params.get('writeannotations', False):
846 video_annotations = self._extract_annotations(video_id)
848 # Decide which formats to download
850 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
852 raise ValueError('Could not find vevo ID')
853 json_code = uppercase_escape(mobj.group(1))
854 ytplayer_config = json.loads(json_code)
855 args = ytplayer_config['args']
856 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
857 # this signatures are encrypted
858 if 'url_encoded_fmt_stream_map' not in args:
859 raise ValueError(u'No stream_map present') # caught below
860 re_signature = re.compile(r'[&,]s=')
861 m_s = re_signature.search(args['url_encoded_fmt_stream_map'])
863 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
864 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
865 m_s = re_signature.search(args.get('adaptive_fmts', u''))
867 if 'adaptive_fmts' in video_info:
868 video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts']
870 video_info['adaptive_fmts'] = [args['adaptive_fmts']]
874 def _map_to_format_list(urlmap):
876 for itag, video_real_url in urlmap.items():
879 'url': video_real_url,
880 'player_url': player_url,
882 if itag in self._formats:
883 dct.update(self._formats[itag])
887 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
888 self.report_rtmp_download()
890 'format_id': '_rtmp',
892 'url': video_info['conn'][0],
893 'player_url': player_url,
895 elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1:
896 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0]
897 if 'rtmpe%3Dyes' in encoded_url_map:
898 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
900 for url_data_str in encoded_url_map.split(','):
901 url_data = compat_parse_qs(url_data_str)
902 if 'itag' not in url_data or 'url' not in url_data:
904 format_id = url_data['itag'][0]
905 url = url_data['url'][0]
907 if 'sig' in url_data:
908 url += '&signature=' + url_data['sig'][0]
909 elif 's' in url_data:
910 encrypted_sig = url_data['s'][0]
913 jsplayer_url_json = self._search_regex(
914 r'"assets
":.+?"js
":\s*("[^
"]+")',
915 video_webpage, u'JS player URL
')
916 player_url = json.loads(jsplayer_url_json)
917 if player_url is None:
918 player_url_json = self._search_regex(
919 r'ytplayer\
.config
.*?
"url"\s
*:\s
*("[^"]+")',
920 video_webpage, u'age gate player URL')
921 player_url = json.loads(player_url_json)
923 if self._downloader.params.get('verbose'):
924 if player_url is None:
925 player_version = 'unknown'
926 player_desc = 'unknown'
928 if player_url.endswith('swf'):
929 player_version = self._search_regex(
930 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
931 u'flash player', fatal=False)
932 player_desc = 'flash player %s' % player_version
934 player_version = self._search_regex(
935 r'html5player-([^/]+?)(?:/html5player)?\.js',
937 'html5 player', fatal=False)
938 player_desc = u'html5 player %s' % player_version
940 parts_sizes = self._signature_cache_id(encrypted_sig)
941 self.to_screen(u'{%s} signature length %s, %s' %
942 (format_id, parts_sizes, player_desc))
944 signature = self._decrypt_signature(
945 encrypted_sig, video_id, player_url, age_gate)
946 url += '&signature=' + signature
947 if 'ratebypass' not in url:
948 url += '&ratebypass=yes'
949 url_map[format_id] = url
950 formats = _map_to_format_list(url_map)
951 elif video_info.get('hlsvp'):
952 manifest_url = video_info['hlsvp'][0]
953 url_map = self._extract_from_m3u8(manifest_url, video_id)
954 formats = _map_to_format_list(url_map)
956 raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
958 # Look for the DASH manifest
959 if (self._downloader.params.get('youtube_include_dash_manifest', False)):
961 # The DASH manifest used needs to be the one from the original video_webpage.
962 # The one found in get_video_info seems to be using different signatures.
963 # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
964 # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
965 # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
967 dash_manifest_url = video_info.get('dashmpd')[0]
969 dash_manifest_url = ytplayer_config['args']['dashmpd']
970 def decrypt_sig(mobj):
972 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
973 return '/signature/%s' % dec_s
974 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
975 dash_doc = self._download_xml(
976 dash_manifest_url, video_id,
977 note=u'Downloading DASH manifest',
978 errnote=u'Could not download DASH manifest')
979 for r in dash_doc.findall(u'.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
980 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
983 format_id = r.attrib['id']
984 video_url = url_el.text
985 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
987 'format_id': format_id,
989 'width': int_or_none(r.attrib.get('width')),
990 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
991 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
992 'filesize': filesize,
995 existing_format = next(
997 if fo['format_id'] == format_id)
998 except StopIteration:
999 f.update(self._formats.get(format_id, {}))
1002 existing_format.update(f)
1004 except (ExtractorError, KeyError) as e:
1005 self.report_warning(u'Skipping DASH manifest: %s' % e, video_id)
1007 self._sort_formats(formats)
1011 'uploader': video_uploader,
1012 'uploader_id': video_uploader_id,
1013 'upload_date': upload_date,
1014 'title': video_title,
1015 'thumbnail': video_thumbnail,
1016 'description': video_description,
1017 'categories': video_categories,
1018 'subtitles': video_subtitles,
1019 'duration': video_duration,
1020 'age_limit': 18 if age_gate else 0,
1021 'annotations': video_annotations,
1022 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1023 'view_count': view_count,
1024 'like_count': like_count,
1025 'dislike_count': dislike_count,
1029 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1030 IE_DESC = u'YouTube.com playlists'
1031 _VALID_URL = r"""(?x)(?:
1036 (?:course|view_play_list|my_playlists|artist|playlist|watch)
1037 \? (?:.*?&)*? (?:p|a|list)=
1041 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1042 # Top tracks, they can also include dots
1047 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1049 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1050 _MORE_PAGES_INDICATOR = r'data-link-type="next
"'
1051 _VIDEO_RE = r'href="\s
*/watch
\?v
=(?P
<id>[0-9A
-Za
-z_
-]{11}
)&
;[^
"]*?index=(?P<index>\d+)'
1052 IE_NAME = u'youtube:playlist'
1054 def _real_initialize(self):
1057 def _ids_to_results(self, ids):
1058 return [self.url_result(vid_id, 'Youtube', video_id=vid_id)
1061 def _extract_mix(self, playlist_id):
1062 # The mixes are generated from a a single video
1063 # the id of the playlist is just 'RD' + video_id
1064 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1065 webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix')
1066 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1067 title_span = (search_title('playlist-title') or
1068 search_title('title long-title') or search_title('title'))
1069 title = clean_html(title_span)
1070 video_re = r'''(?x)data-video-username=".*?
".*?
1071 href="/watch
\?v
=([0-9A
-Za
-z_
-]{11}
)&
;[^
"]*?list=%s''' % re.escape(playlist_id)
1072 ids = orderedSet(re.findall(video_re, webpage, flags=re.DOTALL))
1073 url_results = self._ids_to_results(ids)
1075 return self.playlist_result(url_results, playlist_id, title)
1077 def _real_extract(self, url):
1078 # Extract playlist id
1079 mobj = re.match(self._VALID_URL, url)
1081 raise ExtractorError(u'Invalid URL: %s' % url)
1082 playlist_id = mobj.group(1) or mobj.group(2)
1084 # Check if it's a video-specific URL
1085 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1086 if 'v' in query_dict:
1087 video_id = query_dict['v'][0]
1088 if self._downloader.params.get('noplaylist'):
1089 self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id)
1090 return self.url_result(video_id, 'Youtube', video_id=video_id)
1092 self.to_screen(u'Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1094 if playlist_id.startswith('RD'):
1095 # Mixes require a custom extraction process
1096 return self._extract_mix(playlist_id)
1097 if playlist_id.startswith('TL'):
1098 raise ExtractorError(u'For downloading YouTube.com top lists, use '
1099 u'the "yttoplist
" keyword, for example "youtube
-dl
\'yttoplist
:music
:Top Tracks
\'"', expected=True)
1101 url = self._TEMPLATE_URL % playlist_id
1102 page = self._download_webpage(url, playlist_id)
1103 more_widget_html = content_html = page
1105 # Check if the playlist exists or is private
1106 if re.search(r'<div class="yt
-alert
-message
">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
1107 raise ExtractorError(
1108 u'The playlist doesn\'t exist or is private, use --username or '
1109 '--netrc to access it.',
1112 # Extract the video ids from the playlist pages
1115 for page_num in itertools.count(1):
1116 matches = re.finditer(self._VIDEO_RE, content_html)
1117 # We remove the duplicates and the link with index 0
1118 # (it's not the first video of the playlist)
1119 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1122 mobj = re.search(r'data-uix-load-more-href="/?
(?P
<more
>[^
"]+)"', more_widget_html)
1126 more = self._download_json(
1127 'https
://youtube
.com
/%s' % mobj.group('more
'), playlist_id,
1128 'Downloading page
#%s' % page_num,
1129 transform_source
=uppercase_escape
)
1130 content_html
= more
['content_html']
1131 more_widget_html
= more
['load_more_widget_html']
1133 playlist_title
= self
._html
_search
_regex
(
1134 r
'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1137 url_results
= self
._ids
_to
_results
(ids
)
1138 return self
.playlist_result(url_results
, playlist_id
, playlist_title
)
1141 class YoutubeTopListIE(YoutubePlaylistIE
):
1142 IE_NAME
= u
'youtube:toplist'
1143 IE_DESC
= (u
'YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1144 u
' (Example: "yttoplist:music:Top Tracks")')
1145 _VALID_URL
= r
'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1147 def _real_extract(self
, url
):
1148 mobj
= re
.match(self
._VALID
_URL
, url
)
1149 channel
= mobj
.group('chann')
1150 title
= mobj
.group('title')
1151 query
= compat_urllib_parse
.urlencode({'title': title}
)
1152 playlist_re
= 'href="([^"]+?%s.*?)"' % re
.escape(query
)
1153 channel_page
= self
._download
_webpage
('https://www.youtube.com/%s' % channel
, title
)
1154 link
= self
._html
_search
_regex
(playlist_re
, channel_page
, u
'list')
1155 url
= compat_urlparse
.urljoin('https://www.youtube.com/', link
)
1157 video_re
= r
'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1159 # sometimes the webpage doesn't contain the videos
1160 # retry until we get them
1161 for i
in itertools
.count(0):
1162 msg
= u
'Downloading Youtube mix'
1164 msg
+= ', retry #%d' % i
1165 webpage
= self
._download
_webpage
(url
, title
, msg
)
1166 ids
= orderedSet(re
.findall(video_re
, webpage
))
1169 url_results
= self
._ids
_to
_results
(ids
)
1170 return self
.playlist_result(url_results
, playlist_title
=title
)
1173 class YoutubeChannelIE(InfoExtractor
):
1174 IE_DESC
= u
'YouTube.com channels'
1175 _VALID_URL
= r
"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
1176 _MORE_PAGES_INDICATOR
= 'yt-uix-load-more'
1177 _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'
1178 IE_NAME
= u
'youtube:channel'
1180 def extract_videos_from_page(self
, page
):
1182 for mobj
in re
.finditer(r
'href="/watch\?v=([0-9A-Za-z_-]+)&?', page
):
1183 if mobj
.group(1) not in ids_in_page
:
1184 ids_in_page
.append(mobj
.group(1))
1187 def _real_extract(self
, url
):
1188 # Extract channel id
1189 mobj
= re
.match(self
._VALID
_URL
, url
)
1191 raise ExtractorError(u
'Invalid URL: %s' % url
)
1193 # Download channel page
1194 channel_id
= mobj
.group(1)
1196 url
= 'https://www.youtube.com/channel/%s/videos' % channel_id
1197 channel_page
= self
._download
_webpage
(url
, channel_id
)
1198 autogenerated
= re
.search(r
'''(?x)
1200 channel-header-autogenerated-label|
1201 yt-channel-title-autogenerated
1202 )[^"]*"''', channel_page
) is not None
1205 # The videos are contained in a single page
1206 # the ajax pages can't be used, they are empty
1207 video_ids
= self
.extract_videos_from_page(channel_page
)
1209 # Download all channel pages using the json-based channel_ajax query
1210 for pagenum
in itertools
.count(1):
1211 url
= self
._MORE
_PAGES
_URL
% (pagenum
, channel_id
)
1212 page
= self
._download
_json
(
1213 url
, channel_id
, note
=u
'Downloading page #%s' % pagenum
,
1214 transform_source
=uppercase_escape
)
1216 ids_in_page
= self
.extract_videos_from_page(page
['content_html'])
1217 video_ids
.extend(ids_in_page
)
1219 if self
._MORE
_PAGES
_INDICATOR
not in page
['load_more_widget_html']:
1222 self
._downloader
.to_screen(u
'[youtube] Channel %s: Found %i videos' % (channel_id
, len(video_ids
)))
1224 url_entries
= [self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1225 for video_id
in video_ids
]
1226 return self
.playlist_result(url_entries
, channel_id
)
1229 class YoutubeUserIE(InfoExtractor
):
1230 IE_DESC
= u
'YouTube.com user videos (URL or "ytuser" keyword)'
1231 _VALID_URL
= r
'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
1232 _TEMPLATE_URL
= 'https://gdata.youtube.com/feeds/api/users/%s'
1233 _GDATA_PAGE_SIZE
= 50
1234 _GDATA_URL
= 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1235 IE_NAME
= u
'youtube:user'
1238 def suitable(cls
, url
):
1239 # Don't return True if the url can be extracted with other youtube
1240 # extractor, the regex would is too permissive and it would match.
1241 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1242 if any(ie
.suitable(url
) for ie
in other_ies
): return False
1243 else: return super(YoutubeUserIE
, cls
).suitable(url
)
1245 def _real_extract(self
, url
):
1247 mobj
= re
.match(self
._VALID
_URL
, url
)
1249 raise ExtractorError(u
'Invalid URL: %s' % url
)
1251 username
= mobj
.group(1)
1253 # Download video ids using YouTube Data API. Result size per
1254 # query is limited (currently to 50 videos) so we need to query
1255 # page by page until there are no video ids - it means we got
1258 def download_page(pagenum
):
1259 start_index
= pagenum
* self
._GDATA
_PAGE
_SIZE
+ 1
1261 gdata_url
= self
._GDATA
_URL
% (username
, self
._GDATA
_PAGE
_SIZE
, start_index
)
1262 page
= self
._download
_webpage
(
1263 gdata_url
, username
,
1264 u
'Downloading video ids from %d to %d' % (
1265 start_index
, start_index
+ self
._GDATA
_PAGE
_SIZE
))
1268 response
= json
.loads(page
)
1269 except ValueError as err
:
1270 raise ExtractorError(u
'Invalid JSON in API response: ' + compat_str(err
))
1271 if 'entry' not in response
['feed']:
1274 # Extract video identifiers
1275 entries
= response
['feed']['entry']
1276 for entry
in entries
:
1277 title
= entry
['title']['$t']
1278 video_id
= entry
['id']['$t'].split('/')[-1]
1282 'ie_key': 'Youtube',
1286 url_results
= PagedList(download_page
, self
._GDATA
_PAGE
_SIZE
)
1288 return self
.playlist_result(url_results
, playlist_title
=username
)
1291 class YoutubeSearchIE(SearchInfoExtractor
):
1292 IE_DESC
= u
'YouTube.com searches'
1293 _API_URL
= u
'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1295 IE_NAME
= u
'youtube:search'
1296 _SEARCH_KEY
= 'ytsearch'
1298 def _get_n_results(self
, query
, n
):
1299 """Get a specified number of results for a query"""
1306 while (PAGE_SIZE
* pagenum
) < limit
:
1307 result_url
= self
._API
_URL
% (
1308 compat_urllib_parse
.quote_plus(query
.encode('utf-8')),
1309 (PAGE_SIZE
* pagenum
) + 1)
1310 data_json
= self
._download
_webpage
(
1311 result_url
, video_id
=u
'query "%s"' % query
,
1312 note
=u
'Downloading page %s' % (pagenum
+ 1),
1313 errnote
=u
'Unable to download API page')
1314 data
= json
.loads(data_json
)
1315 api_response
= data
['data']
1317 if 'items' not in api_response
:
1318 raise ExtractorError(
1319 u
'[youtube] No video results', expected
=True)
1321 new_ids
= list(video
['id'] for video
in api_response
['items'])
1322 video_ids
+= new_ids
1324 limit
= min(n
, api_response
['totalItems'])
1327 if len(video_ids
) > n
:
1328 video_ids
= video_ids
[:n
]
1329 videos
= [self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1330 for video_id
in video_ids
]
1331 return self
.playlist_result(videos
, query
)
1334 class YoutubeSearchDateIE(YoutubeSearchIE
):
1335 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1336 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1337 _SEARCH_KEY
= 'ytsearchdate'
1338 IE_DESC
= u
'YouTube.com searches, newest videos first'
1341 class YoutubeSearchURLIE(InfoExtractor
):
1342 IE_DESC
= u
'YouTube.com search URLs'
1343 IE_NAME
= u
'youtube:search_url'
1344 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1346 def _real_extract(self
, url
):
1347 mobj
= re
.match(self
._VALID
_URL
, url
)
1348 query
= compat_urllib_parse
.unquote_plus(mobj
.group('query'))
1350 webpage
= self
._download
_webpage
(url
, query
)
1351 result_code
= self
._search
_regex
(
1352 r
'(?s)<ol class="item-section"(.*?)</ol>', webpage
, u
'result HTML')
1354 part_codes
= re
.findall(
1355 r
'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code
)
1357 for part_code
in part_codes
:
1358 part_title
= self
._html
_search
_regex
(
1359 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1360 part_url_snippet
= self
._html
_search
_regex
(
1361 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1362 part_url
= compat_urlparse
.urljoin(
1363 'https://www.youtube.com/', part_url_snippet
)
1367 'title': part_title
,
1371 '_type': 'playlist',
1377 class YoutubeShowIE(InfoExtractor
):
1378 IE_DESC
= u
'YouTube.com (multi-season) shows'
1379 _VALID_URL
= r
'https?://www\.youtube\.com/show/(.*)'
1380 IE_NAME
= u
'youtube:show'
1382 def _real_extract(self
, url
):
1383 mobj
= re
.match(self
._VALID
_URL
, url
)
1384 show_name
= mobj
.group(1)
1385 webpage
= self
._download
_webpage
(url
, show_name
, u
'Downloading show webpage')
1386 # There's one playlist for each season of the show
1387 m_seasons
= list(re
.finditer(r
'href="(/playlist\?list=.*?)"', webpage
))
1388 self
.to_screen(u
'%s: Found %s seasons' % (show_name
, len(m_seasons
)))
1389 return [self
.url_result('https://www.youtube.com' + season
.group(1), 'YoutubePlaylist') for season
in m_seasons
]
1392 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1394 Base class for extractors that fetch info from
1395 http://www.youtube.com/feed_ajax
1396 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1398 _LOGIN_REQUIRED
= True
1399 # use action_load_personal_feed instead of action_load_system_feed
1400 _PERSONAL_FEED
= False
1403 def _FEED_TEMPLATE(self
):
1404 action
= 'action_load_system_feed'
1405 if self
._PERSONAL
_FEED
:
1406 action
= 'action_load_personal_feed'
1407 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action
, self
._FEED
_NAME
)
1411 return u
'youtube:%s' % self
._FEED
_NAME
1413 def _real_initialize(self
):
1416 def _real_extract(self
, url
):
1419 for i
in itertools
.count(1):
1420 info
= self
._download
_json
(self
._FEED
_TEMPLATE
% paging
,
1421 u
'%s feed' % self
._FEED
_NAME
,
1422 u
'Downloading page %s' % i
)
1423 feed_html
= info
.get('feed_html') or info
.get('content_html')
1424 m_ids
= re
.finditer(r
'"/watch\?v=(.*?)["&]', feed_html
)
1425 ids
= orderedSet(m
.group(1) for m
in m_ids
)
1426 feed_entries
.extend(
1427 self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1428 for video_id
in ids
)
1430 r
'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1434 paging
= mobj
.group('paging')
1435 return self
.playlist_result(feed_entries
, playlist_title
=self
._PLAYLIST
_TITLE
)
1437 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
1438 IE_DESC
= u
'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1439 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1440 _FEED_NAME
= 'recommended'
1441 _PLAYLIST_TITLE
= u
'Youtube Recommended videos'
1443 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor
):
1444 IE_DESC
= u
'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1445 _VALID_URL
= r
'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1446 _FEED_NAME
= 'watch_later'
1447 _PLAYLIST_TITLE
= u
'Youtube Watch Later'
1448 _PERSONAL_FEED
= True
1450 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
1451 IE_DESC
= u
'Youtube watch history, "ythistory" keyword (requires authentication)'
1452 _VALID_URL
= u
'https?://www\.youtube\.com/feed/history|:ythistory'
1453 _FEED_NAME
= 'history'
1454 _PERSONAL_FEED
= True
1455 _PLAYLIST_TITLE
= u
'Youtube Watch History'
1457 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
1458 IE_NAME
= u
'youtube:favorites'
1459 IE_DESC
= u
'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1460 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1461 _LOGIN_REQUIRED
= True
1463 def _real_extract(self
, url
):
1464 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1465 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, u
'favourites playlist id')
1466 return self
.url_result(playlist_id
, 'YoutubePlaylist')
1469 class YoutubeSubscriptionsIE(YoutubePlaylistIE
):
1470 IE_NAME
= u
'youtube:subscriptions'
1471 IE_DESC
= u
'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1472 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1474 def _real_extract(self
, url
):
1475 title
= u
'Youtube Subscriptions'
1476 page
= self
._download
_webpage
('https://www.youtube.com/feed/subscriptions', title
)
1478 # The extraction process is the same as for playlists, but the regex
1479 # for the video ids doesn't contain an index
1481 more_widget_html
= content_html
= page
1483 for page_num
in itertools
.count(1):
1484 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
1485 new_ids
= orderedSet(matches
)
1488 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1492 more
= self
._download
_json
(
1493 'https://youtube.com/%s' % mobj
.group('more'), title
,
1494 'Downloading page #%s' % page_num
,
1495 transform_source
=uppercase_escape
)
1496 content_html
= more
['content_html']
1497 more_widget_html
= more
['load_more_widget_html']
1500 '_type': 'playlist',
1502 'entries': self
._ids
_to
_results
(ids
),
1506 class YoutubeTruncatedURLIE(InfoExtractor
):
1507 IE_NAME
= 'youtube:truncated_url'
1508 IE_DESC
= False # Do not list
1509 _VALID_URL
= r
'''(?x)
1510 (?:https?://)?[^/]+/watch\?(?:
1512 annotation_id=annotation_[^&]+
1514 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1518 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1519 'only_matching': True,
1521 'url': 'http://www.youtube.com/watch?',
1522 'only_matching': True,
1525 def _real_extract(self
, url
):
1526 raise ExtractorError(
1527 u
'Did you forget to quote the URL? Remember that & is a meta '
1528 u
'character in most shells, so you want to put the URL in quotes, '
1530 u
'"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1531 u
' or simply youtube-dl BaW_jenozKc .',