9 from .common
import InfoExtractor
, SearchInfoExtractor
15 compat_urllib_request
,
26 class YoutubeBaseInfoExtractor(InfoExtractor
):
27 """Provide base functions for Youtube extractors"""
28 _LOGIN_URL
= 'https://accounts.google.com/ServiceLogin'
29 _LANG_URL
= r
'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
30 _AGE_URL
= 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
31 _NETRC_MACHINE
= 'youtube'
32 # If True it will raise an error if no login info is provided
33 _LOGIN_REQUIRED
= False
35 def report_lang(self
):
36 """Report attempt to set language."""
37 self
.to_screen(u
'Setting language')
39 def _set_language(self
):
40 request
= compat_urllib_request
.Request(self
._LANG
_URL
)
43 compat_urllib_request
.urlopen(request
).read()
44 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
45 self
._downloader
.report_warning(u
'unable to set language: %s' % compat_str(err
))
50 (username
, password
) = self
._get
_login
_info
()
51 # No authentication to be performed
53 if self
._LOGIN
_REQUIRED
:
54 raise ExtractorError(u
'No login info available, needed for using %s.' % self
.IE_NAME
, expected
=True)
57 request
= compat_urllib_request
.Request(self
._LOGIN
_URL
)
59 login_page
= compat_urllib_request
.urlopen(request
).read().decode('utf-8')
60 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
61 self
._downloader
.report_warning(u
'unable to fetch login page: %s' % compat_str(err
))
66 match
= re
.search(re
.compile(r
'<input.+?name="GALX".+?value="(.+?)"', re
.DOTALL
), login_page
)
69 match
= re
.search(re
.compile(r
'<input.+?name="dsh".+?value="(.+?)"', re
.DOTALL
), login_page
)
75 u
'continue': u
'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
79 u
'PersistentCookie': u
'yes',
81 u
'bgresponse': u
'js_disabled',
82 u
'checkConnection': u
'',
83 u
'checkedDomains': u
'youtube',
89 u
'signIn': u
'Sign in',
91 u
'service': u
'youtube',
95 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
97 login_form
= dict((k
.encode('utf-8'), v
.encode('utf-8')) for k
,v
in login_form_strs
.items())
98 login_data
= compat_urllib_parse
.urlencode(login_form
).encode('ascii')
99 request
= compat_urllib_request
.Request(self
._LOGIN
_URL
, login_data
)
102 login_results
= compat_urllib_request
.urlopen(request
).read().decode('utf-8')
103 if re
.search(r
'(?i)<form[^>]* id="gaia_loginform"', login_results
) is not None:
104 self
._downloader
.report_warning(u
'unable to log in: bad username or password')
106 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
107 self
._downloader
.report_warning(u
'unable to log in: %s' % compat_str(err
))
111 def _confirm_age(self
):
114 'action_confirm': 'Confirm',
116 request
= compat_urllib_request
.Request(self
._AGE
_URL
, compat_urllib_parse
.urlencode(age_form
))
118 self
.report_age_confirmation()
119 compat_urllib_request
.urlopen(request
).read().decode('utf-8')
120 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
121 raise ExtractorError(u
'Unable to confirm age: %s' % compat_str(err
))
124 def _real_initialize(self
):
125 if self
._downloader
is None:
127 if not self
._set
_language
():
129 if not self
._login
():
133 class YoutubeIE(YoutubeBaseInfoExtractor
):
134 IE_DESC
= u
'YouTube.com'
137 (?:https?://)? # http(s):// (optional)
138 (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/|
139 tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains
140 (?:.*?\#/)? # handle anchor (#/) redirect urls
141 (?: # the various things that can precede the ID:
142 (?:(?:v|embed|e)/) # v/ or embed/ or e/
143 |(?: # or the v= param in all its forms
144 (?:(?:watch|movie)(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
145 (?:\?|\#!?) # the params delimiter ? or # or #!
146 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
149 )? # optional -> youtube.com/xxxx is OK
150 )? # all until now is optional -> you can pass the naked ID
151 ([0-9A-Za-z_-]+) # here is it! the YouTube video ID
152 (?(1).+)? # if we found the ID, everything can follow
154 _NEXT_URL_RE
= r
'[\?&]next_url=([^&]+)'
155 # Listed in order of quality
156 _available_formats
= ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13',
157 '95', '94', '93', '92', '132', '151',
158 '85', '84', '102', '83', '101', '82', '100', # 3D
159 '138', '137', '136', '135', '134', '133', '160', # Dash video mp4
160 '141', '140', '139', # Dash auido mp4
161 '248', '247', '246', '245', '244', '243', '242', # Dash video webm
162 '172', '171', # Dash audio webm
164 _available_formats_prefer_free
= ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13',
165 '95', '94', '93', '92', '132', '151',
166 '85', '102', '84', '101', '83', '100', '82',
167 '248', '247', '246', '245', '244', '243', '242', # Dash video webm
168 '172', '171', # Dash audio webm
169 '138', '137', '136', '135', '134', '133', '160', # Dash video mp4
170 '141', '140', '139', # Dash auido mp4
172 _video_extensions
= {
193 # videos that use m3u8
225 _video_dimensions
= {
306 u
"url": u
"http://www.youtube.com/watch?v=BaW_jenozKc",
307 u
"file": u
"BaW_jenozKc.mp4",
309 u
"title": u
"youtube-dl test video \"'/\\ä↭𝕐",
310 u
"uploader": u
"Philipp Hagemeister",
311 u
"uploader_id": u
"phihag",
312 u
"upload_date": u
"20121002",
313 u
"description": u
"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
317 u
"url": u
"http://www.youtube.com/watch?v=1ltcDfZMA3U",
318 u
"file": u
"1ltcDfZMA3U.flv",
319 u
"note": u
"Test VEVO video (#897)",
321 u
"upload_date": u
"20070518",
322 u
"title": u
"Maps - It Will Find You",
323 u
"description": u
"Music video by Maps performing It Will Find You.",
324 u
"uploader": u
"MuteUSA",
325 u
"uploader_id": u
"MuteUSA"
329 u
"url": u
"http://www.youtube.com/watch?v=UxxajLWwzqY",
330 u
"file": u
"UxxajLWwzqY.mp4",
331 u
"note": u
"Test generic use_cipher_signature video (#897)",
333 u
"upload_date": u
"20120506",
334 u
"title": u
"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
335 u
"description": u
"md5:b085c9804f5ab69f4adea963a2dceb3c",
336 u
"uploader": u
"IconaPop",
337 u
"uploader_id": u
"IconaPop"
341 u
"url": u
"https://www.youtube.com/watch?v=07FYdnEawAQ",
342 u
"file": u
"07FYdnEawAQ.mp4",
343 u
"note": u
"Test VEVO video with age protection (#956)",
345 u
"upload_date": u
"20130703",
346 u
"title": u
"Justin Timberlake - Tunnel Vision (Explicit)",
347 u
"description": u
"md5:64249768eec3bc4276236606ea996373",
348 u
"uploader": u
"justintimberlakeVEVO",
349 u
"uploader_id": u
"justintimberlakeVEVO"
353 u
'url': u
'https://www.youtube.com/watch?v=TGi3HqYrWHE',
354 u
'file': u
'TGi3HqYrWHE.mp4',
355 u
'note': u
'm3u8 video',
357 u
'title': u
'Triathlon - Men - London 2012 Olympic Games',
358 u
'description': u
'- Men - TR02 - Triathlon - 07 August 2012 - London 2012 Olympic Games',
359 u
'uploader': u
'olympic',
360 u
'upload_date': u
'20120807',
361 u
'uploader_id': u
'olympic',
364 u
'skip_download': True,
371 def suitable(cls
, url
):
372 """Receives a URL and returns True if suitable for this IE."""
373 if YoutubePlaylistIE
.suitable(url
) or YoutubeSubscriptionsIE
.suitable(url
): return False
374 return re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
) is not None
376 def report_video_webpage_download(self
, video_id
):
377 """Report attempt to download video webpage."""
378 self
.to_screen(u
'%s: Downloading video webpage' % video_id
)
380 def report_video_info_webpage_download(self
, video_id
):
381 """Report attempt to download video info webpage."""
382 self
.to_screen(u
'%s: Downloading video info webpage' % video_id
)
384 def report_video_subtitles_download(self
, video_id
):
385 """Report attempt to download video info webpage."""
386 self
.to_screen(u
'%s: Checking available subtitles' % video_id
)
388 def report_video_subtitles_request(self
, video_id
, sub_lang
, format
):
389 """Report attempt to download video info webpage."""
390 self
.to_screen(u
'%s: Downloading video subtitles for %s.%s' % (video_id
, sub_lang
, format
))
392 def report_video_subtitles_available(self
, video_id
, sub_lang_list
):
393 """Report available subtitles."""
394 sub_lang
= ",".join(list(sub_lang_list
.keys()))
395 self
.to_screen(u
'%s: Available subtitles for video: %s' % (video_id
, sub_lang
))
397 def report_information_extraction(self
, video_id
):
398 """Report attempt to extract video information."""
399 self
.to_screen(u
'%s: Extracting video information' % video_id
)
401 def report_unavailable_format(self
, video_id
, format
):
402 """Report extracted video URL."""
403 self
.to_screen(u
'%s: Format %s not available' % (video_id
, format
))
405 def report_rtmp_download(self
):
406 """Indicate the download will use the RTMP protocol."""
407 self
.to_screen(u
'RTMP download detected')
409 def _decrypt_signature(self
, s
):
410 """Turn the encrypted s field into a working signature"""
413 return s
[25] + s
[3:25] + s
[0] + s
[26:42] + s
[79] + s
[43:79] + s
[91] + s
[80:83]
415 return s
[25] + s
[3:25] + s
[2] + s
[26:40] + s
[77] + s
[41:77] + s
[89] + s
[78:81]
417 return s
[84:78:-1] + s
[87] + s
[77:60:-1] + s
[0] + s
[59:3:-1]
419 return s
[48] + s
[81:67:-1] + s
[82] + s
[66:62:-1] + s
[85] + s
[61:48:-1] + s
[67] + s
[47:12:-1] + s
[3] + s
[11:3:-1] + s
[2] + s
[12]
421 return s
[6:27] + s
[4] + s
[28:39] + s
[27] + s
[40:59] + s
[2] + s
[60:]
423 return s
[5:20] + s
[2] + s
[21:]
425 return s
[83:34:-1] + s
[0] + s
[33:27:-1] + s
[3] + s
[26:19:-1] + s
[34] + s
[18:3:-1] + s
[27]
427 return s
[83:27:-1] + s
[0] + s
[26:5:-1] + s
[2:0:-1] + s
[27]
429 return s
[81:64:-1] + s
[82] + s
[63:52:-1] + s
[45] + s
[51:45:-1] + s
[1] + s
[44:1:-1] + s
[0]
431 return s
[36] + s
[79:67:-1] + s
[81] + s
[66:40:-1] + s
[33] + s
[39:36:-1] + s
[40] + s
[35] + s
[0] + s
[67] + s
[32:0:-1] + s
[34]
433 return s
[56] + s
[79:56:-1] + s
[41] + s
[55:41:-1] + s
[80] + s
[40:34:-1] + s
[0] + s
[33:29:-1] + s
[34] + s
[28:9:-1] + s
[29] + s
[8:0:-1] + s
[9]
435 return s
[54] + s
[77:54:-1] + s
[39] + s
[53:39:-1] + s
[78] + s
[38:34:-1] + s
[0] + s
[33:29:-1] + s
[34] + s
[28:9:-1] + s
[29] + s
[8:0:-1] + s
[9]
438 raise ExtractorError(u
'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s
)))
440 def _decrypt_signature_age_gate(self
, s
):
441 # The videos with age protection use another player, so the algorithms
444 return s
[2:63] + s
[82] + s
[64:82] + s
[63]
446 # Fallback to the other algortihms
447 return self
._decrypt
_signature
(s
)
450 def _get_available_subtitles(self
, video_id
):
451 self
.report_video_subtitles_download(video_id
)
452 request
= compat_urllib_request
.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
)
454 sub_list
= compat_urllib_request
.urlopen(request
).read().decode('utf-8')
455 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
456 return (u
'unable to download video subtitles: %s' % compat_str(err
), None)
457 sub_lang_list
= re
.findall(r
'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list
)
458 sub_lang_list
= dict((l
[1], l
[0]) for l
in sub_lang_list
)
459 if not sub_lang_list
:
460 return (u
'video doesn\'t have subtitles', None)
463 def _list_available_subtitles(self
, video_id
):
464 sub_lang_list
= self
._get
_available
_subtitles
(video_id
)
465 self
.report_video_subtitles_available(video_id
, sub_lang_list
)
467 def _request_subtitle(self
, sub_lang
, sub_name
, video_id
, format
):
470 (error_message, sub_lang, sub)
472 self
.report_video_subtitles_request(video_id
, sub_lang
, format
)
473 params
= compat_urllib_parse
.urlencode({
479 url
= 'http://www.youtube.com/api/timedtext?' + params
481 sub
= compat_urllib_request
.urlopen(url
).read().decode('utf-8')
482 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
483 return (u
'unable to download video subtitles: %s' % compat_str(err
), None, None)
485 return (u
'Did not fetch video subtitles', None, None)
486 return (None, sub_lang
, sub
)
488 def _request_automatic_caption(self
, video_id
, webpage
):
489 """We need the webpage for getting the captions url, pass it as an
490 argument to speed up the process."""
491 sub_lang
= self
._downloader
.params
.get('subtitleslang') or 'en'
492 sub_format
= self
._downloader
.params
.get('subtitlesformat')
493 self
.to_screen(u
'%s: Looking for automatic captions' % video_id
)
494 mobj
= re
.search(r
';ytplayer.config = ({.*?});', webpage
)
495 err_msg
= u
'Couldn\'t find automatic captions for "%s"' % sub_lang
497 return [(err_msg
, None, None)]
498 player_config
= json
.loads(mobj
.group(1))
500 args
= player_config
[u
'args']
501 caption_url
= args
[u
'ttsurl']
502 timestamp
= args
[u
'timestamp']
503 params
= compat_urllib_parse
.urlencode({
510 subtitles_url
= caption_url
+ '&' + params
511 sub
= self
._download
_webpage
(subtitles_url
, video_id
, u
'Downloading automatic captions')
512 return [(None, sub_lang
, sub
)]
514 return [(err_msg
, None, None)]
516 def _extract_subtitle(self
, video_id
):
518 Return a list with a tuple:
519 [(error_message, sub_lang, sub)]
521 sub_lang_list
= self
._get
_available
_subtitles
(video_id
)
522 sub_format
= self
._downloader
.params
.get('subtitlesformat')
523 if isinstance(sub_lang_list
,tuple): #There was some error, it didn't get the available subtitles
524 return [(sub_lang_list
[0], None, None)]
525 if self
._downloader
.params
.get('subtitleslang', False):
526 sub_lang
= self
._downloader
.params
.get('subtitleslang')
527 elif 'en' in sub_lang_list
:
530 sub_lang
= list(sub_lang_list
.keys())[0]
531 if not sub_lang
in sub_lang_list
:
532 return [(u
'no closed captions found in the specified language "%s"' % sub_lang
, None, None)]
534 subtitle
= self
._request
_subtitle
(sub_lang
, sub_lang_list
[sub_lang
].encode('utf-8'), video_id
, sub_format
)
537 def _extract_all_subtitles(self
, video_id
):
538 sub_lang_list
= self
._get
_available
_subtitles
(video_id
)
539 sub_format
= self
._downloader
.params
.get('subtitlesformat')
540 if isinstance(sub_lang_list
,tuple): #There was some error, it didn't get the available subtitles
541 return [(sub_lang_list
[0], None, None)]
543 for sub_lang
in sub_lang_list
:
544 subtitle
= self
._request
_subtitle
(sub_lang
, sub_lang_list
[sub_lang
].encode('utf-8'), video_id
, sub_format
)
545 subtitles
.append(subtitle
)
548 def _print_formats(self
, formats
):
549 print('Available formats:')
551 print('%s\t:\t%s\t[%s]%s' %(x
, self
._video
_extensions
.get(x
, 'flv'),
552 self
._video
_dimensions
.get(x
, '???'),
553 ' ('+self
._special
_itags
[x
]+')' if x
in self
._special
_itags
else ''))
555 def _extract_id(self
, url
):
556 mobj
= re
.match(self
._VALID
_URL
, url
, re
.VERBOSE
)
558 raise ExtractorError(u
'Invalid URL: %s' % url
)
559 video_id
= mobj
.group(2)
562 def _get_video_url_list(self
, url_map
):
564 Transform a dictionary in the format {itag:url} to a list of (itag, url)
565 with the requested formats.
567 req_format
= self
._downloader
.params
.get('format', None)
568 format_limit
= self
._downloader
.params
.get('format_limit', None)
569 available_formats
= self
._available
_formats
_prefer
_free
if self
._downloader
.params
.get('prefer_free_formats', False) else self
._available
_formats
570 if format_limit
is not None and format_limit
in available_formats
:
571 format_list
= available_formats
[available_formats
.index(format_limit
):]
573 format_list
= available_formats
574 existing_formats
= [x
for x
in format_list
if x
in url_map
]
575 if len(existing_formats
) == 0:
576 raise ExtractorError(u
'no known formats available for video')
577 if self
._downloader
.params
.get('listformats', None):
578 self
._print
_formats
(existing_formats
)
580 if req_format
is None or req_format
== 'best':
581 video_url_list
= [(existing_formats
[0], url_map
[existing_formats
[0]])] # Best quality
582 elif req_format
== 'worst':
583 video_url_list
= [(existing_formats
[-1], url_map
[existing_formats
[-1]])] # worst quality
584 elif req_format
in ('-1', 'all'):
585 video_url_list
= [(f
, url_map
[f
]) for f
in existing_formats
] # All formats
587 # Specific formats. We pick the first in a slash-delimeted sequence.
588 # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
589 req_formats
= req_format
.split('/')
590 video_url_list
= None
591 for rf
in req_formats
:
593 video_url_list
= [(rf
, url_map
[rf
])]
595 if video_url_list
is None:
596 raise ExtractorError(u
'requested format not available')
597 return video_url_list
599 def _extract_from_m3u8(self
, manifest_url
, video_id
):
601 def _get_urls(_manifest
):
602 lines
= _manifest
.split('\n')
603 urls
= filter(lambda l
: l
and not l
.startswith('#'),
606 manifest
= self
._download
_webpage
(manifest_url
, video_id
, u
'Downloading formats manifest')
607 formats_urls
= _get_urls(manifest
)
608 for format_url
in formats_urls
:
609 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
610 url_map
[itag
] = format_url
613 def _real_extract(self
, url
):
614 if re
.match(r
'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url
):
615 self
._downloader
.report_warning(u
'Did you forget to quote the URL? Remember that & is a meta-character in most shells, so you want to put the URL in quotes, like youtube-dl \'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\' (or simply youtube-dl BaW_jenozKc ).')
617 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
618 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
620 url
= 'https://www.youtube.com/' + compat_urllib_parse
.unquote(mobj
.group(1)).lstrip('/')
621 video_id
= self
._extract
_id
(url
)
624 self
.report_video_webpage_download(video_id
)
625 url
= 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
626 request
= compat_urllib_request
.Request(url
)
628 video_webpage_bytes
= compat_urllib_request
.urlopen(request
).read()
629 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
630 raise ExtractorError(u
'Unable to download video webpage: %s' % compat_str(err
))
632 video_webpage
= video_webpage_bytes
.decode('utf-8', 'ignore')
634 # Attempt to extract SWF player URL
635 mobj
= re
.search(r
'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
637 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
642 self
.report_video_info_webpage_download(video_id
)
643 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
644 self
.report_age_confirmation()
646 # We simulate the access to the video from www.youtube.com/v/{video_id}
647 # this can be viewed without login into Youtube
648 data
= compat_urllib_parse
.urlencode({'video_id': video_id
,
652 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
656 video_info_url
= 'https://www.youtube.com/get_video_info?' + data
657 video_info_webpage
= self
._download
_webpage
(video_info_url
, video_id
,
659 errnote
='unable to download video info webpage')
660 video_info
= compat_parse_qs(video_info_webpage
)
663 for el_type
in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
664 video_info_url
= ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
665 % (video_id
, el_type
))
666 video_info_webpage
= self
._download
_webpage
(video_info_url
, video_id
,
668 errnote
='unable to download video info webpage')
669 video_info
= compat_parse_qs(video_info_webpage
)
670 if 'token' in video_info
:
672 if 'token' not in video_info
:
673 if 'reason' in video_info
:
674 raise ExtractorError(u
'YouTube said: %s' % video_info
['reason'][0], expected
=True)
676 raise ExtractorError(u
'"token" parameter not in video info for unknown reason')
678 # Check for "rental" videos
679 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
680 raise ExtractorError(u
'"rental" videos not supported')
682 # Start extracting information
683 self
.report_information_extraction(video_id
)
686 if 'author' not in video_info
:
687 raise ExtractorError(u
'Unable to extract uploader name')
688 video_uploader
= compat_urllib_parse
.unquote_plus(video_info
['author'][0])
691 video_uploader_id
= None
692 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
694 video_uploader_id
= mobj
.group(1)
696 self
._downloader
.report_warning(u
'unable to extract uploader nickname')
699 if 'title' not in video_info
:
700 raise ExtractorError(u
'Unable to extract video title')
701 video_title
= compat_urllib_parse
.unquote_plus(video_info
['title'][0])
704 # We try first to get a high quality image:
705 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
706 video_webpage
, re
.DOTALL
)
707 if m_thumb
is not None:
708 video_thumbnail
= m_thumb
.group(1)
709 elif 'thumbnail_url' not in video_info
:
710 self
._downloader
.report_warning(u
'unable to extract video thumbnail')
712 else: # don't panic if we can't find it
713 video_thumbnail
= compat_urllib_parse
.unquote_plus(video_info
['thumbnail_url'][0])
717 mobj
= re
.search(r
'id="eow-date.*?>(.*?)</span>', video_webpage
, re
.DOTALL
)
719 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
720 upload_date
= unified_strdate(upload_date
)
723 video_description
= get_element_by_id("eow-description", video_webpage
)
724 if video_description
:
725 video_description
= clean_html(video_description
)
727 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
729 video_description
= unescapeHTML(fd_mobj
.group(1))
731 video_description
= u
''
734 video_subtitles
= None
736 if self
._downloader
.params
.get('writesubtitles', False):
737 video_subtitles
= self
._extract
_subtitle
(video_id
)
739 (sub_error
, sub_lang
, sub
) = video_subtitles
[0]
741 self
._downloader
.report_warning(sub_error
)
743 if self
._downloader
.params
.get('writeautomaticsub', False):
744 video_subtitles
= self
._request
_automatic
_caption
(video_id
, video_webpage
)
745 (sub_error
, sub_lang
, sub
) = video_subtitles
[0]
747 self
._downloader
.report_warning(sub_error
)
749 if self
._downloader
.params
.get('allsubtitles', False):
750 video_subtitles
= self
._extract
_all
_subtitles
(video_id
)
751 for video_subtitle
in video_subtitles
:
752 (sub_error
, sub_lang
, sub
) = video_subtitle
754 self
._downloader
.report_warning(sub_error
)
756 if self
._downloader
.params
.get('listsubtitles', False):
757 self
._list
_available
_subtitles
(video_id
)
760 if 'length_seconds' not in video_info
:
761 self
._downloader
.report_warning(u
'unable to extract video duration')
764 video_duration
= compat_urllib_parse
.unquote_plus(video_info
['length_seconds'][0])
766 # Decide which formats to download
769 mobj
= re
.search(r
';ytplayer.config = ({.*?});', video_webpage
)
771 raise ValueError('Could not find vevo ID')
772 info
= json
.loads(mobj
.group(1))
774 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
775 # this signatures are encrypted
776 m_s
= re
.search(r
'[&,]s=', args
['url_encoded_fmt_stream_map'])
778 self
.to_screen(u
'%s: Encrypted signatures detected.' % video_id
)
779 video_info
['url_encoded_fmt_stream_map'] = [args
['url_encoded_fmt_stream_map']]
780 m_s
= re
.search(r
'[&,]s=', args
.get('adaptive_fmts', u
''))
782 if 'url_encoded_fmt_stream_map' in video_info
:
783 video_info
['url_encoded_fmt_stream_map'][0] += ',' + args
['adaptive_fmts']
785 video_info
['url_encoded_fmt_stream_map'] = [args
['adaptive_fmts']]
786 elif 'adaptive_fmts' in video_info
:
787 if 'url_encoded_fmt_stream_map' in video_info
:
788 video_info
['url_encoded_fmt_stream_map'][0] += ',' + video_info
['adaptive_fmts'][0]
790 video_info
['url_encoded_fmt_stream_map'] = video_info
['adaptive_fmts']
794 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
795 self
.report_rtmp_download()
796 video_url_list
= [(None, video_info
['conn'][0])]
797 elif 'url_encoded_fmt_stream_map' in video_info
and len(video_info
['url_encoded_fmt_stream_map']) >= 1:
798 if 'rtmpe%3Dyes' in video_info
['url_encoded_fmt_stream_map'][0]:
799 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
801 for url_data_str
in video_info
['url_encoded_fmt_stream_map'][0].split(','):
802 url_data
= compat_parse_qs(url_data_str
)
803 if 'itag' in url_data
and 'url' in url_data
:
804 url
= url_data
['url'][0]
805 if 'sig' in url_data
:
806 url
+= '&signature=' + url_data
['sig'][0]
807 elif 's' in url_data
:
808 if self
._downloader
.params
.get('verbose'):
811 player_version
= self
._search
_regex
(r
'ad3-(.+?)\.swf',
812 video_info
['ad3_module'][0] if 'ad3_module' in video_info
else 'NOT FOUND',
813 'flash player', fatal
=False)
814 player
= 'flash player %s' % player_version
816 player
= u
'html5 player %s' % self
._search
_regex
(r
'html5player-(.+?)\.js', video_webpage
,
817 'html5 player', fatal
=False)
818 parts_sizes
= u
'.'.join(compat_str(len(part
)) for part
in s
.split('.'))
819 self
.to_screen(u
'encrypted signature length %d (%s), itag %s, %s' %
820 (len(s
), parts_sizes
, url_data
['itag'][0], player
))
821 encrypted_sig
= url_data
['s'][0]
823 signature
= self
._decrypt
_signature
_age
_gate
(encrypted_sig
)
825 signature
= self
._decrypt
_signature
(encrypted_sig
)
826 url
+= '&signature=' + signature
827 if 'ratebypass' not in url
:
828 url
+= '&ratebypass=yes'
829 url_map
[url_data
['itag'][0]] = url
830 video_url_list
= self
._get
_video
_url
_list
(url_map
)
831 if not video_url_list
:
833 elif video_info
.get('hlsvp'):
834 manifest_url
= video_info
['hlsvp'][0]
835 url_map
= self
._extract
_from
_m
3u8(manifest_url
, video_id
)
836 video_url_list
= self
._get
_video
_url
_list
(url_map
)
837 if not video_url_list
:
841 raise ExtractorError(u
'no conn or url_encoded_fmt_stream_map information found in video info')
844 for format_param
, video_real_url
in video_url_list
:
846 video_extension
= self
._video
_extensions
.get(format_param
, 'flv')
848 video_format
= '{0} - {1}{2}'.format(format_param
if format_param
else video_extension
,
849 self
._video
_dimensions
.get(format_param
, '???'),
850 ' ('+self
._special
_itags
[format_param
]+')' if format_param
in self
._special
_itags
else '')
854 'url': video_real_url
,
855 'uploader': video_uploader
,
856 'uploader_id': video_uploader_id
,
857 'upload_date': upload_date
,
858 'title': video_title
,
859 'ext': video_extension
,
860 'format': video_format
,
861 'thumbnail': video_thumbnail
,
862 'description': video_description
,
863 'player_url': player_url
,
864 'subtitles': video_subtitles
,
865 'duration': video_duration
869 class YoutubePlaylistIE(InfoExtractor
):
870 IE_DESC
= u
'YouTube.com playlists'
876 (?:course|view_play_list|my_playlists|artist|playlist|watch)
877 \? (?:.*?&)*? (?:p|a|list)=
880 ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,})
883 ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,})
885 _TEMPLATE_URL
= 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
887 IE_NAME
= u
'youtube:playlist'
890 def suitable(cls
, url
):
891 """Receives a URL and returns True if suitable for this IE."""
892 return re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
) is not None
894 def _real_extract(self
, url
):
895 # Extract playlist id
896 mobj
= re
.match(self
._VALID
_URL
, url
, re
.VERBOSE
)
898 raise ExtractorError(u
'Invalid URL: %s' % url
)
900 # Download playlist videos from API
901 playlist_id
= mobj
.group(1) or mobj
.group(2)
904 for page_num
in itertools
.count(1):
905 start_index
= self
._MAX
_RESULTS
* (page_num
- 1) + 1
906 if start_index
>= 1000:
907 self
._downloader
.report_warning(u
'Max number of results reached')
909 url
= self
._TEMPLATE
_URL
% (playlist_id
, self
._MAX
_RESULTS
, start_index
)
910 page
= self
._download
_webpage
(url
, playlist_id
, u
'Downloading page #%s' % page_num
)
913 response
= json
.loads(page
)
914 except ValueError as err
:
915 raise ExtractorError(u
'Invalid JSON in API response: ' + compat_str(err
))
917 if 'feed' not in response
:
918 raise ExtractorError(u
'Got a malformed response from YouTube API')
919 playlist_title
= response
['feed']['title']['$t']
920 if 'entry' not in response
['feed']:
921 # Number of videos is a multiple of self._MAX_RESULTS
924 for entry
in response
['feed']['entry']:
925 index
= entry
['yt$position']['$t']
926 if 'media$group' in entry
and 'media$player' in entry
['media$group']:
927 videos
.append((index
, entry
['media$group']['media$player']['url']))
929 videos
= [v
[1] for v
in sorted(videos
)]
931 url_results
= [self
.url_result(vurl
, 'Youtube') for vurl
in videos
]
932 return [self
.playlist_result(url_results
, playlist_id
, playlist_title
)]
935 class YoutubeChannelIE(InfoExtractor
):
936 IE_DESC
= u
'YouTube.com channels'
937 _VALID_URL
= r
"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
938 _TEMPLATE_URL
= 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
939 _MORE_PAGES_INDICATOR
= 'yt-uix-load-more'
940 _MORE_PAGES_URL
= 'http://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
941 IE_NAME
= u
'youtube:channel'
943 def extract_videos_from_page(self
, page
):
945 for mobj
in re
.finditer(r
'href="/watch\?v=([0-9A-Za-z_-]+)&?', page
):
946 if mobj
.group(1) not in ids_in_page
:
947 ids_in_page
.append(mobj
.group(1))
950 def _real_extract(self
, url
):
952 mobj
= re
.match(self
._VALID
_URL
, url
)
954 raise ExtractorError(u
'Invalid URL: %s' % url
)
956 # Download channel page
957 channel_id
= mobj
.group(1)
961 url
= self
._TEMPLATE
_URL
% (channel_id
, pagenum
)
962 page
= self
._download
_webpage
(url
, channel_id
,
963 u
'Downloading page #%s' % pagenum
)
965 # Extract video identifiers
966 ids_in_page
= self
.extract_videos_from_page(page
)
967 video_ids
.extend(ids_in_page
)
969 # Download any subsequent channel pages using the json-based channel_ajax query
970 if self
._MORE
_PAGES
_INDICATOR
in page
:
971 for pagenum
in itertools
.count(1):
972 url
= self
._MORE
_PAGES
_URL
% (pagenum
, channel_id
)
973 page
= self
._download
_webpage
(url
, channel_id
,
974 u
'Downloading page #%s' % pagenum
)
976 page
= json
.loads(page
)
978 ids_in_page
= self
.extract_videos_from_page(page
['content_html'])
979 video_ids
.extend(ids_in_page
)
981 if self
._MORE
_PAGES
_INDICATOR
not in page
['load_more_widget_html']:
984 self
._downloader
.to_screen(u
'[youtube] Channel %s: Found %i videos' % (channel_id
, len(video_ids
)))
986 urls
= ['http://www.youtube.com/watch?v=%s' % id for id in video_ids
]
987 url_entries
= [self
.url_result(eurl
, 'Youtube') for eurl
in urls
]
988 return [self
.playlist_result(url_entries
, channel_id
)]
991 class YoutubeUserIE(InfoExtractor
):
992 IE_DESC
= u
'YouTube.com user videos (URL or "ytuser" keyword)'
993 _VALID_URL
= r
'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
994 _TEMPLATE_URL
= 'http://gdata.youtube.com/feeds/api/users/%s'
995 _GDATA_PAGE_SIZE
= 50
996 _GDATA_URL
= 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
997 _VIDEO_INDICATOR
= r
'/watch\?v=(.+?)[\<&]'
998 IE_NAME
= u
'youtube:user'
1000 def _real_extract(self
, url
):
1002 mobj
= re
.match(self
._VALID
_URL
, url
)
1004 raise ExtractorError(u
'Invalid URL: %s' % url
)
1006 username
= mobj
.group(1)
1008 # Download video ids using YouTube Data API. Result size per
1009 # query is limited (currently to 50 videos) so we need to query
1010 # page by page until there are no video ids - it means we got
1015 for pagenum
in itertools
.count(0):
1016 start_index
= pagenum
* self
._GDATA
_PAGE
_SIZE
+ 1
1018 gdata_url
= self
._GDATA
_URL
% (username
, self
._GDATA
_PAGE
_SIZE
, start_index
)
1019 page
= self
._download
_webpage
(gdata_url
, username
,
1020 u
'Downloading video ids from %d to %d' % (start_index
, start_index
+ self
._GDATA
_PAGE
_SIZE
))
1022 # Extract video identifiers
1025 for mobj
in re
.finditer(self
._VIDEO
_INDICATOR
, page
):
1026 if mobj
.group(1) not in ids_in_page
:
1027 ids_in_page
.append(mobj
.group(1))
1029 video_ids
.extend(ids_in_page
)
1031 # A little optimization - if current page is not
1032 # "full", ie. does not contain PAGE_SIZE video ids then
1033 # we can assume that this page is the last one - there
1034 # are no more ids on further pages - no need to query
1037 if len(ids_in_page
) < self
._GDATA
_PAGE
_SIZE
:
1040 urls
= ['http://www.youtube.com/watch?v=%s' % video_id
for video_id
in video_ids
]
1041 url_results
= [self
.url_result(rurl
, 'Youtube') for rurl
in urls
]
1042 return [self
.playlist_result(url_results
, playlist_title
= username
)]
1044 class YoutubeSearchIE(SearchInfoExtractor
):
1045 IE_DESC
= u
'YouTube.com searches'
1046 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1048 IE_NAME
= u
'youtube:search'
1049 _SEARCH_KEY
= 'ytsearch'
1051 def report_download_page(self
, query
, pagenum
):
1052 """Report attempt to download search page with given number."""
1053 self
._downloader
.to_screen(u
'[youtube] query "%s": Downloading page %s' % (query
, pagenum
))
1055 def _get_n_results(self
, query
, n
):
1056 """Get a specified number of results for a query"""
1062 while (50 * pagenum
) < limit
:
1063 self
.report_download_page(query
, pagenum
+1)
1064 result_url
= self
._API
_URL
% (compat_urllib_parse
.quote_plus(query
), (50*pagenum
)+1)
1065 request
= compat_urllib_request
.Request(result_url
)
1067 data
= compat_urllib_request
.urlopen(request
).read().decode('utf-8')
1068 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
1069 raise ExtractorError(u
'Unable to download API page: %s' % compat_str(err
))
1070 api_response
= json
.loads(data
)['data']
1072 if not 'items' in api_response
:
1073 raise ExtractorError(u
'[youtube] No video results')
1075 new_ids
= list(video
['id'] for video
in api_response
['items'])
1076 video_ids
+= new_ids
1078 limit
= min(n
, api_response
['totalItems'])
1081 if len(video_ids
) > n
:
1082 video_ids
= video_ids
[:n
]
1083 videos
= [self
.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids
]
1084 return self
.playlist_result(videos
, query
)
1087 class YoutubeShowIE(InfoExtractor
):
1088 IE_DESC
= u
'YouTube.com (multi-season) shows'
1089 _VALID_URL
= r
'https?://www\.youtube\.com/show/(.*)'
1090 IE_NAME
= u
'youtube:show'
1092 def _real_extract(self
, url
):
1093 mobj
= re
.match(self
._VALID
_URL
, url
)
1094 show_name
= mobj
.group(1)
1095 webpage
= self
._download
_webpage
(url
, show_name
, u
'Downloading show webpage')
1096 # There's one playlist for each season of the show
1097 m_seasons
= list(re
.finditer(r
'href="(/playlist\?list=.*?)"', webpage
))
1098 self
.to_screen(u
'%s: Found %s seasons' % (show_name
, len(m_seasons
)))
1099 return [self
.url_result('https://www.youtube.com' + season
.group(1), 'YoutubePlaylist') for season
in m_seasons
]
1102 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1104 Base class for extractors that fetch info from
1105 http://www.youtube.com/feed_ajax
1106 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1108 _LOGIN_REQUIRED
= True
1110 # use action_load_personal_feed instead of action_load_system_feed
1111 _PERSONAL_FEED
= False
1114 def _FEED_TEMPLATE(self
):
1115 action
= 'action_load_system_feed'
1116 if self
._PERSONAL
_FEED
:
1117 action
= 'action_load_personal_feed'
1118 return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action
, self
._FEED
_NAME
)
1122 return u
'youtube:%s' % self
._FEED
_NAME
1124 def _real_initialize(self
):
1127 def _real_extract(self
, url
):
1129 # The step argument is available only in 2.7 or higher
1130 for i
in itertools
.count(0):
1131 paging
= i
*self
._PAGING
_STEP
1132 info
= self
._download
_webpage
(self
._FEED
_TEMPLATE
% paging
,
1133 u
'%s feed' % self
._FEED
_NAME
,
1134 u
'Downloading page %s' % i
)
1135 info
= json
.loads(info
)
1136 feed_html
= info
['feed_html']
1137 m_ids
= re
.finditer(r
'"/watch\?v=(.*?)["&]', feed_html
)
1138 ids
= orderedSet(m
.group(1) for m
in m_ids
)
1139 feed_entries
.extend(self
.url_result(id, 'Youtube') for id in ids
)
1140 if info
['paging'] is None:
1142 return self
.playlist_result(feed_entries
, playlist_title
=self
._PLAYLIST
_TITLE
)
1144 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor
):
1145 IE_DESC
= u
'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
1146 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1147 _FEED_NAME
= 'subscriptions'
1148 _PLAYLIST_TITLE
= u
'Youtube Subscriptions'
1150 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
1151 IE_DESC
= u
'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1152 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1153 _FEED_NAME
= 'recommended'
1154 _PLAYLIST_TITLE
= u
'Youtube Recommended videos'
1156 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor
):
1157 IE_DESC
= u
'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1158 _VALID_URL
= r
'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1159 _FEED_NAME
= 'watch_later'
1160 _PLAYLIST_TITLE
= u
'Youtube Watch Later'
1162 _PERSONAL_FEED
= True
1164 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
1165 IE_NAME
= u
'youtube:favorites'
1166 IE_DESC
= u
'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1167 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:o?rites)?'
1168 _LOGIN_REQUIRED
= True
1170 def _real_extract(self
, url
):
1171 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1172 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, u
'favourites playlist id')
1173 return self
.url_result(playlist_id
, 'YoutubePlaylist')