8 from .common
import InfoExtractor
, SearchInfoExtractor
14 compat_urllib_request
,
25 class YoutubeIE(InfoExtractor
):
26 """Information extractor for youtube.com."""
30 (?:https?://)? # http(s):// (optional)
31 (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/|
32 tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains
33 (?:.*?\#/)? # handle anchor (#/) redirect urls
34 (?: # the various things that can precede the ID:
35 (?:(?:v|embed|e)/) # v/ or embed/ or e/
36 |(?: # or the v= param in all its forms
37 (?:watch(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
38 (?:\?|\#!?) # the params delimiter ? or # or #!
39 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
42 )? # optional -> youtube.com/xxxx is OK
43 )? # all until now is optional -> you can pass the naked ID
44 ([0-9A-Za-z_-]+) # here is it! the YouTube video ID
45 (?(1).+)? # if we found the ID, everything can follow
47 _LANG_URL
= r
'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
48 _LOGIN_URL
= 'https://accounts.google.com/ServiceLogin'
49 _AGE_URL
= 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
50 _NEXT_URL_RE
= r
'[\?&]next_url=([^&]+)'
51 _NETRC_MACHINE
= 'youtube'
52 # Listed in order of quality
53 _available_formats
= ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13']
54 _available_formats_prefer_free
= ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13']
86 u
"url": u
"http://www.youtube.com/watch?v=BaW_jenozKc",
87 u
"file": u
"BaW_jenozKc.mp4",
89 u
"title": u
"youtube-dl test video \"'/\\ä↭𝕐",
90 u
"uploader": u
"Philipp Hagemeister",
91 u
"uploader_id": u
"phihag",
92 u
"upload_date": u
"20121002",
93 u
"description": u
"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
97 u
"url": u
"http://www.youtube.com/watch?v=1ltcDfZMA3U",
98 u
"file": u
"1ltcDfZMA3U.flv",
99 u
"note": u
"Test VEVO video (#897)",
101 u
"upload_date": u
"20070518",
102 u
"title": u
"Maps - It Will Find You",
103 u
"description": u
"Music video by Maps performing It Will Find You.",
104 u
"uploader": u
"MuteUSA",
105 u
"uploader_id": u
"MuteUSA"
109 u
"url": u
"http://www.youtube.com/watch?v=UxxajLWwzqY",
110 u
"file": u
"UxxajLWwzqY.mp4",
111 u
"note": u
"Test generic use_cipher_signature video (#897)",
113 u
"upload_date": u
"20120506",
114 u
"title": u
"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
115 u
"description": u
"md5:b085c9804f5ab69f4adea963a2dceb3c",
116 u
"uploader": u
"IconaPop",
117 u
"uploader_id": u
"IconaPop"
124 def suitable(cls
, url
):
125 """Receives a URL and returns True if suitable for this IE."""
126 if YoutubePlaylistIE
.suitable(url
): return False
127 return re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
) is not None
129 def report_lang(self
):
130 """Report attempt to set language."""
131 self
.to_screen(u
'Setting language')
133 def report_login(self
):
134 """Report attempt to log in."""
135 self
.to_screen(u
'Logging in')
137 def report_video_webpage_download(self
, video_id
):
138 """Report attempt to download video webpage."""
139 self
.to_screen(u
'%s: Downloading video webpage' % video_id
)
141 def report_video_info_webpage_download(self
, video_id
):
142 """Report attempt to download video info webpage."""
143 self
.to_screen(u
'%s: Downloading video info webpage' % video_id
)
145 def report_video_subtitles_download(self
, video_id
):
146 """Report attempt to download video info webpage."""
147 self
.to_screen(u
'%s: Checking available subtitles' % video_id
)
149 def report_video_subtitles_request(self
, video_id
, sub_lang
, format
):
150 """Report attempt to download video info webpage."""
151 self
.to_screen(u
'%s: Downloading video subtitles for %s.%s' % (video_id
, sub_lang
, format
))
153 def report_video_subtitles_available(self
, video_id
, sub_lang_list
):
154 """Report available subtitles."""
155 sub_lang
= ",".join(list(sub_lang_list
.keys()))
156 self
.to_screen(u
'%s: Available subtitles for video: %s' % (video_id
, sub_lang
))
158 def report_information_extraction(self
, video_id
):
159 """Report attempt to extract video information."""
160 self
.to_screen(u
'%s: Extracting video information' % video_id
)
162 def report_unavailable_format(self
, video_id
, format
):
163 """Report extracted video URL."""
164 self
.to_screen(u
'%s: Format %s not available' % (video_id
, format
))
166 def report_rtmp_download(self
):
167 """Indicate the download will use the RTMP protocol."""
168 self
.to_screen(u
'RTMP download detected')
170 def _decrypt_signature(self
, s
):
171 """Decrypt the key"""
174 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]
176 return s
[62] + s
[82:62:-1] + s
[83] + s
[61:52:-1] + s
[0] + s
[51:2:-1]
178 return s
[2:63] + s
[82] + s
[64:82] + s
[63]
180 return s
[76] + s
[82:76:-1] + s
[83] + s
[75:60:-1] + s
[0] + s
[59:50:-1] + s
[1] + s
[49:2:-1]
182 return s
[83:36:-1] + s
[2] + s
[35:26:-1] + s
[3] + s
[25:3:-1] + s
[26]
184 return s
[52] + s
[81:55:-1] + s
[2] + s
[54:52:-1] + s
[82] + s
[51:36:-1] + s
[55] + s
[35:2:-1] + s
[36]
186 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]
189 raise ExtractorError(u
'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s
)))
191 def _get_available_subtitles(self
, video_id
):
192 self
.report_video_subtitles_download(video_id
)
193 request
= compat_urllib_request
.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
)
195 sub_list
= compat_urllib_request
.urlopen(request
).read().decode('utf-8')
196 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
197 return (u
'unable to download video subtitles: %s' % compat_str(err
), None)
198 sub_lang_list
= re
.findall(r
'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list
)
199 sub_lang_list
= dict((l
[1], l
[0]) for l
in sub_lang_list
)
200 if not sub_lang_list
:
201 return (u
'video doesn\'t have subtitles', None)
204 def _list_available_subtitles(self
, video_id
):
205 sub_lang_list
= self
._get
_available
_subtitles
(video_id
)
206 self
.report_video_subtitles_available(video_id
, sub_lang_list
)
208 def _request_subtitle(self
, sub_lang
, sub_name
, video_id
, format
):
211 (error_message, sub_lang, sub)
213 self
.report_video_subtitles_request(video_id
, sub_lang
, format
)
214 params
= compat_urllib_parse
.urlencode({
220 url
= 'http://www.youtube.com/api/timedtext?' + params
222 sub
= compat_urllib_request
.urlopen(url
).read().decode('utf-8')
223 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
224 return (u
'unable to download video subtitles: %s' % compat_str(err
), None, None)
226 return (u
'Did not fetch video subtitles', None, None)
227 return (None, sub_lang
, sub
)
229 def _request_automatic_caption(self
, video_id
, webpage
):
230 """We need the webpage for getting the captions url, pass it as an
231 argument to speed up the process."""
232 sub_lang
= self
._downloader
.params
.get('subtitleslang') or 'en'
233 sub_format
= self
._downloader
.params
.get('subtitlesformat')
234 self
.to_screen(u
'%s: Looking for automatic captions' % video_id
)
235 mobj
= re
.search(r
';ytplayer.config = ({.*?});', webpage
)
236 err_msg
= u
'Couldn\'t find automatic captions for "%s"' % sub_lang
238 return [(err_msg
, None, None)]
239 player_config
= json
.loads(mobj
.group(1))
241 args
= player_config
[u
'args']
242 caption_url
= args
[u
'ttsurl']
243 timestamp
= args
[u
'timestamp']
244 params
= compat_urllib_parse
.urlencode({
251 subtitles_url
= caption_url
+ '&' + params
252 sub
= self
._download
_webpage
(subtitles_url
, video_id
, u
'Downloading automatic captions')
253 return [(None, sub_lang
, sub
)]
255 return [(err_msg
, None, None)]
257 def _extract_subtitle(self
, video_id
):
259 Return a list with a tuple:
260 [(error_message, sub_lang, sub)]
262 sub_lang_list
= self
._get
_available
_subtitles
(video_id
)
263 sub_format
= self
._downloader
.params
.get('subtitlesformat')
264 if isinstance(sub_lang_list
,tuple): #There was some error, it didn't get the available subtitles
265 return [(sub_lang_list
[0], None, None)]
266 if self
._downloader
.params
.get('subtitleslang', False):
267 sub_lang
= self
._downloader
.params
.get('subtitleslang')
268 elif 'en' in sub_lang_list
:
271 sub_lang
= list(sub_lang_list
.keys())[0]
272 if not sub_lang
in sub_lang_list
:
273 return [(u
'no closed captions found in the specified language "%s"' % sub_lang
, None, None)]
275 subtitle
= self
._request
_subtitle
(sub_lang
, sub_lang_list
[sub_lang
].encode('utf-8'), video_id
, sub_format
)
278 def _extract_all_subtitles(self
, video_id
):
279 sub_lang_list
= self
._get
_available
_subtitles
(video_id
)
280 sub_format
= self
._downloader
.params
.get('subtitlesformat')
281 if isinstance(sub_lang_list
,tuple): #There was some error, it didn't get the available subtitles
282 return [(sub_lang_list
[0], None, None)]
284 for sub_lang
in sub_lang_list
:
285 subtitle
= self
._request
_subtitle
(sub_lang
, sub_lang_list
[sub_lang
].encode('utf-8'), video_id
, sub_format
)
286 subtitles
.append(subtitle
)
289 def _print_formats(self
, formats
):
290 print('Available formats:')
292 print('%s\t:\t%s\t[%s]' %(x
, self
._video
_extensions
.get(x
, 'flv'), self
._video
_dimensions
.get(x
, '???')))
294 def _real_initialize(self
):
295 if self
._downloader
is None:
300 downloader_params
= self
._downloader
.params
302 # Attempt to use provided username and password or .netrc data
303 if downloader_params
.get('username', None) is not None:
304 username
= downloader_params
['username']
305 password
= downloader_params
['password']
306 elif downloader_params
.get('usenetrc', False):
308 info
= netrc
.netrc().authenticators(self
._NETRC
_MACHINE
)
313 raise netrc
.NetrcParseError('No authenticators for %s' % self
._NETRC
_MACHINE
)
314 except (IOError, netrc
.NetrcParseError
) as err
:
315 self
._downloader
.report_warning(u
'parsing .netrc: %s' % compat_str(err
))
319 request
= compat_urllib_request
.Request(self
._LANG
_URL
)
322 compat_urllib_request
.urlopen(request
).read()
323 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
324 self
._downloader
.report_warning(u
'unable to set language: %s' % compat_str(err
))
327 # No authentication to be performed
331 request
= compat_urllib_request
.Request(self
._LOGIN
_URL
)
333 login_page
= compat_urllib_request
.urlopen(request
).read().decode('utf-8')
334 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
335 self
._downloader
.report_warning(u
'unable to fetch login page: %s' % compat_str(err
))
340 match
= re
.search(re
.compile(r
'<input.+?name="GALX".+?value="(.+?)"', re
.DOTALL
), login_page
)
342 galx
= match
.group(1)
344 match
= re
.search(re
.compile(r
'<input.+?name="dsh".+?value="(.+?)"', re
.DOTALL
), login_page
)
350 u
'continue': u
'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
354 u
'PersistentCookie': u
'yes',
356 u
'bgresponse': u
'js_disabled',
357 u
'checkConnection': u
'',
358 u
'checkedDomains': u
'youtube',
364 u
'signIn': u
'Sign in',
366 u
'service': u
'youtube',
370 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
372 login_form
= dict((k
.encode('utf-8'), v
.encode('utf-8')) for k
,v
in login_form_strs
.items())
373 login_data
= compat_urllib_parse
.urlencode(login_form
).encode('ascii')
374 request
= compat_urllib_request
.Request(self
._LOGIN
_URL
, login_data
)
377 login_results
= compat_urllib_request
.urlopen(request
).read().decode('utf-8')
378 if re
.search(r
'(?i)<form[^>]* id="gaia_loginform"', login_results
) is not None:
379 self
._downloader
.report_warning(u
'unable to log in: bad username or password')
381 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
382 self
._downloader
.report_warning(u
'unable to log in: %s' % compat_str(err
))
388 'action_confirm': 'Confirm',
390 request
= compat_urllib_request
.Request(self
._AGE
_URL
, compat_urllib_parse
.urlencode(age_form
))
392 self
.report_age_confirmation()
393 compat_urllib_request
.urlopen(request
).read().decode('utf-8')
394 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
395 raise ExtractorError(u
'Unable to confirm age: %s' % compat_str(err
))
397 def _extract_id(self
, url
):
398 mobj
= re
.match(self
._VALID
_URL
, url
, re
.VERBOSE
)
400 raise ExtractorError(u
'Invalid URL: %s' % url
)
401 video_id
= mobj
.group(2)
404 def _real_extract(self
, url
):
405 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
406 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
408 url
= 'https://www.youtube.com/' + compat_urllib_parse
.unquote(mobj
.group(1)).lstrip('/')
409 video_id
= self
._extract
_id
(url
)
412 self
.report_video_webpage_download(video_id
)
413 url
= 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
414 request
= compat_urllib_request
.Request(url
)
416 video_webpage_bytes
= compat_urllib_request
.urlopen(request
).read()
417 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
418 raise ExtractorError(u
'Unable to download video webpage: %s' % compat_str(err
))
420 video_webpage
= video_webpage_bytes
.decode('utf-8', 'ignore')
422 # Attempt to extract SWF player URL
423 mobj
= re
.search(r
'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
425 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
430 self
.report_video_info_webpage_download(video_id
)
431 for el_type
in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
432 video_info_url
= ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
433 % (video_id
, el_type
))
434 video_info_webpage
= self
._download
_webpage
(video_info_url
, video_id
,
436 errnote
='unable to download video info webpage')
437 video_info
= compat_parse_qs(video_info_webpage
)
438 if 'token' in video_info
:
440 if 'token' not in video_info
:
441 if 'reason' in video_info
:
442 raise ExtractorError(u
'YouTube said: %s' % video_info
['reason'][0])
444 raise ExtractorError(u
'"token" parameter not in video info for unknown reason')
446 # Check for "rental" videos
447 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
448 raise ExtractorError(u
'"rental" videos not supported')
450 # Start extracting information
451 self
.report_information_extraction(video_id
)
454 if 'author' not in video_info
:
455 raise ExtractorError(u
'Unable to extract uploader name')
456 video_uploader
= compat_urllib_parse
.unquote_plus(video_info
['author'][0])
459 video_uploader_id
= None
460 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
462 video_uploader_id
= mobj
.group(1)
464 self
._downloader
.report_warning(u
'unable to extract uploader nickname')
467 if 'title' not in video_info
:
468 raise ExtractorError(u
'Unable to extract video title')
469 video_title
= compat_urllib_parse
.unquote_plus(video_info
['title'][0])
472 if 'thumbnail_url' not in video_info
:
473 self
._downloader
.report_warning(u
'unable to extract video thumbnail')
475 else: # don't panic if we can't find it
476 video_thumbnail
= compat_urllib_parse
.unquote_plus(video_info
['thumbnail_url'][0])
480 mobj
= re
.search(r
'id="eow-date.*?>(.*?)</span>', video_webpage
, re
.DOTALL
)
482 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
483 upload_date
= unified_strdate(upload_date
)
486 video_description
= get_element_by_id("eow-description", video_webpage
)
487 if video_description
:
488 video_description
= clean_html(video_description
)
490 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
492 video_description
= unescapeHTML(fd_mobj
.group(1))
494 video_description
= u
''
497 video_subtitles
= None
499 if self
._downloader
.params
.get('writesubtitles', False):
500 video_subtitles
= self
._extract
_subtitle
(video_id
)
502 (sub_error
, sub_lang
, sub
) = video_subtitles
[0]
504 self
._downloader
.report_warning(sub_error
)
506 if self
._downloader
.params
.get('writeautomaticsub', False):
507 video_subtitles
= self
._request
_automatic
_caption
(video_id
, video_webpage
)
508 (sub_error
, sub_lang
, sub
) = video_subtitles
[0]
510 self
._downloader
.report_warning(sub_error
)
512 if self
._downloader
.params
.get('allsubtitles', False):
513 video_subtitles
= self
._extract
_all
_subtitles
(video_id
)
514 for video_subtitle
in video_subtitles
:
515 (sub_error
, sub_lang
, sub
) = video_subtitle
517 self
._downloader
.report_warning(sub_error
)
519 if self
._downloader
.params
.get('listsubtitles', False):
520 self
._list
_available
_subtitles
(video_id
)
523 if 'length_seconds' not in video_info
:
524 self
._downloader
.report_warning(u
'unable to extract video duration')
527 video_duration
= compat_urllib_parse
.unquote_plus(video_info
['length_seconds'][0])
529 # Decide which formats to download
530 req_format
= self
._downloader
.params
.get('format', None)
533 mobj
= re
.search(r
';ytplayer.config = ({.*?});', video_webpage
)
535 raise ValueError('Could not find vevo ID')
536 info
= json
.loads(mobj
.group(1))
538 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
539 # this signatures are encrypted
540 m_s
= re
.search(r
'[&,]s=', args
['url_encoded_fmt_stream_map'])
542 self
.to_screen(u
'%s: Encrypted signatures detected.' % video_id
)
543 video_info
['url_encoded_fmt_stream_map'] = [args
['url_encoded_fmt_stream_map']]
547 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
548 self
.report_rtmp_download()
549 video_url_list
= [(None, video_info
['conn'][0])]
550 elif 'url_encoded_fmt_stream_map' in video_info
and len(video_info
['url_encoded_fmt_stream_map']) >= 1:
552 for url_data_str
in video_info
['url_encoded_fmt_stream_map'][0].split(','):
553 url_data
= compat_parse_qs(url_data_str
)
554 if 'itag' in url_data
and 'url' in url_data
:
555 url
= url_data
['url'][0]
556 if 'sig' in url_data
:
557 url
+= '&signature=' + url_data
['sig'][0]
558 elif 's' in url_data
:
559 if self
._downloader
.params
.get('verbose'):
561 player
= self
._search
_regex
(r
'html5player-(.+?)\.js', video_webpage
,
562 'html5 player', fatal
=False)
563 self
.to_screen('encrypted signature length %d (%d.%d), itag %s, html5 player %s' %
564 (len(s
), len(s
.split('.')[0]), len(s
.split('.')[1]), url_data
['itag'][0], player
))
565 signature
= self
._decrypt
_signature
(url_data
['s'][0])
566 url
+= '&signature=' + signature
567 if 'ratebypass' not in url
:
568 url
+= '&ratebypass=yes'
569 url_map
[url_data
['itag'][0]] = url
571 format_limit
= self
._downloader
.params
.get('format_limit', None)
572 available_formats
= self
._available
_formats
_prefer
_free
if self
._downloader
.params
.get('prefer_free_formats', False) else self
._available
_formats
573 if format_limit
is not None and format_limit
in available_formats
:
574 format_list
= available_formats
[available_formats
.index(format_limit
):]
576 format_list
= available_formats
577 existing_formats
= [x
for x
in format_list
if x
in url_map
]
578 if len(existing_formats
) == 0:
579 raise ExtractorError(u
'no known formats available for video')
580 if self
._downloader
.params
.get('listformats', None):
581 self
._print
_formats
(existing_formats
)
583 if req_format
is None or req_format
== 'best':
584 video_url_list
= [(existing_formats
[0], url_map
[existing_formats
[0]])] # Best quality
585 elif req_format
== 'worst':
586 video_url_list
= [(existing_formats
[len(existing_formats
)-1], url_map
[existing_formats
[len(existing_formats
)-1]])] # worst quality
587 elif req_format
in ('-1', 'all'):
588 video_url_list
= [(f
, url_map
[f
]) for f
in existing_formats
] # All formats
590 # Specific formats. We pick the first in a slash-delimeted sequence.
591 # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
592 req_formats
= req_format
.split('/')
593 video_url_list
= None
594 for rf
in req_formats
:
596 video_url_list
= [(rf
, url_map
[rf
])]
598 if video_url_list
is None:
599 raise ExtractorError(u
'requested format not available')
601 raise ExtractorError(u
'no conn or url_encoded_fmt_stream_map information found in video info')
604 for format_param
, video_real_url
in video_url_list
:
606 video_extension
= self
._video
_extensions
.get(format_param
, 'flv')
608 video_format
= '{0} - {1}'.format(format_param
if format_param
else video_extension
,
609 self
._video
_dimensions
.get(format_param
, '???'))
613 'url': video_real_url
,
614 'uploader': video_uploader
,
615 'uploader_id': video_uploader_id
,
616 'upload_date': upload_date
,
617 'title': video_title
,
618 'ext': video_extension
,
619 'format': video_format
,
620 'thumbnail': video_thumbnail
,
621 'description': video_description
,
622 'player_url': player_url
,
623 'subtitles': video_subtitles
,
624 'duration': video_duration
628 class YoutubePlaylistIE(InfoExtractor
):
629 """Information Extractor for YouTube playlists."""
636 (?:course|view_play_list|my_playlists|artist|playlist|watch)
637 \? (?:.*?&)*? (?:p|a|list)=
640 ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,})
643 ((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
645 _TEMPLATE_URL
= 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
647 IE_NAME
= u
'youtube:playlist'
650 def suitable(cls
, url
):
651 """Receives a URL and returns True if suitable for this IE."""
652 return re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
) is not None
654 def _real_extract(self
, url
):
655 # Extract playlist id
656 mobj
= re
.match(self
._VALID
_URL
, url
, re
.VERBOSE
)
658 raise ExtractorError(u
'Invalid URL: %s' % url
)
660 # Download playlist videos from API
661 playlist_id
= mobj
.group(1) or mobj
.group(2)
666 url
= self
._TEMPLATE
_URL
% (playlist_id
, self
._MAX
_RESULTS
, self
._MAX
_RESULTS
* (page_num
- 1) + 1)
667 page
= self
._download
_webpage
(url
, playlist_id
, u
'Downloading page #%s' % page_num
)
670 response
= json
.loads(page
)
671 except ValueError as err
:
672 raise ExtractorError(u
'Invalid JSON in API response: ' + compat_str(err
))
674 if 'feed' not in response
:
675 raise ExtractorError(u
'Got a malformed response from YouTube API')
676 playlist_title
= response
['feed']['title']['$t']
677 if 'entry' not in response
['feed']:
678 # Number of videos is a multiple of self._MAX_RESULTS
681 for entry
in response
['feed']['entry']:
682 index
= entry
['yt$position']['$t']
683 if 'media$group' in entry
and 'media$player' in entry
['media$group']:
684 videos
.append((index
, entry
['media$group']['media$player']['url']))
686 if len(response
['feed']['entry']) < self
._MAX
_RESULTS
:
690 videos
= [v
[1] for v
in sorted(videos
)]
692 url_results
= [self
.url_result(url
, 'Youtube') for url
in videos
]
693 return [self
.playlist_result(url_results
, playlist_id
, playlist_title
)]
696 class YoutubeChannelIE(InfoExtractor
):
697 """Information Extractor for YouTube channels."""
699 _VALID_URL
= r
"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
700 _TEMPLATE_URL
= 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
701 _MORE_PAGES_INDICATOR
= 'yt-uix-load-more'
702 _MORE_PAGES_URL
= 'http://www.youtube.com/channel_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
703 IE_NAME
= u
'youtube:channel'
705 def extract_videos_from_page(self
, page
):
707 for mobj
in re
.finditer(r
'href="/watch\?v=([0-9A-Za-z_-]+)&?', page
):
708 if mobj
.group(1) not in ids_in_page
:
709 ids_in_page
.append(mobj
.group(1))
712 def _real_extract(self
, url
):
714 mobj
= re
.match(self
._VALID
_URL
, url
)
716 raise ExtractorError(u
'Invalid URL: %s' % url
)
718 # Download channel page
719 channel_id
= mobj
.group(1)
723 url
= self
._TEMPLATE
_URL
% (channel_id
, pagenum
)
724 page
= self
._download
_webpage
(url
, channel_id
,
725 u
'Downloading page #%s' % pagenum
)
727 # Extract video identifiers
728 ids_in_page
= self
.extract_videos_from_page(page
)
729 video_ids
.extend(ids_in_page
)
731 # Download any subsequent channel pages using the json-based channel_ajax query
732 if self
._MORE
_PAGES
_INDICATOR
in page
:
734 pagenum
= pagenum
+ 1
736 url
= self
._MORE
_PAGES
_URL
% (pagenum
, channel_id
)
737 page
= self
._download
_webpage
(url
, channel_id
,
738 u
'Downloading page #%s' % pagenum
)
740 page
= json
.loads(page
)
742 ids_in_page
= self
.extract_videos_from_page(page
['content_html'])
743 video_ids
.extend(ids_in_page
)
745 if self
._MORE
_PAGES
_INDICATOR
not in page
['load_more_widget_html']:
748 self
._downloader
.to_screen(u
'[youtube] Channel %s: Found %i videos' % (channel_id
, len(video_ids
)))
750 urls
= ['http://www.youtube.com/watch?v=%s' % id for id in video_ids
]
751 url_entries
= [self
.url_result(url
, 'Youtube') for url
in urls
]
752 return [self
.playlist_result(url_entries
, channel_id
)]
755 class YoutubeUserIE(InfoExtractor
):
756 """Information Extractor for YouTube users."""
758 _VALID_URL
= r
'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
759 _TEMPLATE_URL
= 'http://gdata.youtube.com/feeds/api/users/%s'
760 _GDATA_PAGE_SIZE
= 50
761 _GDATA_URL
= 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
762 _VIDEO_INDICATOR
= r
'/watch\?v=(.+?)[\<&]'
763 IE_NAME
= u
'youtube:user'
765 def _real_extract(self
, url
):
767 mobj
= re
.match(self
._VALID
_URL
, url
)
769 raise ExtractorError(u
'Invalid URL: %s' % url
)
771 username
= mobj
.group(1)
773 # Download video ids using YouTube Data API. Result size per
774 # query is limited (currently to 50 videos) so we need to query
775 # page by page until there are no video ids - it means we got
782 start_index
= pagenum
* self
._GDATA
_PAGE
_SIZE
+ 1
784 gdata_url
= self
._GDATA
_URL
% (username
, self
._GDATA
_PAGE
_SIZE
, start_index
)
785 page
= self
._download
_webpage
(gdata_url
, username
,
786 u
'Downloading video ids from %d to %d' % (start_index
, start_index
+ self
._GDATA
_PAGE
_SIZE
))
788 # Extract video identifiers
791 for mobj
in re
.finditer(self
._VIDEO
_INDICATOR
, page
):
792 if mobj
.group(1) not in ids_in_page
:
793 ids_in_page
.append(mobj
.group(1))
795 video_ids
.extend(ids_in_page
)
797 # A little optimization - if current page is not
798 # "full", ie. does not contain PAGE_SIZE video ids then
799 # we can assume that this page is the last one - there
800 # are no more ids on further pages - no need to query
803 if len(ids_in_page
) < self
._GDATA
_PAGE
_SIZE
:
808 urls
= ['http://www.youtube.com/watch?v=%s' % video_id
for video_id
in video_ids
]
809 url_results
= [self
.url_result(url
, 'Youtube') for url
in urls
]
810 return [self
.playlist_result(url_results
, playlist_title
= username
)]
812 class YoutubeSearchIE(SearchInfoExtractor
):
813 """Information Extractor for YouTube search queries."""
814 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
816 IE_NAME
= u
'youtube:search'
817 _SEARCH_KEY
= 'ytsearch'
819 def report_download_page(self
, query
, pagenum
):
820 """Report attempt to download search page with given number."""
821 self
._downloader
.to_screen(u
'[youtube] query "%s": Downloading page %s' % (query
, pagenum
))
823 def _get_n_results(self
, query
, n
):
824 """Get a specified number of results for a query"""
830 while (50 * pagenum
) < limit
:
831 self
.report_download_page(query
, pagenum
+1)
832 result_url
= self
._API
_URL
% (compat_urllib_parse
.quote_plus(query
), (50*pagenum
)+1)
833 request
= compat_urllib_request
.Request(result_url
)
835 data
= compat_urllib_request
.urlopen(request
).read().decode('utf-8')
836 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
837 raise ExtractorError(u
'Unable to download API page: %s' % compat_str(err
))
838 api_response
= json
.loads(data
)['data']
840 if not 'items' in api_response
:
841 raise ExtractorError(u
'[youtube] No video results')
843 new_ids
= list(video
['id'] for video
in api_response
['items'])
846 limit
= min(n
, api_response
['totalItems'])
849 if len(video_ids
) > n
:
850 video_ids
= video_ids
[:n
]
851 videos
= [self
.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids
]
852 return self
.playlist_result(videos
, query
)