import itertools
from .common import InfoExtractor, SearchInfoExtractor
+from .subtitles import SubtitlesIE
from ..utils import (
compat_http_client,
compat_parse_qs,
orderedSet,
)
-class YoutubeBaseInfoExtractor(InfoExtractor):
- """Provide base functions for Youtube extractors"""
- _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
- _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
- _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
- _NETRC_MACHINE = 'youtube'
- # If True it will raise an error if no login info is provided
- _LOGIN_REQUIRED = False
-
- def report_lang(self):
- """Report attempt to set language."""
- self.to_screen(u'Setting language')
-
- def _set_language(self):
- request = compat_urllib_request.Request(self._LANG_URL)
- try:
- self.report_lang()
- compat_urllib_request.urlopen(request).read()
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
- return False
- return True
-
- def _login(self):
- (username, password) = self._get_login_info()
- # No authentication to be performed
- if username is None:
- if self._LOGIN_REQUIRED:
- raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
- return False
-
- request = compat_urllib_request.Request(self._LOGIN_URL)
- try:
- login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
- return False
-
- galx = None
- dsh = None
- match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
- if match:
- galx = match.group(1)
- match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
- if match:
- dsh = match.group(1)
-
- # Log in
- login_form_strs = {
- u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
- u'Email': username,
- u'GALX': galx,
- u'Passwd': password,
- u'PersistentCookie': u'yes',
- u'_utf8': u'霱',
- u'bgresponse': u'js_disabled',
- u'checkConnection': u'',
- u'checkedDomains': u'youtube',
- u'dnConn': u'',
- u'dsh': dsh,
- u'pstMsg': u'0',
- u'rmShown': u'1',
- u'secTok': u'',
- u'signIn': u'Sign in',
- u'timeStmp': u'',
- u'service': u'youtube',
- u'uilel': u'3',
- u'hl': u'en_US',
- }
- # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
- # chokes on unicode
- login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
- login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
- request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
+
+class YoutubeSubtitlesIE(SubtitlesIE):
+
+ def _get_available_subtitles(self, video_id):
+ request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
try:
- self.report_login()
- login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
- if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
- self._downloader.report_warning(u'unable to log in: bad username or password')
- return False
+ sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
- return False
- return True
-
- def _confirm_age(self):
- age_form = {
- 'next_url': '/',
- 'action_confirm': 'Confirm',
- }
- request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
+ self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
+ return {}
+ lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
+
+ sub_lang_list = {}
+ for l in lang_list:
+ lang = l[1]
+ params = compat_urllib_parse.urlencode({
+ 'lang': lang,
+ 'v': video_id,
+ 'fmt': self._downloader.params.get('subtitlesformat'),
+ })
+ url = u'http://www.youtube.com/api/timedtext?' + params
+ sub_lang_list[lang] = url
+ if not sub_lang_list:
+ self._downloader.report_warning(u'video doesn\'t have subtitles')
+ return {}
+ return sub_lang_list
+
+ def _request_automatic_caption(self, video_id, webpage):
+ """We need the webpage for getting the captions url, pass it as an
+ argument to speed up the process."""
+ sub_lang = self._downloader.params.get('subtitleslang') or 'en'
+ sub_format = self._downloader.params.get('subtitlesformat')
+ self.to_screen(u'%s: Looking for automatic captions' % video_id)
+ mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
+ err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
+ if mobj is None:
+ self._downloader.report_warning(err_msg)
+ return {}
+ player_config = json.loads(mobj.group(1))
try:
- self.report_age_confirmation()
- compat_urllib_request.urlopen(request).read().decode('utf-8')
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
- return True
+ args = player_config[u'args']
+ caption_url = args[u'ttsurl']
+ timestamp = args[u'timestamp']
+ params = compat_urllib_parse.urlencode({
+ 'lang': 'en',
+ 'tlang': sub_lang,
+ 'fmt': sub_format,
+ 'ts': timestamp,
+ 'kind': 'asr',
+ })
+ subtitles_url = caption_url + '&' + params
+ sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
+ return {sub_lang: sub}
+ # An extractor error can be raise by the download process if there are
+ # no automatic captions but there are subtitles
+ except (KeyError, ExtractorError):
+ self._downloader.report_warning(err_msg)
+ return {}
- def _real_initialize(self):
- if self._downloader is None:
- return
- if not self._set_language():
- return
- if not self._login():
- return
- self._confirm_age()
-class YoutubeIE(YoutubeBaseInfoExtractor):
+class YoutubeIE(YoutubeSubtitlesIE):
IE_DESC = u'YouTube.com'
_VALID_URL = r"""^
(
# Listed in order of quality
_available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13',
'95', '94', '93', '92', '132', '151',
+ # 3D
'85', '84', '102', '83', '101', '82', '100',
+ # Dash video
+ '138', '137', '248', '136', '247', '135', '246',
+ '245', '244', '134', '243', '133', '242', '160',
+ # Dash audio
+ '141', '172', '140', '171', '139',
]
_available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13',
'95', '94', '93', '92', '132', '151',
'85', '102', '84', '101', '83', '100', '82',
+ # Dash video
+ '138', '248', '137', '247', '136', '246', '245',
+ '244', '135', '243', '134', '242', '133', '160',
+ # Dash audio
+ '172', '141', '171', '140', '139',
]
_video_extensions = {
'13': '3gp',
'100': 'webm',
'101': 'webm',
'102': 'webm',
-
+
# videos that use m3u8
'92': 'mp4',
'93': 'mp4',
'96': 'mp4',
'132': 'mp4',
'151': 'mp4',
+
+ # Dash mp4
+ '133': 'mp4',
+ '134': 'mp4',
+ '135': 'mp4',
+ '136': 'mp4',
+ '137': 'mp4',
+ '138': 'mp4',
+ '139': 'mp4',
+ '140': 'mp4',
+ '141': 'mp4',
+ '160': 'mp4',
+
+ # Dash webm
+ '171': 'webm',
+ '172': 'webm',
+ '242': 'webm',
+ '243': 'webm',
+ '244': 'webm',
+ '245': 'webm',
+ '246': 'webm',
+ '247': 'webm',
+ '248': 'webm',
}
_video_dimensions = {
'5': '240x400',
'96': '1080p',
'100': '360p',
'101': '480p',
- '102': '720p',
+ '102': '720p',
'132': '240p',
'151': '72p',
+ '133': '240p',
+ '134': '360p',
+ '135': '480p',
+ '136': '720p',
+ '137': '1080p',
+ '138': '>1080p',
+ '139': '48k',
+ '140': '128k',
+ '141': '256k',
+ '160': '192p',
+ '171': '128k',
+ '172': '256k',
+ '242': '240p',
+ '243': '360p',
+ '244': '480p',
+ '245': '480p',
+ '246': '480p',
+ '247': '720p',
+ '248': '1080p',
}
- _3d_itags = ['85', '84', '102', '83', '101', '82', '100']
+ _special_itags = {
+ '82': '3D',
+ '83': '3D',
+ '84': '3D',
+ '85': '3D',
+ '100': '3D',
+ '101': '3D',
+ '102': '3D',
+ '133': 'DASH Video',
+ '134': 'DASH Video',
+ '135': 'DASH Video',
+ '136': 'DASH Video',
+ '137': 'DASH Video',
+ '138': 'DASH Video',
+ '139': 'DASH Audio',
+ '140': 'DASH Audio',
+ '141': 'DASH Audio',
+ '160': 'DASH Video',
+ '171': 'DASH Audio',
+ '172': 'DASH Audio',
+ '242': 'DASH Video',
+ '243': 'DASH Video',
+ '244': 'DASH Video',
+ '245': 'DASH Video',
+ '246': 'DASH Video',
+ '247': 'DASH Video',
+ '248': 'DASH Video',
+ }
+
IE_NAME = u'youtube'
_TESTS = [
{
u"upload_date": u"20120506",
u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c",
- u"uploader": u"IconaPop",
+ u"uploader": u"Icona Pop",
u"uploader_id": u"IconaPop"
}
},
"""Report attempt to download video info webpage."""
self.to_screen(u'%s: Downloading video info webpage' % video_id)
- def report_video_subtitles_download(self, video_id):
- """Report attempt to download video info webpage."""
- self.to_screen(u'%s: Checking available subtitles' % video_id)
-
- def report_video_subtitles_request(self, video_id, sub_lang, format):
- """Report attempt to download video info webpage."""
- self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
-
- def report_video_subtitles_available(self, video_id, sub_lang_list):
- """Report available subtitles."""
- sub_lang = ",".join(list(sub_lang_list.keys()))
- self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
-
def report_information_extraction(self, video_id):
"""Report attempt to extract video information."""
self.to_screen(u'%s: Extracting video information' % video_id)
return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]
elif len(s) == 90:
return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81]
+ elif len(s) == 89:
+ return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1]
elif len(s) == 88:
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]
elif len(s) == 87:
- return s[83:53:-1] + s[3] + s[52:40:-1] + s[86] + s[39:10:-1] + s[0] + s[9:3:-1] + s[53]
+ return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:]
elif len(s) == 86:
return s[5:20] + s[2] + s[21:]
elif len(s) == 85:
- return s[2:8] + s[0] + s[9:21] + s[65] + s[22:65] + s[84] + s[66:82] + s[21]
+ 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]
elif len(s) == 84:
return s[83:27:-1] + s[0] + s[26:5:-1] + s[2:0:-1] + s[27]
elif len(s) == 83:
- return s[:15] + s[80] + s[16:80] + s[15]
+ 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]
elif len(s) == 82:
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]
elif len(s) == 81:
for x in formats:
print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'),
self._video_dimensions.get(x, '???'),
- ' (3D)' if x in self._3d_itags else ''))
+ ' ('+self._special_itags[x]+')' if x in self._special_itags else ''))
def _extract_id(self, url):
mobj = re.match(self._VALID_URL, url, re.VERBOSE)
# subtitles
video_subtitles = None
- if self._downloader.params.get('writesubtitles', False):
- video_subtitles = self._extract_subtitle(video_id)
- if video_subtitles:
- (sub_error, sub_lang, sub) = video_subtitles[0]
- if sub_error:
- self._downloader.report_warning(sub_error)
-
- if self._downloader.params.get('writeautomaticsub', False):
+ if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
+ video_subtitles = self._extract_subtitles(video_id)
+ elif self._downloader.params.get('writeautomaticsub', False):
video_subtitles = self._request_automatic_caption(video_id, video_webpage)
- (sub_error, sub_lang, sub) = video_subtitles[0]
- if sub_error:
- self._downloader.report_warning(sub_error)
-
- if self._downloader.params.get('allsubtitles', False):
- video_subtitles = self._extract_all_subtitles(video_id)
- for video_subtitle in video_subtitles:
- (sub_error, sub_lang, sub) = video_subtitle
- if sub_error:
- self._downloader.report_warning(sub_error)
if self._downloader.params.get('listsubtitles', False):
self._list_available_subtitles(video_id)
if m_s is not None:
self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
+ m_s = re.search(r'[&,]s=', args.get('adaptive_fmts', u''))
+ if m_s is not None:
+ if 'url_encoded_fmt_stream_map' in video_info:
+ video_info['url_encoded_fmt_stream_map'][0] += ',' + args['adaptive_fmts']
+ else:
+ video_info['url_encoded_fmt_stream_map'] = [args['adaptive_fmts']]
+ elif 'adaptive_fmts' in video_info:
+ if 'url_encoded_fmt_stream_map' in video_info:
+ video_info['url_encoded_fmt_stream_map'][0] += ',' + video_info['adaptive_fmts'][0]
+ else:
+ video_info['url_encoded_fmt_stream_map'] = video_info['adaptive_fmts']
except ValueError:
pass
video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension,
self._video_dimensions.get(format_param, '???'),
- ' (3D)' if format_param in self._3d_itags else '')
+ ' ('+self._special_itags[format_param]+')' if format_param in self._special_itags else '')
results.append({
'id': video_id,