class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
IE_DESC = u'YouTube.com'
- _VALID_URL = r"""^
+ _VALID_URL = r"""(?x)^
(
- (?:https?://)? # http(s):// (optional)
- (?:(?:(?:(?:\w+\.)?youtube(?:-nocookie)?\.com/|
+ (?:https?://|//)? # http(s):// or protocol-independent URL (optional)
+ (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
tube\.majestyc\.net/|
youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
(?:.*?\#/)? # handle anchor (#/) redirect urls
u"uploader_id": u"justintimberlakeVEVO"
}
},
+ {
+ u"url": u"//www.YouTube.com/watch?v=yZIXLfi8CZQ",
+ u"file": u"yZIXLfi8CZQ.mp4",
+ u"note": u"Embed-only video (#1746)",
+ u"info_dict": {
+ u"upload_date": u"20120608",
+ u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012",
+ u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7",
+ u"uploader": u"SET India",
+ u"uploader_id": u"setindia"
+ }
+ },
]
def suitable(cls, url):
"""Receives a URL and returns True if suitable for this IE."""
if YoutubePlaylistIE.suitable(url): return False
- return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
+ return re.match(cls._VALID_URL, url) is not None
def __init__(self, *args, **kwargs):
super(YoutubeIE, self).__init__(*args, **kwargs)
"""Turn the encrypted s field into a working signature"""
if player_url is not None:
+ if player_url.startswith(u'//'):
+ player_url = u'https:' + player_url
try:
player_id = (player_url, len(s))
if player_id not in self._player_cache:
params = compat_urllib_parse.urlencode({
'lang': lang,
'v': video_id,
- 'fmt': self._downloader.params.get('subtitlesformat'),
+ 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
'name': l[0].encode('utf-8'),
})
url = u'http://www.youtube.com/api/timedtext?' + params
def _get_available_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_format = self._downloader.params.get('subtitlesformat')
+ sub_format = self._downloader.params.get('subtitlesformat', 'srt')
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' % video_id
# We simulate the access to the video from www.youtube.com/v/{video_id}
# this can be viewed without login into Youtube
data = compat_urllib_parse.urlencode({'video_id': video_id,
- 'el': 'embedded',
+ 'el': 'player_embedded',
'gl': 'US',
'hl': 'en',
'eurl': 'https://youtube.googleapis.com/v/' + video_id,
else:
raise ExtractorError(u'"token" parameter not in video info for unknown reason')
+ if 'view_count' in video_info:
+ view_count = int(video_info['view_count'][0])
+ else:
+ view_count = None
+
# Check for "rental" videos
if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
raise ExtractorError(u'"rental" videos not supported')
'age_limit': 18 if age_gate else 0,
'annotations': video_annotations,
'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id,
+ 'view_count': view_count,
})
return results
- class YoutubePlaylistIE(InfoExtractor):
+ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
IE_DESC = u'YouTube.com playlists'
_VALID_URL = r"""(?:
(?:https?://)?
|
((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,})
)"""
- _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
- _MAX_RESULTS = 50
+ _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s'
+ _MORE_PAGES_INDICATOR = r'data-link-type="next"'
+ _VIDEO_RE = r'href="/watch\?v=([0-9A-Za-z_-]{11})&'
IE_NAME = u'youtube:playlist'
@classmethod
"""Receives a URL and returns True if suitable for this IE."""
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
+ def _real_initialize(self):
+ self._login()
+
def _real_extract(self, url):
# Extract playlist id
mobj = re.match(self._VALID_URL, url, re.VERBOSE)
else:
self.to_screen(u'Downloading playlist PL%s - add --no-playlist to just download video %s' % (playlist_id, video_id))
- # Download playlist videos from API
- videos = []
+ # Extract the video ids from the playlist pages
+ ids = []
for page_num in itertools.count(1):
- start_index = self._MAX_RESULTS * (page_num - 1) + 1
- if start_index >= 1000:
- self._downloader.report_warning(u'Max number of results reached')
- break
- url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index)
+ url = self._TEMPLATE_URL % (playlist_id, page_num)
page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
+ # The ids are duplicated
+ new_ids = orderedSet(re.findall(self._VIDEO_RE, page))
+ ids.extend(new_ids)
- try:
- response = json.loads(page)
- except ValueError as err:
- raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
-
- if 'feed' not in response:
- raise ExtractorError(u'Got a malformed response from YouTube API')
- playlist_title = response['feed']['title']['$t']
- if 'entry' not in response['feed']:
- # Number of videos is a multiple of self._MAX_RESULTS
+ if re.search(self._MORE_PAGES_INDICATOR, page) is None:
break
- for entry in response['feed']['entry']:
- index = entry['yt$position']['$t']
- if 'media$group' in entry and 'yt$videoid' in entry['media$group']:
- videos.append((
- index,
- 'https://www.youtube.com/watch?v=' + entry['media$group']['yt$videoid']['$t']
- ))
-
- videos = [v[1] for v in sorted(videos)]
+ playlist_title = self._og_search_title(page)
- url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
- return [self.playlist_result(url_results, playlist_id, playlist_title)]
+ url_results = [self.url_result(vid, 'Youtube') for vid in ids]
+ return self.playlist_result(url_results, playlist_id, playlist_title)
class YoutubeChannelIE(InfoExtractor):
# Download channel page
channel_id = mobj.group(1)
video_ids = []
+ url = 'https://www.youtube.com/channel/%s/videos' % channel_id
+ channel_page = self._download_webpage(url, channel_id)
+ if re.search(r'channel-header-autogenerated-label', channel_page) is not None:
+ autogenerated = True
+ else:
+ autogenerated = False
- # Download all channel pages using the json-based channel_ajax query
- for pagenum in itertools.count(1):
- url = self._MORE_PAGES_URL % (pagenum, channel_id)
- page = self._download_webpage(url, channel_id,
- u'Downloading page #%s' % pagenum)
-
- page = json.loads(page)
-
- ids_in_page = self.extract_videos_from_page(page['content_html'])
- video_ids.extend(ids_in_page)
-
- if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
- break
+ if autogenerated:
+ # The videos are contained in a single page
+ # the ajax pages can't be used, they are empty
+ video_ids = self.extract_videos_from_page(channel_page)
+ else:
+ # Download all channel pages using the json-based channel_ajax query
+ for pagenum in itertools.count(1):
+ url = self._MORE_PAGES_URL % (pagenum, channel_id)
+ page = self._download_webpage(url, channel_id,
+ u'Downloading page #%s' % pagenum)
+
+ page = json.loads(page)
+
+ ids_in_page = self.extract_videos_from_page(page['content_html'])
+ video_ids.extend(ids_in_page)
+
+ if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
+ break
self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
projects / youtube-dl / commitdiff