def _extract_video(self, media, programme_id):
formats = []
- vbr = int(media.get('bitrate'))
+ vbr = int_or_none(media.get('bitrate'))
vcodec = media.get('encoding')
service = media.get('service')
- width = int(media.get('width'))
- height = int(media.get('height'))
- file_size = int(media.get('media_file_size'))
+ width = int_or_none(media.get('width'))
+ height = int_or_none(media.get('height'))
+ file_size = int_or_none(media.get('media_file_size'))
for connection in self._extract_connections(media):
conn_formats = self._extract_connection(connection, programme_id)
for format in conn_formats:
def _extract_audio(self, media, programme_id):
formats = []
- abr = int(media.get('bitrate'))
+ abr = int_or_none(media.get('bitrate'))
acodec = media.get('encoding')
service = media.get('service')
for connection in self._extract_connections(media):
if kind != 'programme' and kind != 'radioProgramme':
continue
programme_id = item.get('vpid')
- duration = int(item.get('duration'))
+ duration = int_or_none(item.get('duration'))
formats, subtitles = self._download_media_selector(programme_id)
return programme_id, title, description, duration, formats, subtitles
except ExtractorError as ee:
title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
programme_id = item.get('identifier')
- duration = int(item.get('duration'))
+ duration = int_or_none(item.get('duration'))
formats, subtitles = self._download_media_selector(programme_id)
return programme_id, title, description, duration, formats, subtitles
Potential fields:
* url Mandatory. The URL of the video file
- * ext Will be calculated from url if missing
+ * ext Will be calculated from URL if missing
* format A human-readable description of the format
("mp4 container with h264/opus").
Calculated from the format_id, width, height.
lower to higher preference, each element is a dictionary
with the "ext" entry and one of:
* "data": The subtitles file contents
- * "url": A url pointing to the subtitles file
+ * "url": A URL pointing to the subtitles file
automatic_captions: Like 'subtitles', used by the YoutubeIE for
automatically generated captions
duration: Length of the video in seconds, as an integer.
Set to "root" to indicate that this is a
comment to the original video.
age_limit: Age restriction for the video, as an integer (years)
- webpage_url: The url to the video webpage, if given to youtube-dl it
+ webpage_url: The URL to the video webpage, if given to youtube-dl it
should allow to get the same result again. (It will be set
by YoutubeDL if it's missing)
categories: A list of categories that the video falls in, for example
["Sports", "Berlin"]
is_live: True, False, or None (=unknown). Whether this video is a
live stream that goes on instead of a fixed-length video.
+ start_time: Time in seconds where the reproduction should start, as
+ specified in the URL.
+ end_time: Time in seconds where the reproduction should end, as
+ specified in the URL.
Unless mentioned otherwise, the fields should be Unicode strings.
# Methods for following #608
@staticmethod
def url_result(url, ie=None, video_id=None, video_title=None):
- """Returns a url that points to a page that should be processed"""
+ """Returns a URL that points to a page that should be processed"""
# TODO: ie should be the class used for getting the info
video_info = {'_type': 'url',
'url': url,
return unescapeHTML(escaped)
def _og_search_thumbnail(self, html, **kargs):
- return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs)
+ return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs)
def _og_search_description(self, html, **kargs):
return self._og_search_property('description', html, fatal=False, **kargs)
class SearchInfoExtractor(InfoExtractor):
"""
Base class for paged search queries extractors.
- They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
+ They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query}
Instances should define _SEARCH_KEY and _MAX_RESULTS.
"""
)
from ..utils import (
ExtractorError,
+ determine_ext,
int_or_none,
orderedSet,
+ parse_iso8601,
str_to_int,
unescapeHTML,
)
request.add_header('Cookie', 'family_filter=off; ff=off')
return request
+ def _download_webpage_handle_no_ff(self, url, *args, **kwargs):
+ request = self._build_request(url)
+ return self._download_webpage_handle(request, *args, **kwargs)
+
+ def _download_webpage_no_ff(self, url, *args, **kwargs):
+ request = self._build_request(url)
+ return self._download_webpage(request, *args, **kwargs)
-class DailymotionIE(DailymotionBaseInfoExtractor):
- """Information Extractor for Dailymotion"""
+class DailymotionIE(DailymotionBaseInfoExtractor):
_VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
IE_NAME = 'dailymotion'
'info_dict': {
'id': 'x2iuewm',
'ext': 'mp4',
- 'uploader': 'IGN',
'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
- 'upload_date': '20150306',
+ 'description': 'Several come bundled with the Steam Controller.',
+ 'thumbnail': 're:^https?:.*\.(?:jpg|png)$',
'duration': 74,
+ 'timestamp': 1425657362,
+ 'upload_date': '20150306',
+ 'uploader': 'IGN',
+ 'uploader_id': 'xijv66',
+ 'age_limit': 0,
+ 'view_count': int,
+ 'comment_count': int,
}
},
# Vevo video
def _real_extract(self, url):
video_id = self._match_id(url)
- url = 'https://www.dailymotion.com/video/%s' % video_id
- # Retrieve video webpage to extract further information
- request = self._build_request(url)
- webpage = self._download_webpage(request, video_id)
+ webpage = self._download_webpage_no_ff(
+ 'https://www.dailymotion.com/video/%s' % video_id, video_id)
+
+ age_limit = self._rta_search(webpage)
- # Extract URL, uploader and title from webpage
- self.report_extraction(video_id)
+ description = self._og_search_description(webpage) or self._html_search_meta(
+ 'description', webpage, 'description')
- # It may just embed a vevo video:
- m_vevo = re.search(
+ view_count = str_to_int(self._search_regex(
+ [r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:(\d+)"',
+ r'video_views_count[^>]+>\s+([\d\.,]+)'],
+ webpage, 'view count', fatal=False))
+ comment_count = int_or_none(self._search_regex(
+ r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
+ webpage, 'comment count', fatal=False))
+
+ player_v5 = self._search_regex(
+ r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
+ webpage, 'player v5', default=None)
+ if player_v5:
+ player = self._parse_json(player_v5, video_id)
+ metadata = player['metadata']
+ formats = []
+ for quality, media_list in metadata['qualities'].items():
+ for media in media_list:
+ media_url = media.get('url')
+ if not media_url:
+ continue
+ type_ = media.get('type')
+ if type_ == 'application/vnd.lumberjack.manifest':
+ continue
+ if type_ == 'application/x-mpegURL' or determine_ext(media_url) == 'm3u8':
+ formats.extend(self._extract_m3u8_formats(
+ media_url, video_id, 'mp4', m3u8_id='hls'))
+ else:
+ f = {
+ 'url': media_url,
+ 'format_id': quality,
+ }
+ m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
+ if m:
+ f.update({
+ 'width': int(m.group('width')),
+ 'height': int(m.group('height')),
+ })
+ formats.append(f)
+ self._sort_formats(formats)
+
+ title = metadata['title']
+ duration = int_or_none(metadata.get('duration'))
+ timestamp = int_or_none(metadata.get('created_time'))
+ thumbnail = metadata.get('poster_url')
+ uploader = metadata.get('owner', {}).get('screenname')
+ uploader_id = metadata.get('owner', {}).get('id')
+
+ subtitles = {}
+ for subtitle_lang, subtitle in metadata.get('subtitles', {}).get('data', {}).items():
+ subtitles[subtitle_lang] = [{
+ 'ext': determine_ext(subtitle_url),
+ 'url': subtitle_url,
+ } for subtitle_url in subtitle.get('urls', [])]
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'description': description,
+ 'thumbnail': thumbnail,
+ 'duration': duration,
+ 'timestamp': timestamp,
+ 'uploader': uploader,
+ 'uploader_id': uploader_id,
+ 'age_limit': age_limit,
+ 'view_count': view_count,
+ 'comment_count': comment_count,
+ 'formats': formats,
+ 'subtitles': subtitles,
+ }
+
+ # vevo embed
+ vevo_id = self._search_regex(
r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
- webpage)
- if m_vevo is not None:
- vevo_id = m_vevo.group('id')
- self.to_screen('Vevo video detected: %s' % vevo_id)
- return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
+ webpage, 'vevo embed', default=None)
+ if vevo_id:
+ return self.url_result('vevo:%s' % vevo_id, 'Vevo')
- age_limit = self._rta_search(webpage)
+ # fallback old player
+ embed_page = self._download_webpage_no_ff(
+ 'https://www.dailymotion.com/embed/video/%s' % video_id,
+ video_id, 'Downloading embed page')
+
+ timestamp = parse_iso8601(self._html_search_meta(
+ 'video:release_date', webpage, 'upload date'))
+
+ info = self._parse_json(
+ self._search_regex(
+ r'var info = ({.*?}),$', embed_page,
+ 'video info', flags=re.MULTILINE),
+ video_id)
- video_upload_date = None
- mobj = re.search(r'<meta property="video:release_date" content="([0-9]{4})-([0-9]{2})-([0-9]{2}).+?"/>', webpage)
- if mobj is not None:
- video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
-
- embed_url = 'https://www.dailymotion.com/embed/video/%s' % video_id
- embed_request = self._build_request(embed_url)
- embed_page = self._download_webpage(
- embed_request, video_id, 'Downloading embed page')
- info = self._search_regex(r'var info = ({.*?}),$', embed_page,
- 'video info', flags=re.MULTILINE)
- info = json.loads(info)
if info.get('error') is not None:
msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
raise ExtractorError(msg, expected=True)
'width': width,
'height': height,
})
- if not formats:
- raise ExtractorError('Unable to extract video URL')
+ self._sort_formats(formats)
# subtitles
video_subtitles = self.extract_subtitles(video_id, webpage)
- view_count = str_to_int(self._search_regex(
- r'video_views_count[^>]+>\s+([\d\.,]+)',
- webpage, 'view count', fatal=False))
-
title = self._og_search_title(webpage, default=None)
if title is None:
title = self._html_search_regex(
'id': video_id,
'formats': formats,
'uploader': info['owner.screenname'],
- 'upload_date': video_upload_date,
+ 'timestamp': timestamp,
'title': title,
+ 'description': description,
'subtitles': video_subtitles,
'thumbnail': info['thumbnail_url'],
'age_limit': age_limit,
def _extract_entries(self, id):
video_ids = []
+ processed_urls = set()
for pagenum in itertools.count(1):
- request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum))
- webpage = self._download_webpage(request,
- id, 'Downloading page %s' % pagenum)
+ page_url = self._PAGE_TEMPLATE % (id, pagenum)
+ webpage, urlh = self._download_webpage_handle_no_ff(
+ page_url, id, 'Downloading page %s' % pagenum)
+ if urlh.geturl() in processed_urls:
+ self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
+ page_url, urlh.geturl()), id)
+ break
+
+ processed_urls.add(urlh.geturl())
video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
class DailymotionUserIE(DailymotionPlaylistIE):
IE_NAME = 'dailymotion:user'
- _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?:(?:old/)?user/)?(?P<user>[^/]+)$'
+ _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
_PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
_TESTS = [{
'url': 'https://www.dailymotion.com/user/nqtv',
'title': 'Rémi Gaillard',
},
'playlist_mincount': 100,
+ }, {
+ 'url': 'http://www.dailymotion.com/user/UnderProject',
+ 'info_dict': {
+ 'id': 'UnderProject',
+ 'title': 'UnderProject',
+ },
+ 'playlist_mincount': 1800,
+ 'expected_warnings': [
+ 'Stopped at duplicated page',
+ ],
+ 'skip': 'Takes too long time',
}]
def _real_extract(self, url):
def _real_extract(self, url):
video_id = self._match_id(url)
- request = self._build_request(url)
- webpage = self._download_webpage(request, video_id)
+ webpage = self._download_webpage_no_ff(url, video_id)
title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')