-# coding: utf-8
-from __future__ import unicode_literals
-
import functools
import itertools
import json
import re
+import xml.etree.ElementTree
from .common import InfoExtractor
-from ..compat import (
- compat_etree_Element,
- compat_HTTPError,
- compat_str,
- compat_urlparse,
-)
+from ..compat import compat_str, compat_urlparse
+from ..networking.exceptions import HTTPError
from ..utils import (
ExtractorError,
OnDemandPagedList,
float_or_none,
get_element_by_class,
int_or_none,
+ join_nonempty,
js_to_json,
parse_duration,
parse_iso8601,
parse_qs,
strip_or_none,
+ traverse_obj,
try_get,
unescapeHTML,
unified_timestamp,
class BBCCoUkIE(InfoExtractor):
IE_NAME = 'bbc.co.uk'
IE_DESC = 'BBC iPlayer'
- _ID_REGEX = r'(?:[pbm][\da-z]{7}|w[\da-z]{7,14})'
+ _ID_REGEX = r'(?:[pbml][\da-z]{7}|w[\da-z]{7,14})'
_VALID_URL = r'''(?x)
https?://
(?:www\.)?bbc\.co\.uk/
iplayer(?:/[^/]+)?/(?:episode/|playlist/)|
music/(?:clips|audiovideo/popular)[/#]|
radio/player/|
- sounds/play/|
events/[^/]+/play/[^/]+/
)
(?P<id>%s)(?!/(?:episodes|broadcasts|clips))
''' % _ID_REGEX
+ _EMBED_REGEX = [r'setPlaylist\("(?P<url>https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)']
_LOGIN_URL = 'https://account.bbc.com/signin'
_NETRC_MACHINE = 'bbc'
# rtmp download
'skip_download': True,
},
- }, {
- 'url': 'https://www.bbc.co.uk/sounds/play/m0007jzb',
- 'note': 'Audio',
- 'info_dict': {
- 'id': 'm0007jz9',
- 'ext': 'mp4',
- 'title': 'BBC Proms, 2019, Prom 34: West–Eastern Divan Orchestra',
- 'description': "Live BBC Proms. West–Eastern Divan Orchestra with Daniel Barenboim and Martha Argerich.",
- 'duration': 9840,
- },
- 'params': {
- # rtmp download
- 'skip_download': True,
- }
}, {
'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
'only_matching': True,
'only_matching': True,
}]
- def _login(self):
- username, password = self._get_login_info()
- if username is None:
- return
-
+ def _perform_login(self, username, password):
login_page = self._download_webpage(
self._LOGIN_URL, None, 'Downloading signin page')
post_url, None, 'Logging in', data=urlencode_postdata(login_form),
headers={'Referer': self._LOGIN_URL})
- if self._LOGIN_URL in urlh.geturl():
+ if self._LOGIN_URL in urlh.url:
error = clean_html(get_element_by_class('form-message', response))
if error:
raise ExtractorError(
'Unable to login: %s' % error, expected=True)
raise ExtractorError('Unable to log in')
- def _real_initialize(self):
- self._login()
-
class MediaSelectionError(Exception):
def __init__(self, id):
self.id = id
continue
captions = self._download_xml(
cc_url, programme_id, 'Downloading captions', fatal=False)
- if not isinstance(captions, compat_etree_Element):
+ if not isinstance(captions, xml.etree.ElementTree.Element):
continue
subtitles['en'] = [
{
def _download_media_selector(self, programme_id):
last_exception = None
+ formats, subtitles = [], {}
for media_set in self._MEDIA_SETS:
try:
- return self._download_media_selector_url(
+ fmts, subs = self._download_media_selector_url(
self._MEDIA_SELECTOR_URL_TEMPL % (media_set, programme_id), programme_id)
+ formats.extend(fmts)
+ if subs:
+ self._merge_subtitles(subs, target=subtitles)
except BBCCoUkIE.MediaSelectionError as e:
if e.id in ('notukerror', 'geolocation', 'selectionunavailable'):
last_exception = e
continue
self._raise_extractor_error(e)
- self._raise_extractor_error(last_exception)
+ if last_exception:
+ if formats or subtitles:
+ self.report_warning(f'{self.IE_NAME} returned error: {last_exception.id}')
+ else:
+ self._raise_extractor_error(last_exception)
+ return formats, subtitles
def _download_media_selector_url(self, url, programme_id=None):
media_selection = self._download_json(
formats.extend(self._extract_mpd_formats(
href, programme_id, mpd_id=format_id, fatal=False))
elif transfer_format == 'hls':
- formats.extend(self._extract_m3u8_formats(
- href, programme_id, ext='mp4', entry_protocol='m3u8_native',
- m3u8_id=format_id, fatal=False))
+ # TODO: let expected_status be passed into _extract_xxx_formats() instead
+ try:
+ fmts = self._extract_m3u8_formats(
+ href, programme_id, ext='mp4', entry_protocol='m3u8_native',
+ m3u8_id=format_id, fatal=False)
+ except ExtractorError as e:
+ if not (isinstance(e.exc_info[1], HTTPError)
+ and e.exc_info[1].status in (403, 404)):
+ raise
+ fmts = []
+ formats.extend(fmts)
elif transfer_format == 'hds':
formats.extend(self._extract_f4m_formats(
href, programme_id, f4m_id=format_id, fatal=False))
f['language_preference'] = -10
formats += version_formats
for tag, subformats in (version_subtitles or {}).items():
- subtitles.setdefault(tag, [])
- subtitles[tag] += subformats
+ subtitles.setdefault(tag, []).extend(subformats)
return programme_id, title, description, duration, formats, subtitles
except ExtractorError as ee:
- if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
+ if not (isinstance(ee.cause, HTTPError) and ee.cause.status == 404):
raise
# fallback to legacy playlist
else:
programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
- self._sort_formats(formats)
-
return {
'id': programme_id,
'title': title,
}
-class BBCIE(BBCCoUkIE):
+class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
IE_NAME = 'bbc'
IE_DESC = 'BBC'
- _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
+ _VALID_URL = r'''(?x)
+ https?://(?:www\.)?(?:
+ bbc\.(?:com|co\.uk)|
+ bbcnewsd73hkzno2ini43t4gblxvycyac5aw4gnv7t2rccijh7745uqd\.onion|
+ bbcweb3hytmzhn5d532owbu6oqadra5z3ar726vq5kgwwn6aucdccrad\.onion
+ )/(?:[^/]+/)+(?P<id>[^/#?]+)'''
_MEDIA_SETS = [
'pc',
'url': 'http://www.bbc.com/news/world-europe-32668511',
'info_dict': {
'id': 'world-europe-32668511',
- 'title': 'Russia stages massive WW2 parade',
+ 'title': 'Russia stages massive WW2 parade despite Western boycott',
'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
},
'playlist_count': 2,
'info_dict': {
'id': '3662a707-0af9-3149-963f-47bea720b460',
'title': 'BUGGER',
+ 'description': r're:BUGGER The recent revelations by the whistleblower Edward Snowden were fascinating. .{211}\.{3}$',
},
'playlist_count': 18,
}, {
'info_dict': {
'id': 'p02mprgb',
'ext': 'mp4',
- 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
- 'description': 'md5:2868290467291b37feda7863f7a83f54',
+ 'title': 'Germanwings crash site aerial video',
+ 'description': r're:(?s)Aerial video showed the site where the Germanwings flight 4U 9525, .{156} BFM TV\.$',
'duration': 47,
'timestamp': 1427219242,
'upload_date': '20150324',
+ 'thumbnail': 'https://ichef.bbci.co.uk/news/1024/media/images/81879000/jpg/_81879090_81879089.jpg',
},
'params': {
- # rtmp download
'skip_download': True,
}
}, {
},
'params': {
'skip_download': True,
- }
+ },
+ 'skip': 'now SIMORGH_DATA with no video',
}, {
# single video embedded with data-playable containing XML playlists (regional section)
'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
'info_dict': {
- 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
+ 'id': '39275083',
+ 'display_id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
'ext': 'mp4',
'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
- 'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
+ 'description': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
'timestamp': 1434713142,
'upload_date': '20150619',
+ 'thumbnail': 'https://a.files.bbci.co.uk/worldservice/live/assets/images/2015/06/19/150619132146_honduras_hsopitales_militares_640x360_aptn_nocredit.jpg',
},
'params': {
'skip_download': True,
- }
+ },
}, {
# single video from video playlist embedded with vxp-playlist-data JSON
'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
},
'params': {
'skip_download': True,
- }
+ },
+ 'skip': '404 Not Found',
}, {
- # single video story with digitalData
+ # single video story with __PWA_PRELOADED_STATE__
'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
'info_dict': {
'id': 'p02q6gc4',
- 'ext': 'flv',
- 'title': 'Sri Lanka’s spicy secret',
- 'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
- 'timestamp': 1437674293,
- 'upload_date': '20150723',
+ 'ext': 'mp4',
+ 'title': 'Tasting the spice of life in Jaffna',
+ 'description': r're:(?s)BBC Travel Show’s Henry Golding explores the city of Jaffna .{151} aftertaste\.$',
+ 'timestamp': 1646058397,
+ 'upload_date': '20220228',
+ 'duration': 255,
+ 'thumbnail': 'https://ichef.bbci.co.uk/images/ic/1920xn/p02vxvkn.jpg',
},
- 'params': {
- # rtmp download
- 'skip_download': True,
- }
}, {
# single video story without digitalData
'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
'timestamp': 1415867444,
'upload_date': '20141113',
},
- 'params': {
- # rtmp download
- 'skip_download': True,
- }
+ 'skip': 'redirects to TopGear home page',
}, {
# single video embedded with Morph
+ # TODO: replacement test page
'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
'info_dict': {
'id': 'p041vhd0',
'uploader': 'BBC Sport',
'uploader_id': 'bbc_sport',
},
- 'params': {
- # m3u8 download
- 'skip_download': True,
- },
- 'skip': 'Georestricted to UK',
+ 'skip': 'Video no longer in page',
}, {
- # single video with playlist.sxml URL in playlist param
+ # single video in __INITIAL_DATA__
'url': 'http://www.bbc.com/sport/0/football/33653409',
'info_dict': {
'id': 'p02xycnp',
'ext': 'mp4',
- 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
- 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
+ 'title': 'Ronaldo to Man Utd, Arsenal to spend?',
+ 'description': r're:(?s)BBC Sport\'s David Ornstein rounds up the latest transfer reports, .{359} here\.$',
+ 'timestamp': 1437750175,
+ 'upload_date': '20150724',
+ 'thumbnail': r're:https?://.+/.+media/images/69320000/png/_69320754_mmgossipcolumnextraaugust18.png',
'duration': 140,
},
- 'params': {
- # rtmp download
- 'skip_download': True,
- }
}, {
- # article with multiple videos embedded with playlist.sxml in playlist param
+ # article with multiple videos embedded with Morph.setPayload
'url': 'http://www.bbc.com/sport/0/football/34475836',
'info_dict': {
'id': '34475836',
'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
},
'playlist_count': 3,
+ }, {
+ # Testing noplaylist
+ 'url': 'http://www.bbc.com/sport/0/football/34475836',
+ 'info_dict': {
+ 'id': 'p034ppnv',
+ 'ext': 'mp4',
+ 'title': 'All you need to know about Jurgen Klopp',
+ 'timestamp': 1444335081,
+ 'upload_date': '20151008',
+ 'duration': 122.0,
+ 'thumbnail': 'https://ichef.bbci.co.uk/onesport/cps/976/cpsprodpb/7542/production/_85981003_klopp.jpg',
+ },
+ 'params': {
+ 'noplaylist': True,
+ },
}, {
# school report article with single video
'url': 'http://www.bbc.co.uk/schoolreport/35744779',
'title': 'School which breaks down barriers in Jerusalem',
},
'playlist_count': 1,
+ 'skip': 'redirects to Young Reporter home page https://www.bbc.co.uk/news/topics/cg41ylwv43pt',
}, {
# single video with playlist URL from weather section
'url': 'http://www.bbc.com/weather/features/33601775',
'thumbnail': r're:https?://.+/.+\.jpg',
'timestamp': 1437785037,
'upload_date': '20150725',
+ 'duration': 105,
+ },
+ }, {
+ # video with window.__INITIAL_DATA__ and value as JSON string
+ 'url': 'https://www.bbc.com/news/av/world-europe-59468682',
+ 'info_dict': {
+ 'id': 'p0b779gc',
+ 'ext': 'mp4',
+ 'title': 'Why France is making this woman a national hero',
+ 'description': r're:(?s)France is honouring the US-born 20th Century singer and activist Josephine .{208} Second World War.',
+ 'thumbnail': r're:https?://.+/.+\.jpg',
+ 'timestamp': 1638215626,
+ 'upload_date': '20211129',
+ 'duration': 125,
+ },
+ }, {
+ # video with script id __NEXT_DATA__ and value as JSON string
+ 'url': 'https://www.bbc.com/news/uk-68546268',
+ 'info_dict': {
+ 'id': 'p0hj0lq7',
+ 'ext': 'mp4',
+ 'title': 'Nasser Hospital doctor describes his treatment by IDF',
+ 'description': r're:(?s)Doctor Abu Sabha said he was detained by Israeli forces after .{276} hostages\."$',
+ 'thumbnail': r're:https?://.+/.+\.jpg',
+ 'timestamp': 1710188248,
+ 'upload_date': '20240311',
+ 'duration': 104,
},
}, {
# single video article embedded with data-media-vpid
'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
'only_matching': True,
}, {
+ # bbcthreeConfig
'url': 'https://www.bbc.co.uk/bbcthree/clip/73d0bbd0-abc3-4cea-b3c0-cdae21905eb1',
'info_dict': {
'id': 'p06556y7',
'ext': 'mp4',
- 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
- 'description': 'md5:4b7dfd063d5a789a1512e99662be3ddd',
+ 'title': 'Things Not To Say to people that live on council estates',
+ 'description': "From being labelled a 'chav', to the presumption that they're 'scroungers', people who live on council estates encounter all kinds of prejudices and false assumptions about themselves, their families, and their lifestyles. Here, eight people discuss the common statements, misconceptions, and clichés that they're tired of hearing.",
+ 'duration': 360,
+ 'thumbnail': r're:https?://.+/.+\.jpg',
},
- 'params': {
- 'skip_download': True,
- }
}, {
# window.__PRELOADED_STATE__
'url': 'https://www.bbc.co.uk/radio/play/b0b9z4yl',
'uploader': 'Radio 3',
'uploader_id': 'bbc_radio_three',
},
+ 'skip': '404 Not Found',
}, {
'url': 'http://www.bbc.co.uk/learningenglish/chinese/features/lingohack/ep-181227',
'info_dict': {
'ext': 'mp4',
'title': 'md5:2fabf12a726603193a2879a055f72514',
'description': 'Learn English words and phrases from this story',
+ 'thumbnail': 'https://ichef.bbci.co.uk/images/ic/1200x675/p06pq9gk.jpg',
},
'add_ie': [BBCCoUkIE.ie_key()],
}, {
'info_dict': {
'id': 'p07c6sb9',
'ext': 'mp4',
- 'title': 'How positive thinking is harming your happiness',
- 'alt_title': 'The downsides of positive thinking',
- 'description': 'md5:fad74b31da60d83b8265954ee42d85b4',
+ 'title': 'The downsides of positive thinking',
+ 'description': 'The downsides of positive thinking',
'duration': 235,
- 'thumbnail': r're:https?://.+/p07c9dsr.jpg',
- 'upload_date': '20190604',
- 'categories': ['Psychology'],
+ 'thumbnail': r're:https?://.+/p07c9dsr\.(?:jpg|webp|png)',
+ 'upload_date': '20220223',
+ 'timestamp': 1645632746,
},
+ }, {
+ # BBC Sounds
+ 'url': 'https://www.bbc.co.uk/sounds/play/w3ct5rgx',
+ 'info_dict': {
+ 'id': 'p0hrw4nr',
+ 'ext': 'mp4',
+ 'title': 'Are our coastlines being washed away?',
+ 'description': r're:(?s)Around the world, coastlines are constantly changing .{2000,} Images\)$',
+ 'timestamp': 1713556800,
+ 'upload_date': '20240419',
+ 'duration': 1588,
+ 'thumbnail': 'https://ichef.bbci.co.uk/images/ic/raw/p0hrnxbl.jpg',
+ 'uploader': 'World Service',
+ 'uploader_id': 'bbc_world_service',
+ 'series': 'CrowdScience',
+ 'chapters': [],
+ }
+ }, { # onion routes
+ 'url': 'https://www.bbcnewsd73hkzno2ini43t4gblxvycyac5aw4gnv7t2rccijh7745uqd.onion/news/av/world-europe-63208576',
+ 'only_matching': True,
+ }, {
+ 'url': 'https://www.bbcweb3hytmzhn5d532owbu6oqadra5z3ar726vq5kgwwn6aucdccrad.onion/sport/av/football/63195681',
+ 'only_matching': True,
}]
@classmethod
def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
programme_id, title, description, duration, formats, subtitles = \
self._process_legacy_playlist_url(url, playlist_id)
- self._sort_formats(formats)
return {
'id': programme_id,
'title': title,
json_ld_info = self._search_json_ld(webpage, playlist_id, default={})
timestamp = json_ld_info.get('timestamp')
- playlist_title = json_ld_info.get('title')
- if not playlist_title:
- playlist_title = self._og_search_title(
- webpage, default=None) or self._html_search_regex(
- r'<title>(.+?)</title>', webpage, 'playlist title', default=None)
- if playlist_title:
- playlist_title = re.sub(r'(.+)\s*-\s*BBC.*?$', r'\1', playlist_title).strip()
+ playlist_title = json_ld_info.get('title') or re.sub(
+ r'(.+)\s*-\s*BBC.*?$', r'\1', self._generic_title('', webpage, default='')).strip() or None
playlist_description = json_ld_info.get(
'description') or self._og_search_description(webpage, default=None)
duration = int_or_none(items[0].get('duration'))
programme_id = items[0].get('vpid')
formats, subtitles = self._download_media_selector(programme_id)
- self._sort_formats(formats)
entries.append({
'id': programme_id,
'title': title,
# Some playlist URL may fail with 500, at the same time
# the other one may work fine (e.g.
# http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
- if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
+ if isinstance(e.cause, HTTPError) and e.cause.status == 500:
continue
raise
if entry:
- self._sort_formats(entry['formats'])
entries.append(entry)
if entries:
webpage, 'group id', default=None)
if group_id:
return self.url_result(
- 'https://www.bbc.co.uk/programmes/%s' % group_id,
- ie=BBCCoUkIE.ie_key())
+ f'https://www.bbc.co.uk/programmes/{group_id}', BBCCoUkIE)
# single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
programme_id = self._search_regex(
if programme_id:
formats, subtitles = self._download_media_selector(programme_id)
- self._sort_formats(formats)
# digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
digital_data = self._parse_json(
self._search_regex(
if version_id:
title = smp_data['title']
formats, subtitles = self._download_media_selector(version_id)
- self._sort_formats(formats)
image_url = smp_data.get('holdingImageURL')
display_date = init_data.get('displayDate')
topic_title = init_data.get('topicTitle')
}
# Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
- # There are several setPayload calls may be present but the video
- # seems to be always related to the first one
- morph_payload = self._parse_json(
- self._search_regex(
- r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
- webpage, 'morph payload', default='{}'),
- playlist_id, fatal=False)
+ # Several setPayload calls may be present but the video(s)
+ # should be in one that mentions leadMedia or videoData
+ morph_payload = self._search_json(
+ r'\bMorph\s*\.\s*setPayload\s*\([^,]+,', webpage, 'morph payload', playlist_id,
+ contains_pattern=r'{(?s:(?:(?!</script>).)+(?:"leadMedia"|\\"videoData\\")\s*:.+)}',
+ default={})
if morph_payload:
- components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
- for component in components:
- if not isinstance(component, dict):
- continue
- lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
- if not lead_media:
- continue
- identifiers = lead_media.get('identifiers')
- if not identifiers or not isinstance(identifiers, dict):
- continue
- programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
+ for lead_media in traverse_obj(morph_payload, (
+ 'body', 'components', ..., 'props', 'leadMedia', {dict})):
+ programme_id = traverse_obj(lead_media, ('identifiers', ('vpid', 'playablePid'), {str}, any))
if not programme_id:
continue
- title = lead_media.get('title') or self._og_search_title(webpage)
formats, subtitles = self._download_media_selector(programme_id)
- self._sort_formats(formats)
- description = lead_media.get('summary')
- uploader = lead_media.get('masterBrand')
- uploader_id = lead_media.get('mid')
- duration = None
- duration_d = lead_media.get('duration')
- if isinstance(duration_d, dict):
- duration = parse_duration(dict_get(
- duration_d, ('rawDuration', 'formattedDuration', 'spokenDuration')))
return {
'id': programme_id,
- 'title': title,
- 'description': description,
- 'duration': duration,
- 'uploader': uploader,
- 'uploader_id': uploader_id,
+ 'title': lead_media.get('title') or self._og_search_title(webpage),
+ **traverse_obj(lead_media, {
+ 'description': ('summary', {str}),
+ 'duration': ('duration', ('rawDuration', 'formattedDuration', 'spokenDuration'), {parse_duration}),
+ 'uploader': ('masterBrand', {str}),
+ 'uploader_id': ('mid', {str}),
+ }),
'formats': formats,
'subtitles': subtitles,
}
+ body = self._parse_json(traverse_obj(morph_payload, (
+ 'body', 'content', 'article', 'body')), playlist_id, fatal=False)
+ for video_data in traverse_obj(body, (lambda _, v: v['videoData']['pid'], 'videoData')):
+ if video_data.get('vpid'):
+ video_id = video_data['vpid']
+ formats, subtitles = self._download_media_selector(video_id)
+ entry = {
+ 'id': video_id,
+ 'formats': formats,
+ 'subtitles': subtitles,
+ }
+ else:
+ video_id = video_data['pid']
+ entry = self.url_result(
+ f'https://www.bbc.co.uk/programmes/{video_id}', BBCCoUkIE,
+ video_id, url_transparent=True)
+ entry.update({
+ 'timestamp': traverse_obj(morph_payload, (
+ 'body', 'content', 'article', 'dateTimeInfo', 'dateTime', {parse_iso8601})
+ ),
+ **traverse_obj(video_data, {
+ 'thumbnail': (('iChefImage', 'image'), {url_or_none}, any),
+ 'title': (('title', 'caption'), {str}, any),
+ 'duration': ('duration', {parse_duration}),
+ }),
+ })
+ if video_data.get('isLead') and not self._yes_playlist(playlist_id, video_id):
+ return entry
+ entries.append(entry)
+ if entries:
+ playlist_title = traverse_obj(morph_payload, (
+ 'body', 'content', 'article', 'headline', {str})) or playlist_title
+ return self.playlist_result(
+ entries, playlist_id, playlist_title, playlist_description)
- preload_state = self._parse_json(self._search_regex(
- r'window\.__PRELOADED_STATE__\s*=\s*({.+?});', webpage,
- 'preload state', default='{}'), playlist_id, fatal=False)
- if preload_state:
- current_programme = preload_state.get('programmes', {}).get('current') or {}
- programme_id = current_programme.get('id')
- if current_programme and programme_id and current_programme.get('type') == 'playable_item':
- title = current_programme.get('titles', {}).get('tertiary') or playlist_title
- formats, subtitles = self._download_media_selector(programme_id)
- self._sort_formats(formats)
- synopses = current_programme.get('synopses') or {}
- network = current_programme.get('network') or {}
- duration = int_or_none(
- current_programme.get('duration', {}).get('value'))
- thumbnail = None
- image_url = current_programme.get('image_url')
- if image_url:
- thumbnail = image_url.replace('{recipe}', 'raw')
+ # various PRELOADED_STATE JSON
+ preload_state = self._search_json(
+ r'window\.__(?:PWA_)?PRELOADED_STATE__\s*=', webpage,
+ 'preload state', playlist_id, transform_source=js_to_json, default={})
+ # PRELOADED_STATE with current programmme
+ current_programme = traverse_obj(preload_state, ('programmes', 'current', {dict}))
+ programme_id = traverse_obj(current_programme, ('id', {str}))
+ if programme_id and current_programme.get('type') == 'playable_item':
+ title = traverse_obj(current_programme, ('titles', ('tertiary', 'secondary'), {str}, any)) or playlist_title
+ formats, subtitles = self._download_media_selector(programme_id)
+ return {
+ 'id': programme_id,
+ 'title': title,
+ 'formats': formats,
+ **traverse_obj(current_programme, {
+ 'description': ('synopses', ('long', 'medium', 'short'), {str}, any),
+ 'thumbnail': ('image_url', {lambda u: url_or_none(u.replace('{recipe}', 'raw'))}),
+ 'duration': ('duration', 'value', {int_or_none}),
+ 'uploader': ('network', 'short_title', {str}),
+ 'uploader_id': ('network', 'id', {str}),
+ 'timestamp': ((('availability', 'from'), ('release', 'date')), {parse_iso8601}, any),
+ 'series': ('titles', 'primary', {str}),
+ }),
+ 'subtitles': subtitles,
+ 'chapters': traverse_obj(preload_state, (
+ 'tracklist', 'tracks', lambda _, v: float(v['offset']['start']), {
+ 'title': ('titles', {lambda x: join_nonempty(
+ 'primary', 'secondary', 'tertiary', delim=' - ', from_dict=x)}),
+ 'start_time': ('offset', 'start', {float_or_none}),
+ 'end_time': ('offset', 'end', {float_or_none}),
+ })
+ ),
+ }
+
+ # PWA_PRELOADED_STATE with article video asset
+ asset_id = traverse_obj(preload_state, (
+ 'entities', 'articles', lambda k, _: k.rsplit('/', 1)[-1] == playlist_id,
+ 'assetVideo', 0, {str}, any))
+ if asset_id:
+ video_id = traverse_obj(preload_state, ('entities', 'videos', asset_id, 'vpid', {str}))
+ if video_id:
+ article = traverse_obj(preload_state, (
+ 'entities', 'articles', lambda _, v: v['assetVideo'][0] == asset_id, any))
+
+ def image_url(image_id):
+ return traverse_obj(preload_state, (
+ 'entities', 'images', image_id, 'url',
+ {lambda u: url_or_none(u.replace('$recipe', 'raw'))}))
+
+ formats, subtitles = self._download_media_selector(video_id)
return {
- 'id': programme_id,
- 'title': title,
- 'description': dict_get(synopses, ('long', 'medium', 'short')),
- 'thumbnail': thumbnail,
- 'duration': duration,
- 'uploader': network.get('short_title'),
- 'uploader_id': network.get('id'),
+ 'id': video_id,
+ **traverse_obj(preload_state, ('entities', 'videos', asset_id, {
+ 'title': ('title', {str}),
+ 'description': (('synopsisLong', 'synopsisMedium', 'synopsisShort'), {str}, any),
+ 'thumbnail': (0, {image_url}),
+ 'duration': ('duration', {int_or_none}),
+ })),
'formats': formats,
'subtitles': subtitles,
+ 'timestamp': traverse_obj(article, ('displayDate', {parse_iso8601})),
}
+ else:
+ return self.url_result(
+ f'https://www.bbc.co.uk/programmes/{asset_id}', BBCCoUkIE,
+ asset_id, playlist_title, display_id=playlist_id,
+ description=playlist_description)
bbc3_config = self._parse_json(
self._search_regex(
clip_title = clip.get('title')
if clip_vpid and clip_title:
formats, subtitles = self._download_media_selector(clip_vpid)
- self._sort_formats(formats)
return {
'id': clip_vpid,
'title': clip_title,
if not programme_id:
continue
formats, subtitles = self._download_media_selector(programme_id)
- self._sort_formats(formats)
entries.append({
'id': programme_id,
'title': playlist_title,
return self.playlist_result(
entries, playlist_id, playlist_title, playlist_description)
- initial_data = self._parse_json(self._search_regex(
- r'window\.__INITIAL_DATA__\s*=\s*({.+?});', webpage,
- 'preload state', default='{}'), playlist_id, fatal=False)
+ def parse_model(model):
+ """Extract single video from model structure"""
+ item_id = traverse_obj(model, ('versions', 0, 'versionId', {str}))
+ if not item_id:
+ return
+ formats, subtitles = self._download_media_selector(item_id)
+ return {
+ 'id': item_id,
+ 'formats': formats,
+ 'subtitles': subtitles,
+ **traverse_obj(model, {
+ 'title': ('title', {str}),
+ 'thumbnail': ('imageUrl', {lambda u: urljoin(url, u.replace('$recipe', 'raw'))}),
+ 'description': ('synopses', ('long', 'medium', 'short'), {str}, {lambda x: x or None}, any),
+ 'duration': ('versions', 0, 'duration', {int}),
+ 'timestamp': ('versions', 0, 'availableFrom', {functools.partial(int_or_none, scale=1000)}),
+ })
+ }
+
+ def is_type(*types):
+ return lambda _, v: v['type'] in types
+
+ initial_data = self._search_regex(
+ r'window\.__INITIAL_DATA__\s*=\s*("{.+?}")\s*;', webpage,
+ 'quoted preload state', default=None)
+ if initial_data is None:
+ initial_data = self._search_regex(
+ r'window\.__INITIAL_DATA__\s*=\s*({.+?})\s*;', webpage,
+ 'preload state', default='{}')
+ else:
+ initial_data = self._parse_json(initial_data or '"{}"', playlist_id, fatal=False)
+ initial_data = self._parse_json(initial_data, playlist_id, fatal=False)
if initial_data:
+ for video_data in traverse_obj(initial_data, (
+ 'stores', 'article', 'articleBodyContent', is_type('video'))):
+ model = traverse_obj(video_data, (
+ 'model', 'blocks', is_type('aresMedia'),
+ 'model', 'blocks', is_type('aresMediaMetadata'),
+ 'model', {dict}, any))
+ entry = parse_model(model)
+ if entry:
+ entries.append(entry)
+ if entries:
+ return self.playlist_result(
+ entries, playlist_id, playlist_title, playlist_description)
+
def parse_media(media):
if not media:
return
if not (item_id and item_title):
continue
formats, subtitles = self._download_media_selector(item_id)
- self._sort_formats(formats)
item_desc = None
blocks = try_get(media, lambda x: x['summary']['blocks'], list)
if blocks:
'subtitles': subtitles,
'timestamp': item_time,
'description': strip_or_none(item_desc),
+ 'duration': int_or_none(item.get('duration')),
})
- for resp in (initial_data.get('data') or {}).values():
- name = resp.get('name')
+
+ for resp in traverse_obj(initial_data, ('data', lambda _, v: v['name'])):
+ name = resp['name']
if name == 'media-experience':
parse_media(try_get(resp, lambda x: x['data']['initialItem']['mediaItem'], dict))
elif name == 'article':
- for block in (try_get(resp, lambda x: x['data']['blocks'], list) or []):
- if block.get('type') != 'media':
- continue
- parse_media(block.get('model'))
+ for block in traverse_obj(resp, (
+ 'data', (None, ('content', 'model')), 'blocks',
+ is_type('media', 'video'), 'model', {dict})):
+ parse_media(block)
return self.playlist_result(
entries, playlist_id, playlist_title, playlist_description)
+ # extract from SIMORGH_DATA hydration JSON
+ simorgh_data = self._search_json(
+ r'window\s*\.\s*SIMORGH_DATA\s*=', webpage,
+ 'simorgh data', playlist_id, default={})
+ if simorgh_data:
+ done = False
+ for video_data in traverse_obj(simorgh_data, (
+ 'pageData', 'content', 'model', 'blocks', is_type('video', 'legacyMedia'))):
+ model = traverse_obj(video_data, (
+ 'model', 'blocks', is_type('aresMedia'),
+ 'model', 'blocks', is_type('aresMediaMetadata'),
+ 'model', {dict}, any))
+ if video_data['type'] == 'video':
+ entry = parse_model(model)
+ else: # legacyMedia: no duration, subtitles
+ block_id, entry = traverse_obj(model, ('blockId', {str})), None
+ media_data = traverse_obj(simorgh_data, (
+ 'pageData', 'promo', 'media',
+ {lambda x: x if x['id'] == block_id else None}))
+ formats = traverse_obj(media_data, ('playlist', lambda _, v: url_or_none(v['url']), {
+ 'url': ('url', {url_or_none}),
+ 'ext': ('format', {str}),
+ 'tbr': ('bitrate', {functools.partial(int_or_none, scale=1000)}),
+ }))
+ if formats:
+ entry = {
+ 'id': block_id,
+ 'display_id': playlist_id,
+ 'formats': formats,
+ 'description': traverse_obj(simorgh_data, ('pageData', 'promo', 'summary', {str})),
+ **traverse_obj(model, {
+ 'title': ('title', {str}),
+ 'thumbnail': ('imageUrl', {lambda u: urljoin(url, u.replace('$recipe', 'raw'))}),
+ 'description': ('synopses', ('long', 'medium', 'short'), {str}, any),
+ 'timestamp': ('firstPublished', {functools.partial(int_or_none, scale=1000)}),
+ }),
+ }
+ done = True
+ if entry:
+ entries.append(entry)
+ if done:
+ break
+ if entries:
+ return self.playlist_result(
+ entries, playlist_id, playlist_title, playlist_description)
+
def extract_all(pattern):
return list(filter(None, map(
lambda s: self._parse_json(s, playlist_id, fatal=False),
re.findall(pattern, webpage))))
+ # US accessed article with single embedded video (e.g.
+ # https://www.bbc.com/news/uk-68546268)
+ next_data = traverse_obj(self._search_nextjs_data(webpage, playlist_id, default={}),
+ ('props', 'pageProps', 'page'))
+ model = traverse_obj(next_data, (
+ ..., 'contents', is_type('video'),
+ 'model', 'blocks', is_type('media'),
+ 'model', 'blocks', is_type('mediaMetadata'),
+ 'model', {dict}, any))
+ if model and (entry := parse_model(model)):
+ if not entry.get('timestamp'):
+ entry['timestamp'] = traverse_obj(next_data, (
+ ..., 'contents', is_type('timestamp'), 'model',
+ 'timestamp', {functools.partial(int_or_none, scale=1000)}, any))
+ entries.append(entry)
+ return self.playlist_result(
+ entries, playlist_id, playlist_title, playlist_description)
+
# Multiple video article (e.g.
# http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX
formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
if not formats and not self.get_param('ignore_no_formats'):
continue
- self._sort_formats(formats)
video_id = media_meta.get('externalId')
if not video_id:
projects / yt-dlp.git / blobdiff