Mirroring
/
yt-dlp
mirror of https://github.com/yt-dlp/yt-dlp
1
0
Fork 0
Code Issues Releases Wiki Activity

[ie/bbc] Fix and extend extraction (#9705)

Browse Source 
Closes #9701
Authored by: kylegustavo, dirkf, pukkandan
pull/9952/head
kylegustavo 6 months ago committed by GitHub
parent 6d8a53d870
commit 7975ddf245
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 308 additions and 130 deletions
Show Stats Download Patch File Download Diff File

438
yt_dlp/extractor/bbc.py
Unescape Escape View File

@ -602,7 +602,7 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'url': 'http://www.bbc.com/news/world-europe-32668511', 'url': 'http://www.bbc.com/news/world-europe-32668511',
'info_dict': { 'info_dict': {
'id': 'world-europe-32668511', 'id': 'world-europe-32668511',
'title': 'Russia stages massive WW2 parade', 'title': 'Russia stages massive WW2 parade despite Western boycott',
'description': 'md5:00ff61976f6081841f759a08bf78cc9c', 'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
}, },
'playlist_count': 2, 'playlist_count': 2,
@ -623,6 +623,7 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'info_dict': { 'info_dict': {
'id': '3662a707-0af9-3149-963f-47bea720b460', 'id': '3662a707-0af9-3149-963f-47bea720b460',
'title': 'BUGGER', 'title': 'BUGGER',
'description': r're:BUGGER The recent revelations by the whistleblower Edward Snowden were fascinating. .{211}\.{3}$',
}, },
'playlist_count': 18, 'playlist_count': 18,
}, { }, {
@ -631,14 +632,14 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'info_dict': { 'info_dict': {
'id': 'p02mprgb', 'id': 'p02mprgb',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV', 'title': 'Germanwings crash site aerial video',
'description': 'md5:2868290467291b37feda7863f7a83f54', 'description': r're:(?s)Aerial video showed the site where the Germanwings flight 4U 9525, .{156} BFM TV\.$',
'duration': 47, 'duration': 47,
'timestamp': 1427219242, 'timestamp': 1427219242,
'upload_date': '20150324', 'upload_date': '20150324',
'thumbnail': 'https://ichef.bbci.co.uk/news/1024/media/images/81879000/jpg/_81879090_81879089.jpg',
}, },
'params': { 'params': {
# rtmp download
'skip_download': True, 'skip_download': True,
} }
}, { }, {
@ -656,21 +657,24 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
}, },
'params': { 'params': {
'skip_download': True, 'skip_download': True,
} },
'skip': 'now SIMORGH_DATA with no video',
}, { }, {
# single video embedded with data-playable containing XML playlists (regional section) # 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', 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
'info_dict': { 'info_dict': {
'id': '150619_video_honduras_militares_hospitales_corrupcion_aw', 'id': '39275083',
'display_id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción', '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, 'timestamp': 1434713142,
'upload_date': '20150619', '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': { 'params': {
'skip_download': True, 'skip_download': True,
} },
}, { }, {
# single video from video playlist embedded with vxp-playlist-data JSON # single video from video playlist embedded with vxp-playlist-data JSON
'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376', 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
@ -683,22 +687,21 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
}, },
'params': { 'params': {
'skip_download': True, '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', 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
'info_dict': { 'info_dict': {
'id': 'p02q6gc4', 'id': 'p02q6gc4',
'ext': 'flv', 'ext': 'mp4',
'title': 'Sri Lanka’s spicy secret', 'title': 'Tasting the spice of life in Jaffna',
'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.', 'description': r're:(?s)BBC Travel Show’s Henry Golding explores the city of Jaffna .{151} aftertaste\.$',
'timestamp': 1437674293, 'timestamp': 1646058397,
'upload_date': '20150723', '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 # single video story without digitalData
'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star', 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
@ -710,12 +713,10 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'timestamp': 1415867444, 'timestamp': 1415867444,
'upload_date': '20141113', 'upload_date': '20141113',
}, },
'params': { 'skip': 'redirects to TopGear home page',
# rtmp download
'skip_download': True,
}
}, { }, {
# single video embedded with Morph # single video embedded with Morph
# TODO: replacement test page
'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975', 'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
'info_dict': { 'info_dict': {
'id': 'p041vhd0', 'id': 'p041vhd0',
@ -726,27 +727,22 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'uploader': 'BBC Sport', 'uploader': 'BBC Sport',
'uploader_id': 'bbc_sport', 'uploader_id': 'bbc_sport',
}, },
'params': { 'skip': 'Video no longer in page',
# m3u8 download
'skip_download': True,
},
'skip': 'Georestricted to UK',
}, { }, {
# single video with playlist.sxml URL in playlist param # single video in __INITIAL_DATA__
'url': 'http://www.bbc.com/sport/0/football/33653409', 'url': 'http://www.bbc.com/sport/0/football/33653409',
'info_dict': { 'info_dict': {
'id': 'p02xycnp', 'id': 'p02xycnp',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?', 'title': '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.', '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, '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', 'url': 'http://www.bbc.com/sport/0/football/34475836',
'info_dict': { 'info_dict': {
'id': '34475836', 'id': '34475836',
@ -754,6 +750,21 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.', 'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
}, },
'playlist_count': 3, '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 # school report article with single video
'url': 'http://www.bbc.co.uk/schoolreport/35744779', 'url': 'http://www.bbc.co.uk/schoolreport/35744779',
@ -762,6 +773,7 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'title': 'School which breaks down barriers in Jerusalem', 'title': 'School which breaks down barriers in Jerusalem',
}, },
'playlist_count': 1, '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 # single video with playlist URL from weather section
'url': 'http://www.bbc.com/weather/features/33601775', 'url': 'http://www.bbc.com/weather/features/33601775',
@ -778,18 +790,33 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'thumbnail': r're:https?://.+/.+\.jpg', 'thumbnail': r're:https?://.+/.+\.jpg',
'timestamp': 1437785037, 'timestamp': 1437785037,
'upload_date': '20150725', 'upload_date': '20150725',
'duration': 105,
}, },
}, { }, {
# video with window.__INITIAL_DATA__ and value as JSON string # video with window.__INITIAL_DATA__ and value as JSON string
'url': 'https://www.bbc.com/news/av/world-europe-59468682', 'url': 'https://www.bbc.com/news/av/world-europe-59468682',
'info_dict': { 'info_dict': {
'id': 'p0b71qth', 'id': 'p0b779gc',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Why France is making this woman a national hero', 'title': 'Why France is making this woman a national hero',
'description': 'md5:7affdfab80e9c3a1f976230a1ff4d5e4', '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', 'thumbnail': r're:https?://.+/.+\.jpg',
'timestamp': 1638230731, 'timestamp': 1638215626,
'upload_date': '20211130', '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 # single video article embedded with data-media-vpid
@ -817,6 +844,7 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'uploader': 'Radio 3', 'uploader': 'Radio 3',
'uploader_id': 'bbc_radio_three', 'uploader_id': 'bbc_radio_three',
}, },
'skip': '404 Not Found',
}, { }, {
'url': 'http://www.bbc.co.uk/learningenglish/chinese/features/lingohack/ep-181227', 'url': 'http://www.bbc.co.uk/learningenglish/chinese/features/lingohack/ep-181227',
'info_dict': { 'info_dict': {
@ -824,6 +852,7 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'ext': 'mp4', 'ext': 'mp4',
'title': 'md5:2fabf12a726603193a2879a055f72514', 'title': 'md5:2fabf12a726603193a2879a055f72514',
'description': 'Learn English words and phrases from this story', '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()], 'add_ie': [BBCCoUkIE.ie_key()],
}, { }, {
@ -832,28 +861,30 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'info_dict': { 'info_dict': {
'id': 'p07c6sb9', 'id': 'p07c6sb9',
'ext': 'mp4', 'ext': 'mp4',
'title': 'How positive thinking is harming your happiness', 'title': 'The downsides of positive thinking',
'alt_title': 'The downsides of positive thinking', 'description': 'The downsides of positive thinking',
'description': 'md5:fad74b31da60d83b8265954ee42d85b4',
'duration': 235, 'duration': 235,
'thumbnail': r're:https?://.+/p07c9dsr.jpg', 'thumbnail': r're:https?://.+/p07c9dsr\.(?:jpg|webp|png)',
'upload_date': '20190604', 'upload_date': '20220223',
'categories': ['Psychology'], 'timestamp': 1645632746,
}, },
}, { }, {
# BBC Sounds # BBC Sounds
'url': 'https://www.bbc.co.uk/sounds/play/m001q78b', 'url': 'https://www.bbc.co.uk/sounds/play/w3ct5rgx',
'info_dict': { 'info_dict': {
'id': 'm001q789', 'id': 'p0hrw4nr',
'ext': 'mp4', 'ext': 'mp4',
'title': 'The Night Tracks Mix - Music for the darkling hour', 'title': 'Are our coastlines being washed away?',
'thumbnail': 'https://ichef.bbci.co.uk/images/ic/raw/p0c00hym.jpg', 'description': r're:(?s)Around the world, coastlines are constantly changing .{2000,} Images\)$',
'chapters': 'count:8', 'timestamp': 1713556800,
'description': 'md5:815fb51cbdaa270040aab8145b3f1d67', 'upload_date': '20240419',
'uploader': 'Radio 3', 'duration': 1588,
'duration': 1800, 'thumbnail': 'https://ichef.bbci.co.uk/images/ic/raw/p0hrnxbl.jpg',
'uploader_id': 'bbc_radio_three', 'uploader': 'World Service',
}, 'uploader_id': 'bbc_world_service',
'series': 'CrowdScience',
'chapters': [],
}
}, { # onion routes }, { # onion routes
'url': 'https://www.bbcnewsd73hkzno2ini43t4gblxvycyac5aw4gnv7t2rccijh7745uqd.onion/news/av/world-europe-63208576', 'url': 'https://www.bbcnewsd73hkzno2ini43t4gblxvycyac5aw4gnv7t2rccijh7745uqd.onion/news/av/world-europe-63208576',
'only_matching': True, 'only_matching': True,
@ -1008,8 +1039,7 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
webpage, 'group id', default=None) webpage, 'group id', default=None)
if group_id: if group_id:
return self.url_result( return self.url_result(
'https://www.bbc.co.uk/programmes/%s' % group_id, f'https://www.bbc.co.uk/programmes/{group_id}', BBCCoUkIE)
ie=BBCCoUkIE.ie_key())
# single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret) # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
programme_id = self._search_regex( programme_id = self._search_regex(
@ -1069,83 +1099,133 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
} }
# Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975) # 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 # Several setPayload calls may be present but the video(s)
# seems to be always related to the first one # should be in one that mentions leadMedia or videoData
morph_payload = self._parse_json( morph_payload = self._search_json(
self._search_regex( r'\bMorph\s*\.\s*setPayload\s*\([^,]+,', webpage, 'morph payload', playlist_id,
r'Morph\.setPayload\([^,]+,\s*({.+?})\);', contains_pattern=r'{(?s:(?:(?!</script>).)+(?:"leadMedia"|\\"videoData\\")\s*:.+)}',
webpage, 'morph payload', default='{}'), default={})
playlist_id, fatal=False)
if morph_payload: if morph_payload:
components = try_get(morph_payload, lambda x: x['body']['components'], list) or [] for lead_media in traverse_obj(morph_payload, (
for component in components: 'body', 'components', ..., 'props', 'leadMedia', {dict})):
if not isinstance(component, dict): programme_id = traverse_obj(lead_media, ('identifiers', ('vpid', 'playablePid'), {str}, any))
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')
if not programme_id: if not programme_id:
continue continue
title = lead_media.get('title') or self._og_search_title(webpage)
formats, subtitles = self._download_media_selector(programme_id) formats, subtitles = self._download_media_selector(programme_id)
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 { return {
'id': programme_id, 'id': programme_id,
'title': title, 'title': lead_media.get('title') or self._og_search_title(webpage),
'description': description, **traverse_obj(lead_media, {
'duration': duration, 'description': ('summary', {str}),
'uploader': uploader, 'duration': ('duration', ('rawDuration', 'formattedDuration', 'spokenDuration'), {parse_duration}),
'uploader_id': uploader_id, 'uploader': ('masterBrand', {str}),
'uploader_id': ('mid', {str}),
}),
'formats': formats, 'formats': formats,
'subtitles': subtitles, '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( # various PRELOADED_STATE JSON
r'window\.__PRELOADED_STATE__\s*=\s*({.+?});', webpage, preload_state = self._search_json(
'preload state', default='{}'), playlist_id, fatal=False) r'window\.__(?:PWA_)?PRELOADED_STATE__\s*=', webpage,
if preload_state: 'preload state', playlist_id, transform_source=js_to_json, default={})
current_programme = preload_state.get('programmes', {}).get('current') or {} # PRELOADED_STATE with current programmme
programme_id = current_programme.get('id') current_programme = traverse_obj(preload_state, ('programmes', 'current', {dict}))
if current_programme and programme_id and current_programme.get('type') == 'playable_item': programme_id = traverse_obj(current_programme, ('id', {str}))
title = current_programme.get('titles', {}).get('tertiary') or playlist_title if programme_id and current_programme.get('type') == 'playable_item':
formats, subtitles = self._download_media_selector(programme_id) title = traverse_obj(current_programme, ('titles', ('tertiary', 'secondary'), {str}, any)) or playlist_title
synopses = current_programme.get('synopses') or {} formats, subtitles = self._download_media_selector(programme_id)
network = current_programme.get('network') or {} return {
duration = int_or_none( 'id': programme_id,
current_programme.get('duration', {}).get('value')) 'title': title,
thumbnail = None 'formats': formats,
image_url = current_programme.get('image_url') **traverse_obj(current_programme, {
if image_url: 'description': ('synopses', ('long', 'medium', 'short'), {str}, any),
thumbnail = image_url.replace('{recipe}', 'raw') '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 { return {
'id': programme_id, 'id': video_id,
'title': title, **traverse_obj(preload_state, ('entities', 'videos', asset_id, {
'description': dict_get(synopses, ('long', 'medium', 'short')), 'title': ('title', {str}),
'thumbnail': thumbnail, 'description': (('synopsisLong', 'synopsisMedium', 'synopsisShort'), {str}, any),
'duration': duration, 'thumbnail': (0, {image_url}),
'uploader': network.get('short_title'), 'duration': ('duration', {int_or_none}),
'uploader_id': network.get('id'), })),
'formats': formats, 'formats': formats,
'subtitles': subtitles, 'subtitles': subtitles,
'chapters': traverse_obj(preload_state, ( 'timestamp': traverse_obj(article, ('displayDate', {parse_iso8601})),
'tracklist', 'tracks', lambda _, v: float_or_none(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}),
})) or None,
} }
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( bbc3_config = self._parse_json(
self._search_regex( self._search_regex(
@ -1191,6 +1271,28 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
return self.playlist_result( return self.playlist_result(
entries, playlist_id, playlist_title, playlist_description) entries, playlist_id, playlist_title, playlist_description)
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( initial_data = self._search_regex(
r'window\.__INITIAL_DATA__\s*=\s*("{.+?}")\s*;', webpage, r'window\.__INITIAL_DATA__\s*=\s*("{.+?}")\s*;', webpage,
'quoted preload state', default=None) 'quoted preload state', default=None)
@ -1202,6 +1304,19 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
initial_data = self._parse_json(initial_data or '"{}"', playlist_id, fatal=False) initial_data = self._parse_json(initial_data or '"{}"', playlist_id, fatal=False)
initial_data = self._parse_json(initial_data, playlist_id, fatal=False) initial_data = self._parse_json(initial_data, playlist_id, fatal=False)
if initial_data: 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): def parse_media(media):
if not media: if not media:
return return
@ -1234,27 +1349,90 @@ class BBCIE(BBCCoUkIE): # XXX: Do not subclass from concrete IE
'subtitles': subtitles, 'subtitles': subtitles,
'timestamp': item_time, 'timestamp': item_time,
'description': strip_or_none(item_desc), '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': if name == 'media-experience':
parse_media(try_get(resp, lambda x: x['data']['initialItem']['mediaItem'], dict)) parse_media(try_get(resp, lambda x: x['data']['initialItem']['mediaItem'], dict))
elif name == 'article': elif name == 'article':
for block in (try_get(resp, for block in traverse_obj(resp, (
(lambda x: x['data']['blocks'], 'data', (None, ('content', 'model')), 'blocks',
lambda x: x['data']['content']['model']['blocks'],), is_type('media', 'video'), 'model', {dict})):
list) or []): parse_media(block)
if block.get('type') not in ['media', 'video']:
continue
parse_media(block.get('model'))
return self.playlist_result( return self.playlist_result(
entries, playlist_id, playlist_title, playlist_description) 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): def extract_all(pattern):
return list(filter(None, map( return list(filter(None, map(
lambda s: self._parse_json(s, playlist_id, fatal=False), lambda s: self._parse_json(s, playlist_id, fatal=False),
re.findall(pattern, webpage)))) 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. # Multiple video article (e.g.
# http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460) # 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 EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX

Write Preview
Loading…
Cancel
Save