from __future__ import unicode_literals
import re
-import xml.etree.ElementTree
from .common import InfoExtractor
from ..utils import (
int_or_none,
parse_duration,
parse_iso8601,
+ remove_end,
unescapeHTML,
)
-from ..compat import compat_HTTPError
+from ..compat import (
+ compat_etree_fromstring,
+ compat_HTTPError,
+)
class BBCCoUkIE(InfoExtractor):
IE_NAME = 'bbc.co.uk'
IE_DESC = 'BBC iPlayer'
- _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
+ _ID_REGEX = r'[pb][\da-z]{7}'
+ _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:programmes/(?!articles/)|iplayer(?:/[^/]+)?/(?:episode/|playlist/))|music/clips[/#])(?P<id>%s)' % _ID_REGEX
_MEDIASELECTOR_URLS = [
# Provides HQ HLS streams with even better quality that pc mediaset but fails
# with geolocation in some cases when it's even not geo restricted at all (e.g.
- # http://www.bbc.co.uk/programmes/b06bp7lf)
+ # http://www.bbc.co.uk/programmes/b06bp7lf). Also may fail with selectionunavailable.
'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
]
return self._download_media_selector_url(
mediaselector_url % programme_id, programme_id)
except BBCCoUkIE.MediaSelectionError as e:
- if e.id in ('notukerror', 'geolocation'):
+ if e.id in ('notukerror', 'geolocation', 'selectionunavailable'):
last_exception = e
continue
self._raise_extractor_error(e)
media_selection = self._download_xml(
url, programme_id, 'Downloading media selection XML')
except ExtractorError as ee:
- if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
- media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8'))
+ if isinstance(ee.cause, compat_HTTPError) and ee.cause.code in (403, 404):
+ media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8'))
else:
raise
return self._process_media_selector(media_selection, programme_id)
continue
title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
- description = description_el.text if description_el else None
+ description = description_el.text if description_el is not None else None
def get_programme_id(item):
def get_from_attributes(item):
if not programme_id:
programme_id = self._search_regex(
- r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
+ r'"vpid"\s*:\s*"(%s)"' % self._ID_REGEX, webpage, 'vpid', fatal=False, default=None)
if programme_id:
formats, subtitles = self._download_media_selector(programme_id)
_VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
_MEDIASELECTOR_URLS = [
+ # Provides HQ HLS streams but fails with geolocation in some cases when it's
+ # even not geo restricted at all
+ 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
# Provides more formats, namely direct mp4 links, but fails on some videos with
# notukerror for non UK (?) users (e.g.
# http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
]
_TESTS = [{
- # article with multiple videos embedded with data-media-meta containing
- # playlist.sxml, externalId and no direct video links
+ # article with multiple videos embedded with data-playable containing vpids
'url': 'http://www.bbc.com/news/world-europe-32668511',
'info_dict': {
'id': 'world-europe-32668511',
},
'playlist_count': 2,
}, {
- # article with multiple videos embedded with data-media-meta (more videos)
+ # article with multiple videos embedded with data-playable (more videos)
'url': 'http://www.bbc.com/news/business-28299555',
'info_dict': {
'id': 'business-28299555',
'skip': 'Save time',
}, {
# article with multiple videos embedded with `new SMP()`
+ # broken
'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
'info_dict': {
'id': '3662a707-0af9-3149-963f-47bea720b460',
},
'playlist_count': 18,
}, {
- # single video embedded with mediaAssetPage.init()
+ # single video embedded with data-playable containing vpid
'url': 'http://www.bbc.com/news/world-europe-32041533',
'info_dict': {
'id': 'p02mprgb',
'ext': 'mp4',
'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
+ 'description': 'md5:2868290467291b37feda7863f7a83f54',
'duration': 47,
'timestamp': 1427219242,
'upload_date': '20150324',
'skip_download': True,
}
}, {
- # article with single video embedded with data-media-meta containing
- # direct video links (for now these are extracted) and playlist.xml (with
- # media items as f4m and m3u8 - currently unsupported)
+ # article with single video embedded with data-playable containing XML playlist
+ # with direct video links as progressiveDownloadUrl (for now these are extracted)
+ # and playlist with f4m and m3u8 as streamingUrl
'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
'info_dict': {
'id': '150615_telabyad_kentin_cogu',
'ext': 'mp4',
'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
- 'duration': 47,
'timestamp': 1434397334,
'upload_date': '20150615',
},
'skip_download': True,
}
}, {
- # single video embedded with mediaAssetPage.init() (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',
'info_dict': {
'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
'ext': 'mp4',
'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
- 'duration': 87,
'timestamp': 1434713142,
'upload_date': '20150619',
},
'ext': 'mp4',
'title': 'Hyundai Santa Fe Sport: Rock star',
'description': 'md5:b042a26142c4154a6e472933cf20793d',
- 'timestamp': 1368473503,
- 'upload_date': '20130513',
+ 'timestamp': 1415867444,
+ 'upload_date': '20141113',
},
'params': {
# rtmp download
'skip_download': True,
}
}, {
- # single video with playlist.sxml URL
+ # single video with playlist.sxml URL in playlist param
'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': 'md5:398fca0e2e701c609d726e034fa1fc89',
+ 'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
'duration': 140,
},
'params': {
'skip_download': True,
}
}, {
- # article with multiple videos embedded with playlist.sxml
+ # article with multiple videos embedded with playlist.sxml in playlist param
'url': 'http://www.bbc.com/sport/0/football/34475836',
'info_dict': {
'id': '34475836',
@classmethod
def suitable(cls, url):
- return False if BBCCoUkIE.suitable(url) else super(BBCIE, cls).suitable(url)
+ return False if BBCCoUkIE.suitable(url) or BBCCoUkArticleIE.suitable(url) else super(BBCIE, cls).suitable(url)
def _extract_from_media_meta(self, media_meta, video_id):
# Direct links to media in media metadata (e.g.
webpage = self._download_webpage(url, playlist_id)
- timestamp = parse_iso8601(self._search_regex(
- [r'"datePublished":\s*"([^"]+)',
- r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
- r'itemprop="datePublished"[^>]+datetime="([^"]+)"'],
- webpage, 'date', default=None))
+ timestamp = None
+ playlist_title = None
+ playlist_description = None
+
+ ld = self._parse_json(
+ self._search_regex(
+ r'(?s)<script type="application/ld\+json">(.+?)</script>',
+ webpage, 'ld json', default='{}'),
+ playlist_id, fatal=False)
+ if ld:
+ timestamp = parse_iso8601(ld.get('datePublished'))
+ playlist_title = ld.get('headline')
+ playlist_description = ld.get('articleBody')
+
+ if not timestamp:
+ timestamp = parse_iso8601(self._search_regex(
+ [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
+ r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
+ r'"datePublished":\s*"([^"]+)'],
+ webpage, 'date', default=None))
+
+ entries = []
# article with multiple videos embedded with playlist.sxml (e.g.
# http://www.bbc.com/sport/0/football/34475836)
entries = [
self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
for playlist_url in playlists]
- playlist_title = self._og_search_title(webpage)
- playlist_description = self._og_search_description(webpage, default=None)
- return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
- # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
- programme_id = self._search_regex(
- [r'data-video-player-vpid="([\da-z]{8})"',
- r'<param[^>]+name="externalIdentifier"[^>]+value="([\da-z]{8})"'],
- webpage, 'vpid', default=None)
-
- duration = None
- if not programme_id:
- # single video in news article embedded with data-playable (e.g.
- # http://www.bbc.com/news/world-us-canada-34473351)
- data_playable = self._parse_json(
- unescapeHTML(self._search_regex(
- r'data-playable="({.+?})"', webpage, 'data playable', default='{}')),
- programme_id, fatal=False)
- if data_playable:
- # data-playable has video vpid in settings.playlistObject.items (e.g.
- # http://www.bbc.com/news/world-us-canada-34473351)
+ # news article with multiple videos embedded with data-playable
+ data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
+ if data_playables:
+ for _, data_playable_json in data_playables:
+ data_playable = self._parse_json(
+ unescapeHTML(data_playable_json), playlist_id, fatal=False)
+ if not data_playable:
+ continue
settings = data_playable.get('settings', {})
if settings:
+ # data-playable with video vpid in settings.playlistObject.items (e.g.
+ # http://www.bbc.com/news/world-us-canada-34473351)
playlist_object = settings.get('playlistObject', {})
if playlist_object:
items = playlist_object.get('items')
if items and isinstance(items, list):
+ title = playlist_object['title']
+ description = playlist_object.get('summary')
duration = int_or_none(items[0].get('duration'))
programme_id = items[0].get('vpid')
- if not programme_id:
- # data-playable has no vpid but has a playlist.sxml URLs
- # in otherSettings.playlist (e.g.
- # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
- playlist = data_playable.get('otherSettings', {}).get('playlist', {})
- if playlist:
- return self._extract_from_playlist_sxml(
- playlist.get('progressiveDownloadUrl'), playlist_id, timestamp)
+ formats, subtitles = self._download_media_selector(programme_id)
+ self._sort_formats(formats)
+ entries.append({
+ 'id': programme_id,
+ 'title': title,
+ 'description': description,
+ 'timestamp': timestamp,
+ 'duration': duration,
+ 'formats': formats,
+ 'subtitles': subtitles,
+ })
+ else:
+ # data-playable without vpid but with a playlist.sxml URLs
+ # in otherSettings.playlist (e.g.
+ # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
+ playlist = data_playable.get('otherSettings', {}).get('playlist', {})
+ if playlist:
+ entries.append(self._extract_from_playlist_sxml(
+ playlist.get('progressiveDownloadUrl'), playlist_id, timestamp))
+
+ if entries:
+ playlist_title = playlist_title or remove_end(self._og_search_title(webpage), ' - BBC News')
+ playlist_description = playlist_description or self._og_search_description(webpage, default=None)
+ return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
+
+ # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
+ programme_id = self._search_regex(
+ [r'data-video-player-vpid="(%s)"' % self._ID_REGEX,
+ r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
+ r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
+ webpage, 'vpid', default=None)
if programme_id:
formats, subtitles = self._download_media_selector(programme_id)
'title': title,
'description': description,
'timestamp': timestamp,
- 'duration': duration,
'formats': formats,
'subtitles': subtitles,
}
# 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/(?:[^/]+/)+[\da-z]{8}(?:\b[^"]+)?'
+ EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX
entries = []
for match in extract_all(r'new\s+SMP\(({.+?})\)'):
embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
})
return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
+
+
+class BBCCoUkArticleIE(InfoExtractor):
+ _VALID_URL = 'http://www.bbc.co.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
+ IE_NAME = 'bbc.co.uk:article'
+ IE_DESC = 'BBC articles'
+
+ _TEST = {
+ 'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer',
+ 'info_dict': {
+ 'id': '3jNQLTMrPlYGTBn0WV6M2MS',
+ 'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four',
+ 'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.',
+ },
+ 'playlist_count': 4,
+ 'add_ie': ['BBCCoUk'],
+ }
+
+ def _real_extract(self, url):
+ playlist_id = self._match_id(url)
+
+ webpage = self._download_webpage(url, playlist_id)
+
+ title = self._og_search_title(webpage)
+ description = self._og_search_description(webpage).strip()
+
+ entries = [self.url_result(programme_url) for programme_url in re.findall(
+ r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
+
+ return self.playlist_result(entries, playlist_id, title, description)
projects / youtube-dl / blobdiff