[yt-dlp.git] / youtube_dl / extractor / bbc.py

from __future__ import unicode_literals

import xml.etree.ElementTree

from .common import InfoExtractor
from ..utils import (
    ExtractorError,
    parse_duration,
    int_or_none,
)
from ..compat import compat_HTTPError
import re


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})'

    mediaselector_url = 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s'

    _TESTS = [
        {
            'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
            'info_dict': {
                'id': 'b039d07m',
                'ext': 'flv',
                'title': 'Kaleidoscope, Leonard Cohen',
                'description': 'The Canadian poet and songwriter reflects on his musical career.',
                'duration': 1740,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            }
        },
        {
            'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
            'info_dict': {
                'id': 'b00yng1d',
                'ext': 'flv',
                'title': 'The Man in Black: Series 3: The Printed Name',
                'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
                'duration': 1800,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'Episode is no longer available on BBC iPlayer Radio',
        },
        {
            'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
            'info_dict': {
                'id': 'b00yng1d',
                'ext': 'flv',
                'title': 'The Voice UK: Series 3: Blind Auditions 5',
                'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
                'duration': 5100,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
        },
        {
            'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
            'info_dict': {
                'id': 'b03k3pb7',
                'ext': 'flv',
                'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
                'description': '2. Invasion',
                'duration': 3600,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
        }, {
            'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
            'info_dict': {
                'id': 'b04v209v',
                'ext': 'flv',
                'title': 'Pete Tong, The Essential New Tune Special',
                'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
                'duration': 10800,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            }
        }, {
            'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
            'note': 'Audio',
            'info_dict': {
                'id': 'p02frcch',
                'ext': 'flv',
                'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
                'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
                'duration': 3507,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            }
        }, {
            'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
            'note': 'Video',
            'info_dict': {
                'id': 'p025c103',
                'ext': 'flv',
                'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
                'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
                'duration': 226,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            }
        }, {
            'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
            'info_dict': {
                'id': 'p02n76xf',
                'ext': 'flv',
                'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
                'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
                'duration': 3540,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'geolocation',
        }, {
            'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
            'info_dict': {
                'id': 'b05zmgw1',
                'ext': 'flv',
                'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.',
                'title': 'Royal Academy Summer Exhibition',
                'duration': 3540,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'geolocation',
        }, {
            'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
            'only_matching': True,
        }, {
            'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
            'only_matching': True,
        }, {
            'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
            'only_matching': True,
        }
    ]

    def _extract_asx_playlist(self, connection, programme_id):
        asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
        return [ref.get('href') for ref in asx.findall('./Entry/ref')]

    def _extract_connection(self, connection, programme_id):
        formats = []
        protocol = connection.get('protocol')
        supplier = connection.get('supplier')
        if protocol == 'http':
            href = connection.get('href')
            # ASX playlist
            if supplier == 'asx':
                for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
                    formats.append({
                        'url': ref,
                        'format_id': 'ref%s_%s' % (i, supplier),
                    })
            # Direct link
            else:
                formats.append({
                    'url': href,
                    'format_id': supplier,
                })
        elif protocol == 'rtmp':
            application = connection.get('application', 'ondemand')
            auth_string = connection.get('authString')
            identifier = connection.get('identifier')
            server = connection.get('server')
            formats.append({
                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
                'play_path': identifier,
                'app': '%s?%s' % (application, auth_string),
                'page_url': 'http://www.bbc.co.uk',
                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
                'rtmp_live': False,
                'ext': 'flv',
                'format_id': supplier,
            })
        return formats

    def _extract_items(self, playlist):
        return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')

    def _extract_medias(self, media_selection):
        error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
        if error is not None:
            raise ExtractorError(
                '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
        return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')

    def _extract_connections(self, media):
        return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')

    def _extract_video(self, media, programme_id):
        formats = []
        vbr = int_or_none(media.get('bitrate'))
        vcodec = media.get('encoding')
        service = media.get('service')
        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:
                format.update({
                    'format_id': '%s_%s' % (service, format['format_id']),
                    'width': width,
                    'height': height,
                    'vbr': vbr,
                    'vcodec': vcodec,
                    'filesize': file_size,
                })
            formats.extend(conn_formats)
        return formats

    def _extract_audio(self, media, programme_id):
        formats = []
        abr = int_or_none(media.get('bitrate'))
        acodec = media.get('encoding')
        service = media.get('service')
        for connection in self._extract_connections(media):
            conn_formats = self._extract_connection(connection, programme_id)
            for format in conn_formats:
                format.update({
                    'format_id': '%s_%s' % (service, format['format_id']),
                    'abr': abr,
                    'acodec': acodec,
                })
            formats.extend(conn_formats)
        return formats

    def _get_subtitles(self, media, programme_id):
        subtitles = {}
        for connection in self._extract_connections(media):
            captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
            lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
            subtitles[lang] = [
                {
                    'url': connection.get('href'),
                    'ext': 'ttml',
                },
            ]
        return subtitles

    def _download_media_selector(self, programme_id):
        try:
            media_selection = self._download_xml(
                self.mediaselector_url % programme_id,
                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'))
            else:
                raise

        formats = []
        subtitles = None

        for media in self._extract_medias(media_selection):
            kind = media.get('kind')
            if kind == 'audio':
                formats.extend(self._extract_audio(media, programme_id))
            elif kind == 'video':
                formats.extend(self._extract_video(media, programme_id))
            elif kind == 'captions':
                subtitles = self.extract_subtitles(media, programme_id)

        return formats, subtitles

    def _download_playlist(self, playlist_id):
        try:
            playlist = self._download_json(
                'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
                playlist_id, 'Downloading playlist JSON')

            version = playlist.get('defaultAvailableVersion')
            if version:
                smp_config = version['smpConfig']
                title = smp_config['title']
                description = smp_config['summary']
                for item in smp_config['items']:
                    kind = item['kind']
                    if kind != 'programme' and kind != 'radioProgramme':
                        continue
                    programme_id = item.get('vpid')
                    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:
            if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
                raise

        # fallback to legacy playlist
        playlist = self._download_xml(
            'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id,
            playlist_id, 'Downloading legacy playlist XML')

        no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
        if no_items is not None:
            reason = no_items.get('reason')
            if reason == 'preAvailability':
                msg = 'Episode %s is not yet available' % playlist_id
            elif reason == 'postAvailability':
                msg = 'Episode %s is no longer available' % playlist_id
            elif reason == 'noMedia':
                msg = 'Episode %s is not currently available' % playlist_id
            else:
                msg = 'Episode %s is not available: %s' % (playlist_id, reason)
            raise ExtractorError(msg, expected=True)

        for item in self._extract_items(playlist):
            kind = item.get('kind')
            if kind != 'programme' and kind != 'radioProgramme':
                continue
            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_or_none(item.get('duration'))
            formats, subtitles = self._download_media_selector(programme_id)

        return programme_id, title, description, duration, formats, subtitles

    def _real_extract(self, url):
        group_id = self._match_id(url)

        webpage = self._download_webpage(url, group_id, 'Downloading video page')

        programme_id = None

        tviplayer = self._search_regex(
            r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
            webpage, 'player', default=None)

        if tviplayer:
            player = self._parse_json(tviplayer, group_id).get('player', {})
            duration = int_or_none(player.get('duration'))
            programme_id = player.get('vpid')

        if not programme_id:
            programme_id = self._search_regex(
                r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)

        if programme_id:
            formats, subtitles = self._download_media_selector(programme_id)
            title = self._og_search_title(webpage)
            description = self._search_regex(
                r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
                webpage, 'description', fatal=False)
        else:
            programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)

        self._sort_formats(formats)

        return {
            'id': programme_id,
            'title': title,
            'description': description,
            'thumbnail': self._og_search_thumbnail(webpage, default=None),
            'duration': duration,
            'formats': formats,
            'subtitles': subtitles,
        }


class BBCNewsIE(BBCCoUkIE):
    IE_NAME = 'bbc.com'
    IE_DESC = 'BBC news'
    _VALID_URL = r'https?://(?:www\.)?bbc\.com/.+?/(?P<id>[^/]+)$'

    mediaselector_url = 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s'

    _TESTS = [{
        'url': 'http://www.bbc.com/news/world-europe-32668511',
        'info_dict': {
            'id': 'world-europe-32668511',
            'title': 'Russia stages massive WW2 parade despite Western boycott',
        },
        'playlist_count': 2,
    }, {
        'url': 'http://www.bbc.com/news/business-28299555',
        'info_dict': {
            'id': 'business-28299555',
            'title': 'Farnborough Airshow: Video highlights',
        },
        'playlist_count': 9,
    }, {
        'url': 'http://www.bbc.com/news/world-europe-32041533',
        'note': 'Video',
        'info_dict': {
            'id': 'p02mprgb',
            'ext': 'mp4',
            'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
            'description': 'Germanwings plane crash site in aerial video - Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
            'duration': 47,
            'upload_date': '20150324',
            'uploader': 'BBC News',
        },
        'params': {
            'skip_download': True,
        }
    }, {
        'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
        'note': 'Video',
        'info_dict': {
            'id': 'NA',
            'ext': 'mp4',
            'title': 'YPG: Tel Abyad\'\u0131n tamam\u0131 kontrol\xfcm\xfczde',
            'description': 'YPG: Tel Abyad\'\u0131n tamam\u0131 kontrol\xfcm\xfczde',
            'duration': 47,
            'upload_date': '20150615',
            'uploader': 'BBC News',
        },
        'params': {
            'skip_download': True,
        }
    }, {
        'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
        'note': 'Video',
        'info_dict': {
            'id': '39275083',
            'ext': 'mp4',
            'title': 'Honduras militariza sus hospitales por nuevo esc\xe1ndalo de corrupci\xf3n',
            'description': 'Honduras militariza sus hospitales por nuevo esc\xe1ndalo de corrupci\xf3n',
            'duration': 87,
            'upload_date': '20150619',
            'uploader': 'BBC News',
        },
        'params': {
            'skip_download': True,
        }
    }]

    def _real_extract(self, url):
        list_id = self._match_id(url)
        webpage = self._download_webpage(url, list_id)

        list_title = self._html_search_regex(r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'list title')

        pubdate = self._html_search_regex(r'"datePublished":\s*"(\d+-\d+-\d+)', webpage, 'date', default=None)
        if pubdate:
            pubdate = pubdate.replace('-', '')

        ret = []
        jsent = []

        # works with bbc.com/news/something-something-123456 articles
        jsent = map(
            lambda m: self._parse_json(m, list_id),
            re.findall(r"data-media-meta='({[^']+})'", webpage)
        )

        if len(jsent) == 0:
            # http://www.bbc.com/news/video_and_audio/international
            # and single-video articles
            masset = self._html_search_regex(r'mediaAssetPage\.init\(\s*({.+?}), "/', webpage, 'mediaassets', default=None)
            if masset:
                jmasset = self._parse_json(masset, list_id)
                for key, val in jmasset.get('videos', {}).items():
                    for skey, sval in val.items():
                        sval['id'] = skey
                        jsent.append(sval)

        if len(jsent) == 0:
            # stubbornly generic extractor for {json with "image":{allvideoshavethis},etc}
            # in http://www.bbc.com/news/video_and_audio/international
            # prone to breaking if entries have sourceFiles list
            jsent = map(
                lambda m: self._parse_json(m, list_id),
                re.findall(r"({[^{}]+image\":{[^}]+}[^}]+})", webpage)
            )

        if len(jsent) == 0:
            raise ExtractorError('No video found', expected=True)

        for jent in jsent:
            programme_id = jent.get('externalId')
            xml_url = jent.get('href')

            title = jent.get('caption', '')
            if title == '':
                title = list_title

            duration = parse_duration(jent.get('duration'))
            description = list_title
            if jent.get('caption', '') != '':
                description += ' - ' + jent.get('caption')
            thumbnail = None
            if jent.get('image') is not None:
                thumbnail = jent['image'].get('href')

            formats = []
            subtitles = []

            if programme_id:
                formats, subtitles = self._download_media_selector(programme_id)
            elif jent.get('sourceFiles') is not None:
                # mediaselector not used at
                # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu
                for key, val in jent['sourceFiles'].items():
                    formats.append({
                        'ext': val.get('encoding'),
                        'url': val.get('url'),
                        'filesize': int(val.get('filesize')),
                        'format_id': key
                    })
            elif xml_url:
                # Cheap fallback
                # http://playlists.bbc.co.uk/news/(list_id)[ABC..]/playlist.sxml
                xml = self._download_webpage(xml_url, programme_id, 'Downloading playlist.sxml for externalId (fallback)')
                programme_id = self._search_regex(r'<mediator [^>]*identifier="(.+?)"', xml, 'playlist.sxml (externalId fallback)')
                formats, subtitles = self._download_media_selector(programme_id)

            if len(formats) == 0:
                raise ExtractorError('unsupported json media entry.\n    ' + str(jent) + '\n')

            self._sort_formats(formats)

            id = jent.get('id') if programme_id is None else programme_id
            if id is None:
                id = 'NA'

            ret.append({
                'id': id,
                'uploader': 'BBC News',
                'upload_date': pubdate,
                'title': title,
                'description': description,
                'thumbnail': thumbnail,
                'duration': duration,
                'formats': formats,
                'subtitles': subtitles,
            })

        if len(ret) > 0:
            return self.playlist_result(ret, list_id, list_title)
        raise ExtractorError('No video found', expected=True)
Commit	Line	Data
082c6c86 S	1	from __future__ import unicode_literals
082c6c86 S	2
c056efa2	3	import xml.etree.ElementTree
082c6c86	4
f13b1e7d	5	from .common import InfoExtractor
8683b4d8 S	6	from ..utils import (
8683b4d8 S	7	ExtractorError,
10273d6e	8	parse_duration,
8683b4d8 S	9	int_or_none,
8683b4d8 S	10	)
c056efa2	11	from ..compat import compat_HTTPError
10273d6e	12	import re
082c6c86 S	13
082c6c86 S	14
f13b1e7d	15	class BBCCoUkIE(InfoExtractor):
082c6c86	16	IE_NAME = 'bbc.co.uk'
2e3fd9ec	17	IE_DESC = 'BBC iPlayer'
0692ef86	18	_VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes\|iplayer(?:/[^/]+)?/(?:episode\|playlist))/)\|music/clips[/#])(?P<id>[\da-z]{8})'
082c6c86	19
a8b081a0	20	mediaselector_url = 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s'
a8b081a0	21
2e3fd9ec S	22	_TESTS = [
2e3fd9ec S	23	{
f2d0fc68	24	'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
2e3fd9ec	25	'info_dict': {
f2d0fc68	26	'id': 'b039d07m',
2e3fd9ec	27	'ext': 'flv',
c4914185 S	28	'title': 'Kaleidoscope, Leonard Cohen',
c4914185 S	29	'description': 'The Canadian poet and songwriter reflects on his musical career.',
f2d0fc68	30	'duration': 1740,
2e3fd9ec S	31	},
	32	'params': {
	33	# rtmp download
	34	'skip_download': True,
	35	}
082c6c86	36	},
2e3fd9ec S	37	{
	38	'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
	39	'info_dict': {
	40	'id': 'b00yng1d',
	41	'ext': 'flv',
	42	'title': 'The Man in Black: Series 3: The Printed Name',
	43	'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
	44	'duration': 1800,
	45	},
	46	'params': {
	47	# rtmp download
	48	'skip_download': True,
c7f0177f S	49	},
c7f0177f S	50	'skip': 'Episode is no longer available on BBC iPlayer Radio',
2e3fd9ec S	51	},
	52	{
	53	'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
	54	'info_dict': {
	55	'id': 'b00yng1d',
	56	'ext': 'flv',
17968e44 S	57	'title': 'The Voice UK: Series 3: Blind Auditions 5',
	58	'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
	59	'duration': 5100,
2e3fd9ec S	60	},
	61	'params': {
	62	# rtmp download
	63	'skip_download': True,
	64	},
	65	'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
c056efa2 S	66	},
	67	{
	68	'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
	69	'info_dict': {
	70	'id': 'b03k3pb7',
	71	'ext': 'flv',
	72	'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
	73	'description': '2. Invasion',
	74	'duration': 3600,
	75	},
	76	'params': {
	77	# rtmp download
	78	'skip_download': True,
	79	},
	80	'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
ae6986fb S	81	}, {
	82	'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
	83	'info_dict': {
	84	'id': 'b04v209v',
	85	'ext': 'flv',
	86	'title': 'Pete Tong, The Essential New Tune Special',
	87	'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
	88	'duration': 10800,
	89	},
	90	'params': {
	91	# rtmp download
	92	'skip_download': True,
	93	}
c7e67594 S	94	}, {
	95	'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
	96	'note': 'Audio',
	97	'info_dict': {
	98	'id': 'p02frcch',
	99	'ext': 'flv',
	100	'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
	101	'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
	102	'duration': 3507,
	103	},
	104	'params': {
	105	# rtmp download
	106	'skip_download': True,
	107	}
	108	}, {
	109	'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
	110	'note': 'Video',
	111	'info_dict': {
	112	'id': 'p025c103',
	113	'ext': 'flv',
	114	'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
	115	'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
	116	'duration': 226,
	117	},
	118	'params': {
	119	# rtmp download
	120	'skip_download': True,
	121	}
e68ae99a S	122	}, {
	123	'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
	124	'info_dict': {
	125	'id': 'p02n76xf',
	126	'ext': 'flv',
	127	'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
	128	'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
	129	'duration': 3540,
	130	},
	131	'params': {
	132	# rtmp download
	133	'skip_download': True,
	134	},
	135	'skip': 'geolocation',
25fa8d66 YCH	136	}, {
	137	'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
	138	'info_dict': {
	139	'id': 'b05zmgw1',
	140	'ext': 'flv',
	141	'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.',
	142	'title': 'Royal Academy Summer Exhibition',
	143	'duration': 3540,
	144	},
	145	'params': {
	146	# rtmp download
	147	'skip_download': True,
	148	},
	149	'skip': 'geolocation',
31763975 S	150	}, {
	151	'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
	152	'only_matching': True,
c7e67594 S	153	}, {
	154	'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
	155	'only_matching': True,
0692ef86 S	156	}, {
	157	'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
	158	'only_matching': True,
ae6986fb	159	}
2e3fd9ec S	160	]
	161
	162	def _extract_asx_playlist(self, connection, programme_id):
	163	asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
	164	return [ref.get('href') for ref in asx.findall('./Entry/ref')]
	165
	166	def _extract_connection(self, connection, programme_id):
	167	formats = []
	168	protocol = connection.get('protocol')
	169	supplier = connection.get('supplier')
	170	if protocol == 'http':
	171	href = connection.get('href')
	172	# ASX playlist
	173	if supplier == 'asx':
	174	for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
	175	formats.append({
	176	'url': ref,
	177	'format_id': 'ref%s_%s' % (i, supplier),
	178	})
	179	# Direct link
	180	else:
	181	formats.append({
	182	'url': href,
	183	'format_id': supplier,
	184	})
	185	elif protocol == 'rtmp':
	186	application = connection.get('application', 'ondemand')
	187	auth_string = connection.get('authString')
	188	identifier = connection.get('identifier')
	189	server = connection.get('server')
	190	formats.append({
	191	'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
	192	'play_path': identifier,
	193	'app': '%s?%s' % (application, auth_string),
	194	'page_url': 'http://www.bbc.co.uk',
	195	'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
	196	'rtmp_live': False,
	197	'ext': 'flv',
	198	'format_id': supplier,
	199	})
	200	return formats
	201
	202	def _extract_items(self, playlist):
	203	return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
	204
	205	def _extract_medias(self, media_selection):
c056efa2 S	206	error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
	207	if error is not None:
	208	raise ExtractorError(
	209	'%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
2e3fd9ec S	210	return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
	211
	212	def _extract_connections(self, media):
	213	return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
	214
	215	def _extract_video(self, media, programme_id):
	216	formats = []
d97f5cd7	217	vbr = int_or_none(media.get('bitrate'))
2e3fd9ec S	218	vcodec = media.get('encoding')
2e3fd9ec S	219	service = media.get('service')
d97f5cd7 S	220	width = int_or_none(media.get('width'))
	221	height = int_or_none(media.get('height'))
	222	file_size = int_or_none(media.get('media_file_size'))
2e3fd9ec S	223	for connection in self._extract_connections(media):
	224	conn_formats = self._extract_connection(connection, programme_id)
	225	for format in conn_formats:
	226	format.update({
	227	'format_id': '%s_%s' % (service, format['format_id']),
	228	'width': width,
	229	'height': height,
	230	'vbr': vbr,
	231	'vcodec': vcodec,
	232	'filesize': file_size,
	233	})
	234	formats.extend(conn_formats)
	235	return formats
	236
	237	def _extract_audio(self, media, programme_id):
	238	formats = []
d97f5cd7	239	abr = int_or_none(media.get('bitrate'))
2e3fd9ec S	240	acodec = media.get('encoding')
	241	service = media.get('service')
	242	for connection in self._extract_connections(media):
	243	conn_formats = self._extract_connection(connection, programme_id)
	244	for format in conn_formats:
	245	format.update({
	246	'format_id': '%s_%s' % (service, format['format_id']),
	247	'abr': abr,
	248	'acodec': acodec,
	249	})
	250	formats.extend(conn_formats)
	251	return formats
	252
f13b1e7d	253	def _get_subtitles(self, media, programme_id):
2e3fd9ec S	254	subtitles = {}
	255	for connection in self._extract_connections(media):
	256	captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
	257	lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
f13b1e7d JMF	258	subtitles[lang] = [
	259	{
	260	'url': connection.get('href'),
	261	'ext': 'ttml',
	262	},
f13b1e7d	263	]
2e3fd9ec	264	return subtitles
082c6c86	265
c056efa2 S	266	def _download_media_selector(self, programme_id):
	267	try:
	268	media_selection = self._download_xml(
a8b081a0	269	self.mediaselector_url % programme_id,
c056efa2 S	270	programme_id, 'Downloading media selection XML')
	271	except ExtractorError as ee:
	272	if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
974a6146	273	media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8'))
2e3fd9ec	274	else:
c056efa2	275	raise
082c6c86 S	276
082c6c86 S	277	formats = []
2e3fd9ec S	278	subtitles = None
2e3fd9ec S	279
c056efa2 S	280	for media in self._extract_medias(media_selection):
	281	kind = media.get('kind')
	282	if kind == 'audio':
	283	formats.extend(self._extract_audio(media, programme_id))
	284	elif kind == 'video':
	285	formats.extend(self._extract_video(media, programme_id))
	286	elif kind == 'captions':
f13b1e7d	287	subtitles = self.extract_subtitles(media, programme_id)
2e3fd9ec	288
c056efa2	289	return formats, subtitles
2e3fd9ec	290
ae6986fb S	291	def _download_playlist(self, playlist_id):
	292	try:
	293	playlist = self._download_json(
	294	'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
	295	playlist_id, 'Downloading playlist JSON')
	296
	297	version = playlist.get('defaultAvailableVersion')
	298	if version:
	299	smp_config = version['smpConfig']
	300	title = smp_config['title']
	301	description = smp_config['summary']
	302	for item in smp_config['items']:
	303	kind = item['kind']
	304	if kind != 'programme' and kind != 'radioProgramme':
	305	continue
	306	programme_id = item.get('vpid')
d97f5cd7	307	duration = int_or_none(item.get('duration'))
ae6986fb S	308	formats, subtitles = self._download_media_selector(programme_id)
	309	return programme_id, title, description, duration, formats, subtitles
	310	except ExtractorError as ee:
f813928e	311	if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
ae6986fb S	312	raise
	313
	314	# fallback to legacy playlist
	315	playlist = self._download_xml(
931e2d1d PH	316	'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id,
931e2d1d PH	317	playlist_id, 'Downloading legacy playlist XML')
ae6986fb S	318
	319	no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
	320	if no_items is not None:
	321	reason = no_items.get('reason')
	322	if reason == 'preAvailability':
	323	msg = 'Episode %s is not yet available' % playlist_id
	324	elif reason == 'postAvailability':
	325	msg = 'Episode %s is no longer available' % playlist_id
	326	elif reason == 'noMedia':
	327	msg = 'Episode %s is not currently available' % playlist_id
	328	else:
	329	msg = 'Episode %s is not available: %s' % (playlist_id, reason)
	330	raise ExtractorError(msg, expected=True)
	331
	332	for item in self._extract_items(playlist):
	333	kind = item.get('kind')
	334	if kind != 'programme' and kind != 'radioProgramme':
	335	continue
	336	title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
	337	description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
	338	programme_id = item.get('identifier')
d97f5cd7	339	duration = int_or_none(item.get('duration'))
ae6986fb S	340	formats, subtitles = self._download_media_selector(programme_id)
	341
	342	return programme_id, title, description, duration, formats, subtitles
	343
c056efa2 S	344	def _real_extract(self, url):
	345	group_id = self._match_id(url)
	346
	347	webpage = self._download_webpage(url, group_id, 'Downloading video page')
	348
8683b4d8 S	349	programme_id = None
	350
	351	tviplayer = self._search_regex(
	352	r'mediator\.bind\(({.+?})\s,\sdocument\.getElementById',
	353	webpage, 'player', default=None)
	354
	355	if tviplayer:
	356	player = self._parse_json(tviplayer, group_id).get('player', {})
	357	duration = int_or_none(player.get('duration'))
	358	programme_id = player.get('vpid')
	359
	360	if not programme_id:
	361	programme_id = self._search_regex(
	362	r'"vpid"\s:\s"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None)
	363
c056efa2	364	if programme_id:
c056efa2	365	formats, subtitles = self._download_media_selector(programme_id)
8683b4d8 S	366	title = self._og_search_title(webpage)
8683b4d8 S	367	description = self._search_regex(
25fa8d66	368	r'<p class="[^"]medium-description[^"]">([^<]+)</p>',
8683b4d8	369	webpage, 'description', fatal=False)
c056efa2	370	else:
ae6986fb	371	programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
2e3fd9ec	372
082c6c86 S	373	self._sort_formats(formats)
	374
	375	return {
2e3fd9ec	376	'id': programme_id,
082c6c86 S	377	'title': title,
082c6c86 S	378	'description': description,
650cfd0c	379	'thumbnail': self._og_search_thumbnail(webpage, default=None),
082c6c86 S	380	'duration': duration,
082c6c86 S	381	'formats': formats,
2e3fd9ec	382	'subtitles': subtitles,
5f6a1245	383	}
10273d6e	384
	385
	386	class BBCNewsIE(BBCCoUkIE):
	387	IE_NAME = 'bbc.com'
	388	IE_DESC = 'BBC news'
de939d89	389	_VALID_URL = r'https?://(?:www\.)?bbc\.com/.+?/(?P<id>[^/]+)$'
10273d6e	390
	391	mediaselector_url = 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s'
	392
	393	_TESTS = [{
	394	'url': 'http://www.bbc.com/news/world-europe-32668511',
	395	'info_dict': {
	396	'id': 'world-europe-32668511',
	397	'title': 'Russia stages massive WW2 parade despite Western boycott',
	398	},
	399	'playlist_count': 2,
a3bfddfa	400	}, {
10273d6e	401	'url': 'http://www.bbc.com/news/business-28299555',
	402	'info_dict': {
	403	'id': 'business-28299555',
	404	'title': 'Farnborough Airshow: Video highlights',
	405	},
	406	'playlist_count': 9,
a3bfddfa	407	}, {
10273d6e	408	'url': 'http://www.bbc.com/news/world-europe-32041533',
	409	'note': 'Video',
	410	'info_dict': {
	411	'id': 'p02mprgb',
	412	'ext': 'mp4',
	413	'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
	414	'description': 'Germanwings plane crash site in aerial video - Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
	415	'duration': 47,
da92eeae	416	'upload_date': '20150324',
da92eeae	417	'uploader': 'BBC News',
10273d6e	418	},
	419	'params': {
	420	'skip_download': True,
	421	}
a3bfddfa	422	}, {
de939d89	423	'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
	424	'note': 'Video',
	425	'info_dict': {
	426	'id': 'NA',
	427	'ext': 'mp4',
da92eeae	428	'title': 'YPG: Tel Abyad\'\u0131n tamam\u0131 kontrol\xfcm\xfczde',
da92eeae	429	'description': 'YPG: Tel Abyad\'\u0131n tamam\u0131 kontrol\xfcm\xfczde',
de939d89	430	'duration': 47,
da92eeae	431	'upload_date': '20150615',
da92eeae	432	'uploader': 'BBC News',
de939d89	433	},
	434	'params': {
	435	'skip_download': True,
	436	}
a3bfddfa	437	}, {
de939d89	438	'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
	439	'note': 'Video',
	440	'info_dict': {
	441	'id': '39275083',
	442	'ext': 'mp4',
da92eeae	443	'title': 'Honduras militariza sus hospitales por nuevo esc\xe1ndalo de corrupci\xf3n',
da92eeae	444	'description': 'Honduras militariza sus hospitales por nuevo esc\xe1ndalo de corrupci\xf3n',
de939d89	445	'duration': 87,
da92eeae	446	'upload_date': '20150619',
da92eeae	447	'uploader': 'BBC News',
de939d89	448	},
	449	'params': {
	450	'skip_download': True,
	451	}
10273d6e	452	}]
	453
	454	def _real_extract(self, url):
	455	list_id = self._match_id(url)
	456	webpage = self._download_webpage(url, list_id)
	457
de939d89	458	list_title = self._html_search_regex(r'<title>(.?)(?:\s-\s*BBC [^ ]+)?</title>', webpage, 'list title')
10273d6e	459
	460	pubdate = self._html_search_regex(r'"datePublished":\s*"(\d+-\d+-\d+)', webpage, 'date', default=None)
	461	if pubdate:
a3bfddfa	462	pubdate = pubdate.replace('-', '')
10273d6e	463
10273d6e	464	ret = []
de939d89	465	jsent = []
de939d89	466
10273d6e	467	# works with bbc.com/news/something-something-123456 articles
de939d89	468	jsent = map(
a3bfddfa	469	lambda m: self._parse_json(m, list_id),
a3bfddfa	470	re.findall(r"data-media-meta='({[^']+})'", webpage)
de939d89	471	)
	472
	473	if len(jsent) == 0:
a3bfddfa	474	# http://www.bbc.com/news/video_and_audio/international
	475	# and single-video articles
	476	masset = self._html_search_regex(r'mediaAssetPage\.init\(\s*({.+?}), "/', webpage, 'mediaassets', default=None)
	477	if masset:
	478	jmasset = self._parse_json(masset, list_id)
	479	for key, val in jmasset.get('videos', {}).items():
	480	for skey, sval in val.items():
	481	sval['id'] = skey
	482	jsent.append(sval)
de939d89	483
de939d89	484	if len(jsent) == 0:
a3bfddfa	485	# stubbornly generic extractor for {json with "image":{allvideoshavethis},etc}
	486	# in http://www.bbc.com/news/video_and_audio/international
	487	# prone to breaking if entries have sourceFiles list
	488	jsent = map(
	489	lambda m: self._parse_json(m, list_id),
	490	re.findall(r"({[^{}]+image\":{[^}]+}[^}]+})", webpage)
	491	)
10273d6e	492
de939d89	493	if len(jsent) == 0:
a3bfddfa	494	raise ExtractorError('No video found', expected=True)
10273d6e	495
de939d89	496	for jent in jsent:
75ab0ebc	497	programme_id = jent.get('externalId')
2a282a3b	498	xml_url = jent.get('href')
de939d89	499
a3bfddfa	500	title = jent.get('caption', '')
36da4879	501	if title == '':
a3bfddfa	502	title = list_title
10273d6e	503
77c975f5	504	duration = parse_duration(jent.get('duration'))
da92eeae	505	description = list_title
36da4879	506	if jent.get('caption', '') != '':
a3bfddfa	507	description += ' - ' + jent.get('caption')
10273d6e	508	thumbnail = None
a3bfddfa	509	if jent.get('image') is not None:
a3bfddfa	510	thumbnail = jent['image'].get('href')
10273d6e	511
de939d89	512	formats = []
	513	subtitles = []
	514
10273d6e	515	if programme_id:
a3bfddfa	516	formats, subtitles = self._download_media_selector(programme_id)
	517	elif jent.get('sourceFiles') is not None:
	518	# mediaselector not used at
	519	# http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu
	520	for key, val in jent['sourceFiles'].items():
	521	formats.append({
	522	'ext': val.get('encoding'),
	523	'url': val.get('url'),
	524	'filesize': int(val.get('filesize')),
	525	'format_id': key
	526	})
10273d6e	527	elif xml_url:
a3bfddfa	528	# Cheap fallback
	529	# http://playlists.bbc.co.uk/news/(list_id)[ABC..]/playlist.sxml
	530	xml = self._download_webpage(xml_url, programme_id, 'Downloading playlist.sxml for externalId (fallback)')
	531	programme_id = self._search_regex(r'<mediator [^>]*identifier="(.+?)"', xml, 'playlist.sxml (externalId fallback)')
	532	formats, subtitles = self._download_media_selector(programme_id)
de939d89	533
de939d89	534	if len(formats) == 0:
a3bfddfa	535	raise ExtractorError('unsupported json media entry.\n ' + str(jent) + '\n')
a3bfddfa	536
10273d6e	537	self._sort_formats(formats)
10273d6e	538
a3bfddfa	539	id = jent.get('id') if programme_id is None else programme_id
	540	if id is None:
	541	id = 'NA'
da92eeae	542
a3bfddfa	543	ret.append({
da92eeae	544	'id': id,
10273d6e	545	'uploader': 'BBC News',
	546	'upload_date': pubdate,
	547	'title': title,
	548	'description': description,
	549	'thumbnail': thumbnail,
	550	'duration': duration,
	551	'formats': formats,
	552	'subtitles': subtitles,
a3bfddfa	553	})
10273d6e	554
10273d6e	555	if len(ret) > 0:
a3bfddfa	556	return self.playlist_result(ret, list_id, list_title)
10273d6e	557	raise ExtractorError('No video found', expected=True)