youtube-dl/youtube_dl/extractor/bbc.py

# coding: utf-8
from __future__ import unicode_literals

import re
import itertools

from .common import InfoExtractor
from ..utils import (
    clean_html,
    dict_get,
    ExtractorError,
    float_or_none,
    get_element_by_class,
    int_or_none,
    js_to_json,
    parse_duration,
    parse_iso8601,
    try_get,
    unescapeHTML,
    urlencode_postdata,
    urljoin,
)
from ..compat import (
    compat_HTTPError,
    compat_urlparse,
)


class BBCCoUkIE(InfoExtractor):
    IE_NAME = 'bbc.co.uk'
    IE_DESC = 'BBC iPlayer'
    _ID_REGEX = r'[pbw][\da-z]{7}'
    _VALID_URL = r'''(?x)
                    https?://
                        (?:www\.)?bbc\.co\.uk/
                        (?:
                            programmes/(?!articles/)|
                            iplayer(?:/[^/]+)?/(?:episode/|playlist/)|
                            music/(?:clips|audiovideo/popular)[/#]|
                            radio/player/|
                            events/[^/]+/play/[^/]+/
                        )
                        (?P<id>%s)(?!/(?:episodes|broadcasts|clips))
                    ''' % _ID_REGEX

    _LOGIN_URL = 'https://account.bbc.com/signin'
    _NETRC_MACHINE = 'bbc'

    _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). 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',
    ]

    _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
    _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'

    _NAMESPACES = (
        _MEDIASELECTION_NS,
        _EMP_PLAYLIST_NS,
    )

    _TESTS = [
        {
            'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
            'info_dict': {
                'id': 'b039d07m',
                'ext': 'flv',
                'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
                'description': 'The Canadian poet and songwriter reflects on his musical career.',
            },
            '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,
            },
            'skip': 'Episode is no longer available on BBC iPlayer Radio',
        }, {
            'url': 'http://www.bbc.co.uk/music/clips/p022h44b',
            'note': 'Audio',
            'info_dict': {
                'id': 'p022h44j',
                'ext': 'flv',
                'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
                'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
                'duration': 227,
            },
            '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