[yt-dlp.git] / youtube_dlc / extractor / wdr.py

# coding: utf-8
from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..compat import (
    compat_str,
    compat_urlparse,
)
from ..utils import (
    determine_ext,
    ExtractorError,
    js_to_json,
    strip_jsonp,
    try_get,
    unified_strdate,
    update_url_query,
    urlhandle_detect_ext,
)


class WDRIE(InfoExtractor):
    _VALID_URL = r'https?://deviceids-medp\.wdr\.de/ondemand/\d+/(?P<id>\d+)\.js'
    _GEO_COUNTRIES = ['DE']
    _TEST = {
        'url': 'http://deviceids-medp.wdr.de/ondemand/155/1557833.js',
        'info_dict': {
            'id': 'mdb-1557833',
            'ext': 'mp4',
            'title': 'Biathlon-Staffel verpasst Podest bei Olympia-Generalprobe',
            'upload_date': '20180112',
        },
    }

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

        metadata = self._download_json(
            url, video_id, transform_source=strip_jsonp)

        is_live = metadata.get('mediaType') == 'live'

        tracker_data = metadata['trackerData']
        media_resource = metadata['mediaResource']

        formats = []
        subtitles = {}

        # check if the metadata contains a direct URL to a file
        for kind, media_resource in media_resource.items():
            if kind == 'captionsHash':
                for ext, url in media_resource.items():
                    subtitles.setdefault('de', []).append({
                        'url': url,
                        'ext': ext,
                    })
                continue

            if kind not in ('dflt', 'alt'):
                continue

            for tag_name, medium_url in media_resource.items():
                if tag_name not in ('videoURL', 'audioURL'):
                    continue

                ext = determine_ext(medium_url)
                if ext == 'm3u8':
                    formats.extend(self._extract_m3u8_formats(
                        medium_url, video_id, 'mp4', 'm3u8_native',
                        m3u8_id='hls'))
                elif ext == 'f4m':
                    manifest_url = update_url_query(
                        medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
                    formats.extend(self._extract_f4m_formats(
                        manifest_url, video_id, f4m_id='hds', fatal=False))
                elif ext == 'smil':
                    formats.extend(self._extract_smil_formats(
                        medium_url, 'stream', fatal=False))
                else:
                    a_format = {
                        'url': medium_url
                    }
                    if ext == 'unknown_video':
                        urlh = self._request_webpage(
                            medium_url, video_id, note='Determining extension')
                        ext = urlhandle_detect_ext(urlh)
                        a_format['ext'] = ext
                    formats.append(a_format)

        self._sort_formats(formats)

        title = tracker_data['trackerClipTitle']

        return {
            'id': tracker_data.get('trackerClipId', video_id),
            'title': self._live_title(title) if is_live else title,
            'alt_title': tracker_data.get('trackerClipSubcategory'),
            'formats': formats,
            'subtitles': subtitles,
            'upload_date': unified_strdate(tracker_data.get('trackerClipAirTime')),
            'is_live': is_live,
        }


class WDRPageIE(InfoExtractor):
    _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
    _PAGE_REGEX = r'/(?:mediathek/)?(?:[^/]+/)*(?P<display_id>[^/]+)\.html'
    _VALID_URL = r'https?://(?:www\d?\.)?(?:wdr\d?|sportschau)\.de' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL

    _TESTS = [
        {
            'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
            # HDS download, MD5 is unstable
            'info_dict': {
                'id': 'mdb-1058683',
                'ext': 'flv',
                'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
                'title': 'Geheimnis Aachener Dom',
                'alt_title': 'Doku am Freitag',
                'upload_date': '20160304',
                'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
                'is_live': False,
                'subtitles': {'de': [{
                    'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
                    'ext': 'ttml',
                }]},
            },
            'skip': 'HTTP Error 404: Not Found',
        },
        {
            'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
            'md5': 'f4c1f96d01cf285240f53ea4309663d8',
            'info_dict': {
                'id': 'mdb-1072000',
                'ext': 'mp3',
                'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
                'title': 'Schriftstellerin Juli Zeh',
                'alt_title': 'WDR 3 Gespräch am Samstag',
                'upload_date': '20160312',
                'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
                'is_live': False,
                'subtitles': {}
            },
            'skip': 'HTTP Error 404: Not Found',
        },
        {
            'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
            'info_dict': {
                'id': 'mdb-1406149',
                'ext': 'mp4',
                'title': r're:^WDR Fernsehen im Livestream \(nur in Deutschland erreichbar\) [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
                'alt_title': 'WDR Fernsehen Live',
                'upload_date': '20150101',
                'is_live': True,
            },
            'params': {
                'skip_download': True,  # m3u8 download
            },
        },
        {
            'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
            'playlist_mincount': 7,
            'info_dict': {
                'id': 'aktuelle-stunde-120',
            },
        },
        {
            'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
            'info_dict': {
                'id': 'mdb-1552552',
                'ext': 'mp4',
                'upload_date': 're:^[0-9]{8}$',
                'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
            },
            'skip': 'The id changes from week to week because of the new episode'
        },
        {
            'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
            'md5': '803138901f6368ee497b4d195bb164f2',
            'info_dict': {
                'id': 'mdb-186083',
                'ext': 'mp4',
                'upload_date': '20130919',
                'title': 'Sachgeschichte - Achterbahn ',
            },
        },
        {
            'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
            # Live stream, MD5 unstable
            'info_dict': {
                'id': 'mdb-869971',
                'ext': 'mp4',
                'title': r're:^COSMO Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
                'upload_date': '20160101',
            },
            'params': {
                'skip_download': True,  # m3u8 download
            }
        },
        {
            'url': 'http://www.sportschau.de/handballem2018/handball-nationalmannschaft-em-stolperstein-vorrunde-100.html',
            'info_dict': {
                'id': 'mdb-1556012',
                'ext': 'mp4',
                'title': 'DHB-Vizepräsident Bob Hanning - "Die Weltspitze ist extrem breit"',
                'upload_date': '20180111',
            },
            'params': {
                'skip_download': True,
            },
        },
        {
            'url': 'http://www.sportschau.de/handballem2018/audio-vorschau---die-handball-em-startet-mit-grossem-favoritenfeld-100.html',
            'only_matching': True,
        }
    ]

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        display_id = mobj.group('display_id')
        webpage = self._download_webpage(url, display_id)

        entries = []

        # Article with several videos

        # for wdr.de the data-extension is in a tag with the class "mediaLink"
        # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
        # for wdrmaus, in a tag with the class "videoButton" (previously a link
        # to the page in a multiline "videoLink"-tag)
        for mobj in re.finditer(
            r'''(?sx)class=
                    (?:
                        (["\'])(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b.*?\1[^>]+|
                        (["\'])videoLink\b.*?\2[\s]*>\n[^\n]*
                    )data-extension=(["\'])(?P<data>(?:(?!\3).)+)\3
                    ''', webpage):
            media_link_obj = self._parse_json(
                mobj.group('data'), display_id, transform_source=js_to_json,
                fatal=False)
            if not media_link_obj:
                continue
            jsonp_url = try_get(
                media_link_obj, lambda x: x['mediaObj']['url'], compat_str)
            if jsonp_url:
                entries.append(self.url_result(jsonp_url, ie=WDRIE.ie_key()))

        # Playlist (e.g. https://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html)
        if not entries:
            entries = [
                self.url_result(
                    compat_urlparse.urljoin(url, mobj.group('href')),
                    ie=WDRPageIE.ie_key())
                for mobj in re.finditer(
                    r'<a[^>]+\bhref=(["\'])(?P<href>(?:(?!\1).)+)\1[^>]+\bdata-extension=',
                    webpage) if re.match(self._PAGE_REGEX, mobj.group('href'))
            ]

        return self.playlist_result(entries, playlist_id=display_id)


class WDRElefantIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)wdrmaus\.de/elefantenseite/#(?P<id>.+)'
    _TEST = {
        'url': 'http://www.wdrmaus.de/elefantenseite/#folge_ostern_2015',
        'info_dict': {
            'title': 'Folge Oster-Spezial 2015',
            'id': 'mdb-1088195',
            'ext': 'mp4',
            'age_limit': None,
            'upload_date': '20150406'
        },
        'params': {
            'skip_download': True,
        },
    }

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

        # Table of Contents seems to always be at this address, so fetch it directly.
        # The website fetches configurationJS.php5, which links to tableOfContentsJS.php5.
        table_of_contents = self._download_json(
            'https://www.wdrmaus.de/elefantenseite/data/tableOfContentsJS.php5',
            display_id)
        if display_id not in table_of_contents:
            raise ExtractorError(
                'No entry in site\'s table of contents for this URL. '
                'Is the fragment part of the URL (after the #) correct?',
                expected=True)
        xml_metadata_path = table_of_contents[display_id]['xmlPath']
        xml_metadata = self._download_xml(
            'https://www.wdrmaus.de/elefantenseite/' + xml_metadata_path,
            display_id)
        zmdb_url_element = xml_metadata.find('./movie/zmdb_url')
        if zmdb_url_element is None:
            raise ExtractorError(
                '%s is not a video' % display_id, expected=True)
        return self.url_result(zmdb_url_element.text, ie=WDRIE.ie_key())


class WDRMobileIE(InfoExtractor):
    _VALID_URL = r'''(?x)
        https?://mobile-ondemand\.wdr\.de/
        .*?/fsk(?P<age_limit>[0-9]+)
        /[0-9]+/[0-9]+/
        (?P<id>[0-9]+)_(?P<title>[0-9]+)'''
    IE_NAME = 'wdr:mobile'
    _TEST = {
        'url': 'http://mobile-ondemand.wdr.de/CMS2010/mdb/ondemand/weltweit/fsk0/42/421735/421735_4283021.mp4',
        'info_dict': {
            'title': '4283021',
            'id': '421735',
            'ext': 'mp4',
            'age_limit': 0,
        },
        'skip': 'Problems with loading data.'
    }

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        return {
            'id': mobj.group('id'),
            'title': mobj.group('title'),
            'age_limit': int(mobj.group('age_limit')),
            'url': url,
            'http_headers': {
                'User-Agent': 'mobile',
            },
        }
Commit	Line	Data
dcdb292f	1	# coding: utf-8
b461641f S	2	from __future__ import unicode_literals
	3
	4	import re
	5
	6	from .common import InfoExtractor
54e8f62e S	7	from ..compat import (
	8	compat_str,
	9	compat_urlparse,
	10	)
1cc79574	11	from ..utils import (
bec2c14f	12	determine_ext,
1594a493	13	ExtractorError,
33a1ff71	14	js_to_json,
37f97295	15	strip_jsonp,
54e8f62e	16	try_get,
cd7481a3	17	unified_strdate,
6869d634	18	update_url_query,
50918c4e	19	urlhandle_detect_ext,
b461641f S	20	)
	21
	22
54e8f62e S	23	class WDRIE(InfoExtractor):
54e8f62e S	24	_VALID_URL = r'https?://deviceids-medp\.wdr\.de/ondemand/\d+/(?P<id>\d+)\.js'
1915662d	25	_GEO_COUNTRIES = ['DE']
54e8f62e S	26	_TEST = {
	27	'url': 'http://deviceids-medp.wdr.de/ondemand/155/1557833.js',
	28	'info_dict': {
0ce39bc5	29	'id': 'mdb-1557833',
54e8f62e	30	'ext': 'mp4',
0ce39bc5 S	31	'title': 'Biathlon-Staffel verpasst Podest bei Olympia-Generalprobe',
0ce39bc5 S	32	'upload_date': '20180112',
54e8f62e	33	},
54e8f62e	34	}
6d0d4fc2	35
54e8f62e S	36	def _real_extract(self, url):
54e8f62e S	37	video_id = self._match_id(url)
6d0d4fc2 YCH	38
6d0d4fc2 YCH	39	metadata = self._download_json(
54e8f62e	40	url, video_id, transform_source=strip_jsonp)
6d0d4fc2	41
54e8f62e S	42	is_live = metadata.get('mediaType') == 'live'
	43
	44	tracker_data = metadata['trackerData']
	45	media_resource = metadata['mediaResource']
6d0d4fc2 YCH	46
6d0d4fc2 YCH	47	formats = []
6de54568	48	subtitles = {}
6d0d4fc2 YCH	49
6d0d4fc2 YCH	50	# check if the metadata contains a direct URL to a file
54e8f62e	51	for kind, media_resource in media_resource.items():
6de54568	52	if kind == 'captionsHash':
	53	for ext, url in media_resource.items():
	54	subtitles.setdefault('de', []).append({
	55	'url': url,
	56	'ext': ext,
	57	})
	58	continue
	59
6d0d4fc2 YCH	60	if kind not in ('dflt', 'alt'):
	61	continue
	62
	63	for tag_name, medium_url in media_resource.items():
	64	if tag_name not in ('videoURL', 'audioURL'):
	65	continue
	66
	67	ext = determine_ext(medium_url)
	68	if ext == 'm3u8':
	69	formats.extend(self._extract_m3u8_formats(
54e8f62e	70	medium_url, video_id, 'mp4', 'm3u8_native',
6d0d4fc2 YCH	71	m3u8_id='hls'))
	72	elif ext == 'f4m':
	73	manifest_url = update_url_query(
	74	medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
	75	formats.extend(self._extract_f4m_formats(
54e8f62e	76	manifest_url, video_id, f4m_id='hds', fatal=False))
6d0d4fc2 YCH	77	elif ext == 'smil':
	78	formats.extend(self._extract_smil_formats(
	79	medium_url, 'stream', fatal=False))
	80	else:
	81	a_format = {
	82	'url': medium_url
	83	}
	84	if ext == 'unknown_video':
	85	urlh = self._request_webpage(
54e8f62e	86	medium_url, video_id, note='Determining extension')
6d0d4fc2 YCH	87	ext = urlhandle_detect_ext(urlh)
	88	a_format['ext'] = ext
	89	formats.append(a_format)
	90
	91	self._sort_formats(formats)
	92
54e8f62e	93	title = tracker_data['trackerClipTitle']
6d0d4fc2 YCH	94
6d0d4fc2 YCH	95	return {
54e8f62e S	96	'id': tracker_data.get('trackerClipId', video_id),
	97	'title': self._live_title(title) if is_live else title,
	98	'alt_title': tracker_data.get('trackerClipSubcategory'),
6d0d4fc2 YCH	99	'formats': formats,
6d0d4fc2 YCH	100	'subtitles': subtitles,
54e8f62e S	101	'upload_date': unified_strdate(tracker_data.get('trackerClipAirTime')),
54e8f62e S	102	'is_live': is_live,
6d0d4fc2 YCH	103	}
	104
	105
54e8f62e	106	class WDRPageIE(InfoExtractor):
949fc42e	107	_CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
54e8f62e S	108	_PAGE_REGEX = r'/(?:mediathek/)?(?:[^/]+/)*(?P<display_id>[^/]+)\.html'
54e8f62e S	109	_VALID_URL = r'https?://(?:www\d?\.)?(?:wdr\d?\|sportschau)\.de' + _PAGE_REGEX + '\|' + _CURRENT_MAUS_URL
c0837a12	110
b461641f S	111	_TESTS = [
b461641f S	112	{
c0837a12	113	'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
a4a8201c	114	# HDS download, MD5 is unstable
b461641f	115	'info_dict': {
c0837a12	116	'id': 'mdb-1058683',
b461641f	117	'ext': 'flv',
c0837a12 BW	118	'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
	119	'title': 'Geheimnis Aachener Dom',
	120	'alt_title': 'Doku am Freitag',
	121	'upload_date': '20160304',
	122	'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
	123	'is_live': False,
	124	'subtitles': {'de': [{
bc7e7adf YCH	125	'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
bc7e7adf YCH	126	'ext': 'ttml',
c0837a12	127	}]},
b461641f	128	},
54e8f62e	129	'skip': 'HTTP Error 404: Not Found',
b461641f S	130	},
b461641f S	131	{
c0837a12 BW	132	'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
c0837a12 BW	133	'md5': 'f4c1f96d01cf285240f53ea4309663d8',
b461641f	134	'info_dict': {
c0837a12	135	'id': 'mdb-1072000',
b461641f	136	'ext': 'mp3',
c0837a12 BW	137	'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
	138	'title': 'Schriftstellerin Juli Zeh',
	139	'alt_title': 'WDR 3 Gespräch am Samstag',
	140	'upload_date': '20160312',
	141	'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
	142	'is_live': False,
	143	'subtitles': {}
b461641f	144	},
54e8f62e	145	'skip': 'HTTP Error 404: Not Found',
b461641f	146	},
176cf9e0	147	{
c0837a12	148	'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
dd8982f1	149	'info_dict': {
54e8f62e	150	'id': 'mdb-1406149',
bec2c14f	151	'ext': 'mp4',
54e8f62e	152	'title': r're:^WDR Fernsehen im Livestream \(nur in Deutschland erreichbar\) [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
c0837a12	153	'alt_title': 'WDR Fernsehen Live',
54e8f62e	154	'upload_date': '20150101',
c0837a12	155	'is_live': True,
bec2c14f BW	156	},
	157	'params': {
	158	'skip_download': True, # m3u8 download
	159	},
b8988b63 AA	160	},
b8988b63 AA	161	{
c0837a12	162	'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
54e8f62e	163	'playlist_mincount': 7,
b8988b63	164	'info_dict': {
54e8f62e	165	'id': 'aktuelle-stunde-120',
b8988b63	166	},
14f7a2b8 BW	167	},
	168	{
	169	'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
	170	'info_dict': {
54e8f62e	171	'id': 'mdb-1552552',
f802c486	172	'ext': 'mp4',
14f7a2b8 BW	173	'upload_date': 're:^[0-9]{8}$',
14f7a2b8 BW	174	'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
14f7a2b8 BW	175	},
	176	'skip': 'The id changes from week to week because of the new episode'
	177	},
949fc42e	178	{
f802c486	179	'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
6d0d4fc2	180	'md5': '803138901f6368ee497b4d195bb164f2',
949fc42e BW	181	'info_dict': {
949fc42e BW	182	'id': 'mdb-186083',
6d0d4fc2	183	'ext': 'mp4',
949fc42e BW	184	'upload_date': '20130919',
949fc42e BW	185	'title': 'Sachgeschichte - Achterbahn ',
949fc42e	186	},
949fc42e	187	},
50918c4e YCH	188	{
50918c4e YCH	189	'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
6869d634	190	# Live stream, MD5 unstable
50918c4e YCH	191	'info_dict': {
50918c4e YCH	192	'id': 'mdb-869971',
54e8f62e S	193	'ext': 'mp4',
54e8f62e S	194	'title': r're:^COSMO Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
50918c4e YCH	195	'upload_date': '20160101',
50918c4e YCH	196	},
54e8f62e S	197	'params': {
	198	'skip_download': True, # m3u8 download
	199	}
	200	},
	201	{
	202	'url': 'http://www.sportschau.de/handballem2018/handball-nationalmannschaft-em-stolperstein-vorrunde-100.html',
	203	'info_dict': {
	204	'id': 'mdb-1556012',
	205	'ext': 'mp4',
	206	'title': 'DHB-Vizepräsident Bob Hanning - "Die Weltspitze ist extrem breit"',
	207	'upload_date': '20180111',
	208	},
	209	'params': {
	210	'skip_download': True,
	211	},
	212	},
	213	{
	214	'url': 'http://www.sportschau.de/handballem2018/audio-vorschau---die-handball-em-startet-mit-grossem-favoritenfeld-100.html',
	215	'only_matching': True,
50918c4e	216	}
b461641f S	217	]
	218
	219	def _real_extract(self, url):
	220	mobj = re.match(self._VALID_URL, url)
c0837a12 BW	221	display_id = mobj.group('display_id')
c0837a12 BW	222	webpage = self._download_webpage(url, display_id)
b461641f	223
54e8f62e	224	entries = []
b8988b63	225
54e8f62e	226	# Article with several videos
b461641f	227
54e8f62e S	228	# for wdr.de the data-extension is in a tag with the class "mediaLink"
	229	# for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
	230	# for wdrmaus, in a tag with the class "videoButton" (previously a link
	231	# to the page in a multiline "videoLink"-tag)
	232	for mobj in re.finditer(
	233	r'''(?sx)class=
	234	(?:
	235	(["\'])(?:mediaLink\|wdrrPlayerPlayBtn\|videoButton)\b.*?\1[^>]+\|
	236	(["\'])videoLink\b.?\2[\s]>\n[^\n]*
	237	)data-extension=(["\'])(?P<data>(?:(?!\3).)+)\3
	238	''', webpage):
	239	media_link_obj = self._parse_json(
	240	mobj.group('data'), display_id, transform_source=js_to_json,
	241	fatal=False)
	242	if not media_link_obj:
	243	continue
	244	jsonp_url = try_get(
	245	media_link_obj, lambda x: x['mediaObj']['url'], compat_str)
	246	if jsonp_url:
	247	entries.append(self.url_result(jsonp_url, ie=WDRIE.ie_key()))
6d0d4fc2	248
54e8f62e S	249	# Playlist (e.g. https://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html)
	250	if not entries:
	251	entries = [
	252	self.url_result(
	253	compat_urlparse.urljoin(url, mobj.group('href')),
	254	ie=WDRPageIE.ie_key())
	255	for mobj in re.finditer(
	256	r'<a[^>]+\bhref=(["\'])(?P<href>(?:(?!\1).)+)\1[^>]+\bdata-extension=',
	257	webpage) if re.match(self._PAGE_REGEX, mobj.group('href'))
	258	]
cd7481a3	259
54e8f62e	260	return self.playlist_result(entries, playlist_id=display_id)
cd7481a3	261
2d8bb80c	262
54e8f62e S	263	class WDRElefantIE(InfoExtractor):
	264	_VALID_URL = r'https?://(?:www\.)wdrmaus\.de/elefantenseite/#(?P<id>.+)'
	265	_TEST = {
	266	'url': 'http://www.wdrmaus.de/elefantenseite/#folge_ostern_2015',
	267	'info_dict': {
	268	'title': 'Folge Oster-Spezial 2015',
	269	'id': 'mdb-1088195',
	270	'ext': 'mp4',
	271	'age_limit': None,
	272	'upload_date': '20150406'
2d8bb80c	273	},
54e8f62e S	274	'params': {
	275	'skip_download': True,
	276	},
	277	}
2d8bb80c SL	278
2d8bb80c SL	279	def _real_extract(self, url):
54e8f62e	280	display_id = self._match_id(url)
2d8bb80c SL	281
	282	# Table of Contents seems to always be at this address, so fetch it directly.
	283	# The website fetches configurationJS.php5, which links to tableOfContentsJS.php5.
	284	table_of_contents = self._download_json(
54e8f62e S	285	'https://www.wdrmaus.de/elefantenseite/data/tableOfContentsJS.php5',
54e8f62e S	286	display_id)
2d8bb80c SL	287	if display_id not in table_of_contents:
	288	raise ExtractorError(
	289	'No entry in site\'s table of contents for this URL. '
	290	'Is the fragment part of the URL (after the #) correct?',
	291	expected=True)
	292	xml_metadata_path = table_of_contents[display_id]['xmlPath']
	293	xml_metadata = self._download_xml(
54e8f62e S	294	'https://www.wdrmaus.de/elefantenseite/' + xml_metadata_path,
54e8f62e S	295	display_id)
2d8bb80c SL	296	zmdb_url_element = xml_metadata.find('./movie/zmdb_url')
	297	if zmdb_url_element is None:
	298	raise ExtractorError(
54e8f62e S	299	'%s is not a video' % display_id, expected=True)
54e8f62e S	300	return self.url_result(zmdb_url_element.text, ie=WDRIE.ie_key())
2d8bb80c SL	301
2d8bb80c SL	302
e4cbb5f3 PH	303	class WDRMobileIE(InfoExtractor):
	304	_VALID_URL = r'''(?x)
	305	https?://mobile-ondemand\.wdr\.de/
	306	.*?/fsk(?P<age_limit>[0-9]+)
	307	/[0-9]+/[0-9]+/
	308	(?P<id>[0-9]+)_(?P<title>[0-9]+)'''
	309	IE_NAME = 'wdr:mobile'
	310	_TEST = {
	311	'url': 'http://mobile-ondemand.wdr.de/CMS2010/mdb/ondemand/weltweit/fsk0/42/421735/421735_4283021.mp4',
	312	'info_dict': {
	313	'title': '4283021',
	314	'id': '421735',
7807ee66	315	'ext': 'mp4',
e4cbb5f3 PH	316	'age_limit': 0,
e4cbb5f3 PH	317	},
7807ee66	318	'skip': 'Problems with loading data.'
e4cbb5f3 PH	319	}
	320
	321	def _real_extract(self, url):
	322	mobj = re.match(self._VALID_URL, url)
	323	return {
	324	'id': mobj.group('id'),
	325	'title': mobj.group('title'),
	326	'age_limit': int(mobj.group('age_limit')),
	327	'url': url,
e1554a40 JMF	328	'http_headers': {
	329	'User-Agent': 'mobile',
	330	},
e4cbb5f3	331	}