2 from __future__
import unicode_literals
6 from .common
import InfoExtractor
11 compat_urllib_parse_urlparse
,
19 class ARDMediathekIE(InfoExtractor
):
20 IE_NAME
= 'ARD:mediathek'
21 _VALID_URL
= r
'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
24 'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
25 'file': '22429276.mp4',
26 'md5': '469751912f1de0816a9fc9df8336476c',
28 'title': 'Vertrauen ist gut, Spionieren ist besser - Geht so deutsch-amerikanische Freundschaft?',
29 'description': 'Das Erste Mediathek [ARD]: Vertrauen ist gut, Spionieren ist besser - Geht so deutsch-amerikanische Freundschaft?, Anne Will, Über die Spionage-Affäre diskutieren Clemens Binninger, Katrin Göring-Eckardt, Georg Mascolo, Andrew B. Denison und Constanze Kurz.. Das Video zur Sendung Anne Will am Mittwoch, 16.07.2014',
31 'skip': 'Blocked outside of Germany',
33 'url': 'http://www.ardmediathek.de/tv/Tatort/Das-Wunder-von-Wolbeck-Video-tgl-ab-20/Das-Erste/Video?documentId=22490580&bcastId=602916',
37 'title': 'Das Wunder von Wolbeck (Video tgl. ab 20 Uhr)',
38 'description': 'Auf einem restaurierten Hof bei Wolbeck wird der Heilpraktiker Raffael Lembeck eines morgens von seiner Frau Stella tot aufgefunden. Das Opfer war offensichtlich in seiner Praxis zu Fall gekommen und ist dann verblutet, erklärt Prof. Boerne am Tatort.',
40 'skip': 'Blocked outside of Germany',
43 def _real_extract(self
, url
):
44 # determine video id from url
45 m
= re
.match(self
._VALID
_URL
, url
)
47 numid
= re
.search(r
'documentId=([0-9]+)', url
)
49 video_id
= numid
.group(1)
51 video_id
= m
.group('video_id')
53 urlp
= compat_urllib_parse_urlparse(url
)
54 url
= urlp
._replace
(path
=compat_urllib_parse
.quote(urlp
.path
.encode('utf-8'))).geturl()
56 webpage
= self
._download
_webpage
(url
, video_id
)
58 if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage
:
59 raise ExtractorError('Video %s is no longer available' % video_id
, expected
=True)
61 title
= self
._html
_search
_regex
(
62 [r
'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
63 r
'<meta name="dcterms.title" content="(.*?)"/>',
64 r
'<h4 class="headline">(.*?)</h4>'],
66 description
= self
._html
_search
_meta
(
67 'dcterms.abstract', webpage
, 'description', default
=None)
68 if description
is None:
69 description
= self
._html
_search
_meta
(
70 'description', webpage
, 'meta description')
72 # Thumbnail is sometimes not present.
73 # It is in the mobile version, but that seems to use a different URL
74 # structure altogether.
75 thumbnail
= self
._og
_search
_thumbnail
(webpage
, default
=None)
77 media_streams
= re
.findall(r
'''(?x)
78 mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
79 "([^"]+)"''', webpage
)
82 QUALITIES
= qualities(['lo', 'hi', 'hq'])
84 for furl
in set(media_streams
):
85 if furl
.endswith('.f4m'):
88 fid_m
= re
.match(r
'.*\.([^.]+)\.[^.]+$', furl
)
89 fid
= fid_m
.group(1) if fid_m
else None
91 'quality': QUALITIES(fid
),
95 else: # request JSON file
96 media_info
= self
._download
_json
(
97 'http://www.ardmediathek.de/play/media/%s' % video_id
, video_id
)
98 # The second element of the _mediaArray contains the standard http urls
99 streams
= media_info
['_mediaArray'][1]['_mediaStreamArray']
101 if '"fsk"' in webpage
:
102 raise ExtractorError('This video is only available after 20:00')
106 if type(s
['_stream']) == list:
107 for index
, url
in enumerate(s
['_stream'][::-1]):
108 quality
= s
['_quality'] + index
112 'format_id': '%s-%s' % (determine_ext(url
), quality
)
117 'quality': s
['_quality'],
121 format
['format_id'] = '%s-%s' % (
122 determine_ext(format
['url']), format
['quality'])
124 formats
.append(format
)
126 self
._sort
_formats
(formats
)
131 'description': description
,
133 'thumbnail': thumbnail
,
137 class ARDIE(InfoExtractor
):
138 _VALID_URL
= '(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
140 'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
141 'md5': 'd216c3a86493f9322545e045ddc3eb35',
143 'display_id': 'die-story-im-ersten-mission-unter-falscher-flagge',
147 'title': 'Die Story im Ersten: Mission unter falscher Flagge',
148 'upload_date': '20140804',
149 'thumbnail': 're:^https?://.*\.jpg$',
153 def _real_extract(self
, url
):
154 mobj
= re
.match(self
._VALID
_URL
, url
)
155 display_id
= mobj
.group('display_id')
157 player_url
= mobj
.group('mainurl') + '~playerXml.xml'
158 doc
= self
._download
_xml
(player_url
, display_id
)
159 video_node
= doc
.find('./video')
160 upload_date
= unified_strdate(video_node
.find('./broadcastDate').text
)
161 thumbnail
= video_node
.find('.//teaserImage//variant/url').text
164 for a
in video_node
.findall('.//asset'):
166 'format_id': a
.attrib
['type'],
167 'width': int_or_none(a
.find('./frameWidth').text
),
168 'height': int_or_none(a
.find('./frameHeight').text
),
169 'vbr': int_or_none(a
.find('./bitrateVideo').text
),
170 'abr': int_or_none(a
.find('./bitrateAudio').text
),
171 'vcodec': a
.find('./codecVideo').text
,
172 'tbr': int_or_none(a
.find('./totalBitrate').text
),
174 if a
.find('./serverPrefix').text
:
175 f
['url'] = a
.find('./serverPrefix').text
176 f
['playpath'] = a
.find('./fileName').text
178 f
['url'] = a
.find('./fileName').text
180 self
._sort
_formats
(formats
)
183 'id': mobj
.group('id'),
185 'display_id': display_id
,
186 'title': video_node
.find('./title').text
,
187 'duration': parse_duration(video_node
.find('./duration').text
),
188 'upload_date': upload_date
,
189 'thumbnail': thumbnail
,