1 from __future__ import unicode_literals
5 from .common import InfoExtractor
17 class DreiSatIE(InfoExtractor):
19 _VALID_URL = r'(?:https?://)?(?:www\.)?3sat\.de/mediathek/(?:index\.php|mediathek\.php)?\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
22 'url': 'http://www.3sat.de/mediathek/index.php?mode=play&obj=45918',
23 'md5': 'be37228896d30a88f315b638900a026e',
27 'title': 'Waidmannsheil',
28 'description': 'md5:cce00ca1d70e21425e72c86a98a56817',
29 'uploader': 'SCHWEIZWEIT',
30 'uploader_id': '100000210',
31 'upload_date': '20140913'
34 'skip_download': True, # m3u8 downloads
38 'url': 'http://www.3sat.de/mediathek/mediathek.php?mode=play&obj=51066',
39 'only_matching': True,
43 def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
45 for param_group in smil.findall(self._xpath_ns('./head/paramGroup', namespace)):
46 group_id = param_group.attrib.get(self._xpath_ns('id', 'http://www.w3.org/XML/1998/namespace'))
48 for param in param_group:
49 params[param.get('name')] = param.get('value')
50 param_groups[group_id] = params
53 for video in smil.findall(self._xpath_ns('.//video', namespace)):
54 src = video.get('src')
57 bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
58 group_id = video.get('paramGroup')
59 param_group = param_groups[group_id]
60 for proto in param_group['protocols'].split(','):
62 'url': '%s://%s' % (proto, param_group['host']),
63 'app': param_group['app'],
66 'format_id': '%s-%d' % (proto, bitrate),
69 self._sort_formats(formats)
72 def extract_from_xml_url(self, video_id, xml_url):
73 doc = self._download_xml(
75 note='Downloading video info',
76 errnote='Failed to download video info')
78 status_code = doc.find('./status/statuscode')
79 if status_code is not None and status_code.text != 'ok':
80 code = status_code.text
81 if code == 'notVisibleAnymore':
82 message = 'Video %s is not available' % video_id
84 message = '%s returned error: %s' % (self.IE_NAME, code)
85 raise ExtractorError(message, expected=True)
87 title = doc.find('.//information/title').text
88 description = xpath_text(doc, './/information/detail', 'description')
89 duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration'))
90 uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader')
91 uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id')
92 upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date'))
94 def xml_to_thumbnails(fnode):
97 thumbnail_url = node.text
101 'url': thumbnail_url,
103 if 'key' in node.attrib:
104 m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key'])
106 thumbnail['width'] = int(m.group(1))
107 thumbnail['height'] = int(m.group(2))
108 thumbnails.append(thumbnail)
111 thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage'))
113 format_nodes = doc.findall('.//formitaeten/formitaet')
114 quality = qualities(['veryhigh', 'high', 'med', 'low'])
116 def get_quality(elem):
117 return quality(xpath_text(elem, 'quality'))
118 format_nodes.sort(key=get_quality)
121 for fnode in format_nodes:
122 video_url = fnode.find('url').text
123 is_available = 'http://www.metafilegenerator' not in video_url
126 format_id = fnode.attrib['basetype']
127 quality = xpath_text(fnode, './quality', 'quality')
128 format_m = re.match(r'''(?x)
129 (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
130 (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
133 ext = determine_ext(video_url, None) or format_m.group('container')
134 if ext not in ('smil', 'f4m', 'm3u8'):
135 format_id = format_id + '-' + quality
136 if format_id in format_ids:
142 formats.extend(self._extract_smil_formats(
143 video_url, video_id, fatal=False))
145 # the certificates are misconfigured (see
146 # https://github.com/rg3/youtube-dl/issues/8665)
147 if video_url.startswith('https://'):
149 formats.extend(self._extract_m3u8_formats(
150 video_url, video_id, 'mp4', m3u8_id=format_id, fatal=False))
152 formats.extend(self._extract_f4m_formats(
153 video_url, video_id, f4m_id=format_id, fatal=False))
155 proto = format_m.group('proto').lower()
157 abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000)
158 vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000)
160 width = int_or_none(xpath_text(fnode, './width', 'width'))
161 height = int_or_none(xpath_text(fnode, './height', 'height'))
163 filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize'))
170 'format_id': format_id,
173 'acodec': format_m.group('acodec'),
174 'vcodec': format_m.group('vcodec'),
179 'filesize': filesize,
180 'format_note': format_note,
182 '_available': is_available,
184 format_ids.append(format_id)
186 self._sort_formats(formats)
191 'description': description,
192 'duration': duration,
193 'thumbnails': thumbnails,
194 'uploader': uploader,
195 'uploader_id': uploader_id,
196 'upload_date': upload_date,
200 def _real_extract(self, url):
201 mobj = re.match(self._VALID_URL, url)
202 video_id = mobj.group('id')
203 details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
204 return self.extract_from_xml_url(video_id, details_url)