_ Git - youtube-dl/blob - youtube_dl/extractor/dreisat.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     unified_strdate,
   9 )
  10
  11
  12 class DreiSatIE(InfoExtractor):
  13     IE_NAME = '3sat'
  14     _VALID_URL = r'(?:http://)?(?:www\.)?3sat\.de/mediathek/(?:index\.php)?\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
  15     _TEST = {
  16         'url': 'http://www.3sat.de/mediathek/index.php?mode=play&obj=45918',
  17         'md5': 'be37228896d30a88f315b638900a026e',
  18         'info_dict': {
  19             'id': '45918',
  20             'ext': 'mp4',
  21             'title': 'Waidmannsheil',
  22             'description': 'md5:cce00ca1d70e21425e72c86a98a56817',
  23             'uploader': '3sat',
  24             'upload_date': '20140913'
  25         }
  26     }
  27
  28     def _real_extract(self, url):
  29         mobj = re.match(self._VALID_URL, url)
  30         video_id = mobj.group('id')
  31         details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
  32         details_doc = self._download_xml(details_url, video_id, 'Downloading video details')
  33
  34         status_code = details_doc.find('./status/statuscode')
  35         if status_code is not None and status_code.text != 'ok':
  36             code = status_code.text
  37             if code == 'notVisibleAnymore':
  38                 message = 'Video %s is not available' % video_id
  39             else:
  40                 message = '%s returned error: %s' % (self.IE_NAME, code)
  41             raise ExtractorError(message, expected=True)
  42
  43         thumbnail_els = details_doc.findall('.//teaserimage')
  44         thumbnails = [{
  45             'width': int(te.attrib['key'].partition('x')[0]),
  46             'height': int(te.attrib['key'].partition('x')[2]),
  47             'url': te.text,
  48         } for te in thumbnail_els]
  49
  50         information_el = details_doc.find('.//information')
  51         video_title = information_el.find('./title').text
  52         video_description = information_el.find('./detail').text
  53
  54         details_el = details_doc.find('.//details')
  55         video_uploader = details_el.find('./channel').text
  56         upload_date = unified_strdate(details_el.find('./airtime').text)
  57
  58         format_els = details_doc.findall('.//formitaet')
  59         formats = [{
  60             'format_id': fe.attrib['basetype'],
  61             'width': int(fe.find('./width').text),
  62             'height': int(fe.find('./height').text),
  63             'url': fe.find('./url').text,
  64             'filesize': int(fe.find('./filesize').text),
  65             'video_bitrate': int(fe.find('./videoBitrate').text),
  66         } for fe in format_els
  67             if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
  68
  69         self._sort_formats(formats)
  70
  71         return {
  72             '_type': 'video',
  73             'id': video_id,
  74             'title': video_title,
  75             'formats': formats,
  76             'description': video_description,
  77             'thumbnails': thumbnails,
  78             'thumbnail': thumbnails[-1]['url'],
  79             'uploader': video_uploader,
  80             'upload_date': upload_date,
  81         }