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|mediathek\.php)?\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
  15     _TESTS = [
  16         {
  17             'url': 'http://www.3sat.de/mediathek/index.php?mode=play&obj=45918',
  18             'md5': 'be37228896d30a88f315b638900a026e',
  19             'info_dict': {
  20                 'id': '45918',
  21                 'ext': 'mp4',
  22                 'title': 'Waidmannsheil',
  23                 'description': 'md5:cce00ca1d70e21425e72c86a98a56817',
  24                 'uploader': '3sat',
  25                 'upload_date': '20140913'
  26             }
  27         },
  28         {
  29             'url': 'http://www.3sat.de/mediathek/mediathek.php?mode=play&obj=51066',
  30             'only_matching': True,
  31         },
  32     ]
  33
  34     def _real_extract(self, url):
  35         mobj = re.match(self._VALID_URL, url)
  36         video_id = mobj.group('id')
  37         details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
  38         details_doc = self._download_xml(details_url, video_id, 'Downloading video details')
  39
  40         status_code = details_doc.find('./status/statuscode')
  41         if status_code is not None and status_code.text != 'ok':
  42             code = status_code.text
  43             if code == 'notVisibleAnymore':
  44                 message = 'Video %s is not available' % video_id
  45             else:
  46                 message = '%s returned error: %s' % (self.IE_NAME, code)
  47             raise ExtractorError(message, expected=True)
  48
  49         thumbnail_els = details_doc.findall('.//teaserimage')
  50         thumbnails = [{
  51             'width': int(te.attrib['key'].partition('x')[0]),
  52             'height': int(te.attrib['key'].partition('x')[2]),
  53             'url': te.text,
  54         } for te in thumbnail_els]
  55
  56         information_el = details_doc.find('.//information')
  57         video_title = information_el.find('./title').text
  58         video_description = information_el.find('./detail').text
  59
  60         details_el = details_doc.find('.//details')
  61         video_uploader = details_el.find('./channel').text
  62         upload_date = unified_strdate(details_el.find('./airtime').text)
  63
  64         format_els = details_doc.findall('.//formitaet')
  65         formats = [{
  66             'format_id': fe.attrib['basetype'],
  67             'width': int(fe.find('./width').text),
  68             'height': int(fe.find('./height').text),
  69             'url': fe.find('./url').text,
  70             'filesize': int(fe.find('./filesize').text),
  71             'video_bitrate': int(fe.find('./videoBitrate').text),
  72         } for fe in format_els
  73             if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
  74
  75         self._sort_formats(formats)
  76
  77         return {
  78             '_type': 'video',
  79             'id': video_id,
  80             'title': video_title,
  81             'formats': formats,
  82             'description': video_description,
  83             'thumbnails': thumbnails,
  84             'thumbnail': thumbnails[-1]['url'],
  85             'uploader': video_uploader,
  86             'upload_date': upload_date,
  87         }