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 unified_strdate
   7
   8
   9 class DreiSatIE(InfoExtractor):
  10     IE_NAME = '3sat'
  11     _VALID_URL = r'(?:http://)?(?:www\.)?3sat\.de/mediathek/(?:index\.php)?\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
  12     _TEST = {
  13         'url': 'http://www.3sat.de/mediathek/index.php?obj=36983',
  14         'md5': '9dcfe344732808dbfcc901537973c922',
  15         'info_dict': {
  16             'id': '36983',
  17             'ext': 'mp4',
  18             'title': 'Kaffeeland Schweiz',
  19             'description': 'md5:cc4424b18b75ae9948b13929a0814033',
  20             'uploader': '3sat',
  21             'upload_date': '20130622'
  22         }
  23     }
  24
  25     def _real_extract(self, url):
  26         mobj = re.match(self._VALID_URL, url)
  27         video_id = mobj.group('id')
  28         details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
  29         details_doc = self._download_xml(details_url, video_id, 'Downloading video details')
  30
  31         thumbnail_els = details_doc.findall('.//teaserimage')
  32         thumbnails = [{
  33             'width': int(te.attrib['key'].partition('x')[0]),
  34             'height': int(te.attrib['key'].partition('x')[2]),
  35             'url': te.text,
  36         } for te in thumbnail_els]
  37
  38         information_el = details_doc.find('.//information')
  39         video_title = information_el.find('./title').text
  40         video_description = information_el.find('./detail').text
  41
  42         details_el = details_doc.find('.//details')
  43         video_uploader = details_el.find('./channel').text
  44         upload_date = unified_strdate(details_el.find('./airtime').text)
  45
  46         format_els = details_doc.findall('.//formitaet')
  47         formats = [{
  48             'format_id': fe.attrib['basetype'],
  49             'width': int(fe.find('./width').text),
  50             'height': int(fe.find('./height').text),
  51             'url': fe.find('./url').text,
  52             'filesize': int(fe.find('./filesize').text),
  53             'video_bitrate': int(fe.find('./videoBitrate').text),
  54         } for fe in format_els
  55             if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
  56
  57         self._sort_formats(formats)
  58
  59         return {
  60             '_type': 'video',
  61             'id': video_id,
  62             'title': video_title,
  63             'formats': formats,
  64             'description': video_description,
  65             'thumbnails': thumbnails,
  66             'thumbnail': thumbnails[-1]['url'],
  67             'uploader': video_uploader,
  68             'upload_date': upload_date,
  69         }