Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/dreisat.py

   1 # coding: utf-8
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     unified_strdate,
   8 )
   9
  10
  11 class DreiSatIE(InfoExtractor):
  12     IE_NAME = '3sat'
  13     _VALID_URL = r'(?:http://)?(?:www\.)?3sat\.de/mediathek/(?:index\.php)?\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
  14     _TEST = {
  15         u"url": u"http://www.3sat.de/mediathek/index.php?obj=36983",
  16         u'file': u'36983.mp4',
  17         u'md5': u'9dcfe344732808dbfcc901537973c922',
  18         u'info_dict': {
  19             u"title": u"Kaffeeland Schweiz",
  20             u"description": u"Über 80 Kaffeeröstereien liefern in der Schweiz das Getränk, in das das Land so vernarrt ist: Mehr als 1000 Tassen trinkt ein Schweizer pro Jahr. SCHWEIZWEIT nimmt die Kaffeekultur unter die...",
  21             u"uploader": u"3sat",
  22             u"upload_date": u"20130622"
  23         }
  24     }
  25
  26
  27     def _real_extract(self, url):
  28         mobj = re.match(self._VALID_URL, url)
  29         video_id = mobj.group('id')
  30         details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
  31         details_doc = self._download_xml(details_url, video_id, note=u'Downloading video details')
  32
  33         thumbnail_els = details_doc.findall('.//teaserimage')
  34         thumbnails = [{
  35             'width': te.attrib['key'].partition('x')[0],
  36             'height': te.attrib['key'].partition('x')[2],
  37             'url': te.text,
  38         } for te in thumbnail_els]
  39
  40         information_el = details_doc.find('.//information')
  41         video_title = information_el.find('./title').text
  42         video_description = information_el.find('./detail').text
  43
  44         details_el = details_doc.find('.//details')
  45         video_uploader = details_el.find('./channel').text
  46         upload_date = unified_strdate(details_el.find('./airtime').text)
  47
  48         format_els = details_doc.findall('.//formitaet')
  49         formats = [{
  50             'format_id': fe.attrib['basetype'],
  51             'width': int(fe.find('./width').text),
  52             'height': int(fe.find('./height').text),
  53             'url': fe.find('./url').text,
  54             'filesize': int(fe.find('./filesize').text),
  55             'video_bitrate': int(fe.find('./videoBitrate').text),
  56         } for fe in format_els
  57             if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
  58
  59         self._sort_formats(formats)
  60
  61         return {
  62             '_type': 'video',
  63             'id': video_id,
  64             'title': video_title,
  65             'formats': formats,
  66             'description': video_description,
  67             'thumbnails': thumbnails,
  68             'thumbnail': thumbnails[-1]['url'],
  69             'uploader': video_uploader,
  70             'upload_date': upload_date,
  71         }