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     determine_ext,
   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         u"url": u"http://www.3sat.de/mediathek/index.php?obj=36983",
  17         u'file': u'36983.webm',
  18         u'md5': u'57c97d0469d71cf874f6815aa2b7c944',
  19         u'info_dict': {
  20             u"title": u"Kaffeeland Schweiz",
  21             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...",
  22             u"uploader": u"3sat",
  23             u"upload_date": u"20130622"
  24         }
  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, note=u'Downloading video details')
  33
  34         thumbnail_els = details_doc.findall('.//teaserimage')
  35         thumbnails = [{
  36             'width': te.attrib['key'].partition('x')[0],
  37             'height': te.attrib['key'].partition('x')[2],
  38             'url': te.text,
  39         } for te in thumbnail_els]
  40
  41         information_el = details_doc.find('.//information')
  42         video_title = information_el.find('./title').text
  43         video_description = information_el.find('./detail').text
  44
  45         details_el = details_doc.find('.//details')
  46         video_uploader = details_el.find('./channel').text
  47         upload_date = unified_strdate(details_el.find('./airtime').text)
  48
  49         format_els = details_doc.findall('.//formitaet')
  50         formats = [{
  51             'format_id': fe.attrib['basetype'],
  52             'width': int(fe.find('./width').text),
  53             'height': int(fe.find('./height').text),
  54             'url': fe.find('./url').text,
  55             'ext': determine_ext(fe.find('./url').text),
  56             'filesize': int(fe.find('./filesize').text),
  57             'video_bitrate': int(fe.find('./videoBitrate').text),
  58             '3sat_qualityname': fe.find('./quality').text,
  59         } for fe in format_els
  60             if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
  61
  62         def _sortkey(format):
  63             qidx = ['low', 'med', 'high', 'veryhigh'].index(format['3sat_qualityname'])
  64             prefer_http = 1 if 'rtmp' in format['url'] else 0
  65             return (qidx, prefer_http, format['video_bitrate'])
  66         formats.sort(key=_sortkey)
  67
  68         return {
  69             '_type': 'video',
  70             'id': video_id,
  71             'title': video_title,
  72             'formats': formats,
  73             'description': video_description,
  74             'thumbnails': thumbnails,
  75             'thumbnail': thumbnails[-1]['url'],
  76             'uploader': video_uploader,
  77             'upload_date': upload_date,
  78         }