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

   1 import operator
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     parse_xml_doc,
   7     unified_strdate,
   8 )
   9
  10
  11 class ZDFIE(InfoExtractor):
  12     _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek(?P<hash>#)?\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?'
  13
  14     def _real_extract(self, url):
  15         mobj = re.match(self._VALID_URL, url)
  16         video_id = mobj.group('video_id')
  17
  18         xml_url = u'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
  19         info_xml = self._download_webpage(
  20             xml_url, video_id, note=u'Downloading video info')
  21         doc = parse_xml_doc(info_xml)
  22
  23         title = doc.find('.//information/title').text
  24         description = doc.find('.//information/detail').text
  25         uploader_node = doc.find('.//details/originChannelTitle')
  26         uploader = None if uploader_node is None else uploader_node.text
  27         duration_str = doc.find('.//details/length').text
  28         duration_m = re.match(r'''(?x)^
  29             (?P<hours>[0-9]{2})
  30             :(?P<minutes>[0-9]{2})
  31             :(?P<seconds>[0-9]{2})
  32             (?:\.(?P<ms>[0-9]+)?)
  33             ''', duration_str)
  34         duration = (
  35             (
  36                 (int(duration_m.group('hours')) * 60 * 60) +
  37                 (int(duration_m.group('minutes')) * 60) +
  38                 int(duration_m.group('seconds'))
  39             )
  40             if duration_m
  41             else None
  42         )
  43         upload_date = unified_strdate(doc.find('.//details/airtime').text)
  44
  45         def xml_to_format(fnode):
  46             video_url = fnode.find('url').text
  47             is_available = u'http://www.metafilegenerator' not in video_url
  48
  49             format_id = fnode.attrib['basetype']
  50             format_m = re.match(r'''(?x)
  51                 (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
  52                 (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
  53             ''', format_id)
  54
  55             ext = format_m.group('container')
  56             is_supported = ext != 'f4f'
  57
  58             PROTO_ORDER = ['http', 'rtmp', 'rtsp']
  59             try:
  60                 proto_pref = -PROTO_ORDER.index(format_m.group('proto'))
  61             except ValueError:
  62                 proto_pref = 999
  63
  64             quality = fnode.find('./quality').text
  65             QUALITY_ORDER = ['veryhigh', '300', 'high', 'med', 'low']
  66             try:
  67                 quality_pref = -QUALITY_ORDER.index(quality)
  68             except ValueError:
  69                 quality_pref = 999
  70
  71             abr = int(fnode.find('./audioBitrate').text) // 1000
  72             vbr = int(fnode.find('./videoBitrate').text) // 1000
  73             pref = (is_available, is_supported,
  74                     proto_pref, quality_pref, vbr, abr)
  75
  76             format_note = u''
  77             if not is_supported:
  78                 format_note += u'(unsupported)'
  79             if not format_note:
  80                 format_note = None
  81
  82             return {
  83                 'format_id': format_id + u'-' + quality,
  84                 'url': video_url,
  85                 'ext': ext,
  86                 'acodec': format_m.group('acodec'),
  87                 'vcodec': format_m.group('vcodec'),
  88                 'abr': abr,
  89                 'vbr': vbr,
  90                 'width': int(fnode.find('./width').text),
  91                 'height': int(fnode.find('./height').text),
  92                 'filesize': int(fnode.find('./filesize').text),
  93                 'format_note': format_note,
  94                 '_pref': pref,
  95                 '_available': is_available,
  96             }
  97
  98         format_nodes = doc.findall('.//formitaeten/formitaet')
  99         formats = sorted(filter(lambda f: f['_available'],
 100                                 map(xml_to_format, format_nodes)),
 101                          key=operator.itemgetter('_pref'))
 102
 103         return {
 104             'id': video_id,
 105             'title': title,
 106             'formats': formats,
 107             'description': description,
 108             'uploader': uploader,
 109             'duration': duration,
 110             'upload_date': upload_date,
 111         }