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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     unified_strdate,
  10 )
  11
  12
  13 class ZDFIE(InfoExtractor):
  14     _VALID_URL = r'^https?://www\.zdf\.de/ZDFmediathek(?P<hash>#)?/(.*beitrag/(?:video/)?)(?P<video_id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
  15
  16     _TEST = {
  17         'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt',
  18         'info_dict': {
  19             'id': '2037704',
  20             'ext': 'webm',
  21             'title': 'ZDFspezial - Ende des Machtpokers',
  22             'description': 'Union und SPD haben sich auf einen Koalitionsvertrag geeinigt. Aber was bedeutet das für die Bürger? Sehen Sie hierzu das ZDFspezial "Ende des Machtpokers - Große Koalition für Deutschland".',
  23             'duration': 1022,
  24             'uploader': 'spezial',
  25             'uploader_id': '225948',
  26             'upload_date': '20131127',
  27         },
  28         'skip': 'Videos on ZDF.de are depublicised in short order',
  29     }
  30
  31     def _real_extract(self, url):
  32         mobj = re.match(self._VALID_URL, url)
  33         video_id = mobj.group('video_id')
  34
  35         xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
  36         doc = self._download_xml(
  37             xml_url, video_id,
  38             note='Downloading video info',
  39             errnote='Failed to download video info')
  40
  41         title = doc.find('.//information/title').text
  42         description = doc.find('.//information/detail').text
  43         duration = int(doc.find('.//details/lengthSec').text)
  44         uploader_node = doc.find('.//details/originChannelTitle')
  45         uploader = None if uploader_node is None else uploader_node.text
  46         uploader_id_node = doc.find('.//details/originChannelId')
  47         uploader_id = None if uploader_id_node is None else uploader_id_node.text
  48         upload_date = unified_strdate(doc.find('.//details/airtime').text)
  49
  50         def xml_to_format(fnode):
  51             video_url = fnode.find('url').text
  52             is_available = 'http://www.metafilegenerator' not in video_url
  53
  54             format_id = fnode.attrib['basetype']
  55             format_m = re.match(r'''(?x)
  56                 (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
  57                 (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
  58             ''', format_id)
  59
  60             ext = format_m.group('container')
  61             proto = format_m.group('proto').lower()
  62
  63             quality = fnode.find('./quality').text
  64             abr = int(fnode.find('./audioBitrate').text) // 1000
  65             vbr_node = fnode.find('./videoBitrate')
  66             vbr = None if vbr_node is None else int(vbr_node.text) // 1000
  67
  68             width_node = fnode.find('./width')
  69             width = None if width_node is None else int_or_none(width_node.text)
  70             height_node = fnode.find('./height')
  71             height = None if height_node is None else int_or_none(height_node.text)
  72
  73             format_note = ''
  74             if not format_note:
  75                 format_note = None
  76
  77             return {
  78                 'format_id': format_id + '-' + quality,
  79                 'url': video_url,
  80                 'ext': ext,
  81                 'acodec': format_m.group('acodec'),
  82                 'vcodec': format_m.group('vcodec'),
  83                 'abr': abr,
  84                 'vbr': vbr,
  85                 'width': width,
  86                 'height': height,
  87                 'filesize': int_or_none(fnode.find('./filesize').text),
  88                 'format_note': format_note,
  89                 'protocol': proto,
  90                 '_available': is_available,
  91             }
  92
  93         format_nodes = doc.findall('.//formitaeten/formitaet')
  94         formats = list(filter(
  95             lambda f: f['_available'],
  96             map(xml_to_format, format_nodes)))
  97
  98         self._sort_formats(formats)
  99
 100         return {
 101             'id': video_id,
 102             'title': title,
 103             'description': description,
 104             'duration': duration,
 105             'uploader': uploader,
 106             'uploader_id': uploader_id,
 107             'upload_date': upload_date,
 108             'formats': formats,
 109         }