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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import functools
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     int_or_none,
  10     unified_strdate,
  11     OnDemandPagedList,
  12 )
  13
  14
  15 def extract_from_xml_url(ie, video_id, xml_url):
  16     doc = ie._download_xml(
  17         xml_url, video_id,
  18         note='Downloading video info',
  19         errnote='Failed to download video info')
  20
  21     title = doc.find('.//information/title').text
  22     description = doc.find('.//information/detail').text
  23     duration = int(doc.find('.//details/lengthSec').text)
  24     uploader_node = doc.find('.//details/originChannelTitle')
  25     uploader = None if uploader_node is None else uploader_node.text
  26     uploader_id_node = doc.find('.//details/originChannelId')
  27     uploader_id = None if uploader_id_node is None else uploader_id_node.text
  28     upload_date = unified_strdate(doc.find('.//details/airtime').text)
  29
  30     def xml_to_format(fnode):
  31         video_url = fnode.find('url').text
  32         is_available = 'http://www.metafilegenerator' not in video_url
  33
  34         format_id = fnode.attrib['basetype']
  35         format_m = re.match(r'''(?x)
  36             (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
  37             (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
  38         ''', format_id)
  39
  40         ext = format_m.group('container')
  41         proto = format_m.group('proto').lower()
  42
  43         quality = fnode.find('./quality').text
  44         abr = int(fnode.find('./audioBitrate').text) // 1000
  45         vbr_node = fnode.find('./videoBitrate')
  46         vbr = None if vbr_node is None else int(vbr_node.text) // 1000
  47
  48         width_node = fnode.find('./width')
  49         width = None if width_node is None else int_or_none(width_node.text)
  50         height_node = fnode.find('./height')
  51         height = None if height_node is None else int_or_none(height_node.text)
  52
  53         format_note = ''
  54         if not format_note:
  55             format_note = None
  56
  57         return {
  58             'format_id': format_id + '-' + quality,
  59             'url': video_url,
  60             'ext': ext,
  61             'acodec': format_m.group('acodec'),
  62             'vcodec': format_m.group('vcodec'),
  63             'abr': abr,
  64             'vbr': vbr,
  65             'width': width,
  66             'height': height,
  67             'filesize': int_or_none(fnode.find('./filesize').text),
  68             'format_note': format_note,
  69             'protocol': proto,
  70             '_available': is_available,
  71         }
  72
  73     format_nodes = doc.findall('.//formitaeten/formitaet')
  74     formats = list(filter(
  75         lambda f: f['_available'],
  76         map(xml_to_format, format_nodes)))
  77     ie._sort_formats(formats)
  78
  79     return {
  80         'id': video_id,
  81         'title': title,
  82         'description': description,
  83         'duration': duration,
  84         'uploader': uploader,
  85         'uploader_id': uploader_id,
  86         'upload_date': upload_date,
  87         'formats': formats,
  88     }
  89
  90
  91 class ZDFIE(InfoExtractor):
  92     _VALID_URL = r'(?:zdf:|zdf:video:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/(.*beitrag/(?:video/)?))(?P<id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
  93
  94     _TEST = {
  95         'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt',
  96         'info_dict': {
  97             'id': '2037704',
  98             'ext': 'webm',
  99             'title': 'ZDFspezial - Ende des Machtpokers',
 100             '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".',
 101             'duration': 1022,
 102             'uploader': 'spezial',
 103             'uploader_id': '225948',
 104             'upload_date': '20131127',
 105         },
 106         'skip': 'Videos on ZDF.de are depublicised in short order',
 107     }
 108
 109     def _real_extract(self, url):
 110         video_id = self._match_id(url)
 111         xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
 112         return extract_from_xml_url(self, video_id, xml_url)
 113
 114
 115 class ZDFChannelIE(InfoExtractor):
 116     _VALID_URL = r'(?:zdf:topic:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/)(?P<id>[0-9]+)'
 117     _TEST = {
 118         'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic',
 119         'info_dict': {
 120             'id': '1586442',
 121         },
 122         'playlist_count': 3,
 123     }
 124     _PAGE_SIZE = 50
 125
 126     def _fetch_page(self, channel_id, page):
 127         offset = page * self._PAGE_SIZE
 128         xml_url = (
 129             'http://www.zdf.de/ZDFmediathek/xmlservice/web/aktuellste?ak=web&offset=%d&maxLength=%d&id=%s'
 130             % (offset, self._PAGE_SIZE, channel_id))
 131         doc = self._download_xml(
 132             xml_url, channel_id,
 133             note='Downloading channel info',
 134             errnote='Failed to download channel info')
 135
 136         title = doc.find('.//information/title').text
 137         description = doc.find('.//information/detail').text
 138         for asset in doc.findall('.//teasers/teaser'):
 139             a_type = asset.find('./type').text
 140             a_id = asset.find('./details/assetId').text
 141             if a_type not in ('video', 'topic'):
 142                 continue
 143             yield {
 144                 '_type': 'url',
 145                 'playlist_title': title,
 146                 'playlist_description': description,
 147                 'url': 'zdf:%s:%s' % (a_type, a_id),
 148             }
 149
 150     def _real_extract(self, url):
 151         channel_id = self._match_id(url)
 152         entries = OnDemandPagedList(
 153             functools.partial(self._fetch_page, channel_id), self._PAGE_SIZE)
 154
 155         return {
 156             '_type': 'playlist',
 157             'id': channel_id,
 158             'entries': entries,
 159         }