_ Git - youtube-dl/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     xpath_text,
  13 )
  14
  15
  16 def extract_from_xml_url(ie, video_id, xml_url):
  17     doc = ie._download_xml(
  18         xml_url, video_id,
  19         note='Downloading video info',
  20         errnote='Failed to download video info')
  21
  22     title = doc.find('.//information/title').text
  23     description = xpath_text(doc, './/information/detail', 'description')
  24     duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration'))
  25     uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader')
  26     uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id')
  27     upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date'))
  28
  29     def xml_to_format(fnode):
  30         video_url = fnode.find('url').text
  31         is_available = 'http://www.metafilegenerator' not in video_url
  32
  33         format_id = fnode.attrib['basetype']
  34         format_m = re.match(r'''(?x)
  35             (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
  36             (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
  37         ''', format_id)
  38
  39         ext = format_m.group('container')
  40         proto = format_m.group('proto').lower()
  41
  42         quality = xpath_text(fnode, './quality', 'quality')
  43         abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000)
  44         vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000)
  45
  46         width = int_or_none(xpath_text(fnode, './width', 'width'))
  47         height = int_or_none(xpath_text(fnode, './height', 'height'))
  48
  49         filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize'))
  50
  51         format_note = ''
  52         if not format_note:
  53             format_note = None
  54
  55         return {
  56             'format_id': format_id + '-' + quality,
  57             'url': video_url,
  58             'ext': ext,
  59             'acodec': format_m.group('acodec'),
  60             'vcodec': format_m.group('vcodec'),
  61             'abr': abr,
  62             'vbr': vbr,
  63             'width': width,
  64             'height': height,
  65             'filesize': filesize,
  66             'format_note': format_note,
  67             'protocol': proto,
  68             '_available': is_available,
  69         }
  70
  71     def xml_to_thumbnails(fnode):
  72         thumbnails = []
  73         for node in fnode:
  74             thumbnail_url = node.text
  75             if not thumbnail_url:
  76                 continue
  77             thumbnail = {
  78                 'url': thumbnail_url,
  79             }
  80             if 'key' in node.attrib:
  81                 m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key'])
  82                 if m:
  83                     thumbnail['width'] = int(m.group(1))
  84                     thumbnail['height'] = int(m.group(2))
  85             thumbnails.append(thumbnail)
  86         return thumbnails
  87
  88     thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage'))
  89
  90     format_nodes = doc.findall('.//formitaeten/formitaet')
  91     formats = list(filter(
  92         lambda f: f['_available'],
  93         map(xml_to_format, format_nodes)))
  94     ie._sort_formats(formats)
  95
  96     return {
  97         'id': video_id,
  98         'title': title,
  99         'description': description,
 100         'duration': duration,
 101         'thumbnails': thumbnails,
 102         'uploader': uploader,
 103         'uploader_id': uploader_id,
 104         'upload_date': upload_date,
 105         'formats': formats,
 106     }
 107
 108
 109 class ZDFIE(InfoExtractor):
 110     _VALID_URL = r'(?:zdf:|zdf:video:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/(.*beitrag/(?:video/)?))(?P<id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
 111
 112     _TEST = {
 113         'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt',
 114         'info_dict': {
 115             'id': '2037704',
 116             'ext': 'webm',
 117             'title': 'ZDFspezial - Ende des Machtpokers',
 118             '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".',
 119             'duration': 1022,
 120             'uploader': 'spezial',
 121             'uploader_id': '225948',
 122             'upload_date': '20131127',
 123         },
 124         'skip': 'Videos on ZDF.de are depublicised in short order',
 125     }
 126
 127     def _real_extract(self, url):
 128         video_id = self._match_id(url)
 129         xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
 130         return extract_from_xml_url(self, video_id, xml_url)
 131
 132
 133 class ZDFChannelIE(InfoExtractor):
 134     _VALID_URL = r'(?:zdf:topic:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/)(?P<id>[0-9]+)'
 135     _TEST = {
 136         'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic',
 137         'info_dict': {
 138             'id': '1586442',
 139         },
 140         'playlist_count': 3,
 141     }
 142     _PAGE_SIZE = 50
 143
 144     def _fetch_page(self, channel_id, page):
 145         offset = page * self._PAGE_SIZE
 146         xml_url = (
 147             'http://www.zdf.de/ZDFmediathek/xmlservice/web/aktuellste?ak=web&offset=%d&maxLength=%d&id=%s'
 148             % (offset, self._PAGE_SIZE, channel_id))
 149         doc = self._download_xml(
 150             xml_url, channel_id,
 151             note='Downloading channel info',
 152             errnote='Failed to download channel info')
 153
 154         title = doc.find('.//information/title').text
 155         description = doc.find('.//information/detail').text
 156         for asset in doc.findall('.//teasers/teaser'):
 157             a_type = asset.find('./type').text
 158             a_id = asset.find('./details/assetId').text
 159             if a_type not in ('video', 'topic'):
 160                 continue
 161             yield {
 162                 '_type': 'url',
 163                 'playlist_title': title,
 164                 'playlist_description': description,
 165                 'url': 'zdf:%s:%s' % (a_type, a_id),
 166             }
 167
 168     def _real_extract(self, url):
 169         channel_id = self._match_id(url)
 170         entries = OnDemandPagedList(
 171             functools.partial(self._fetch_page, channel_id), self._PAGE_SIZE)
 172
 173         return {
 174             '_type': 'playlist',
 175             'id': channel_id,
 176             'entries': entries,
 177         }