_ 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     determine_ext,
  14     qualities,
  15     float_or_none,
  16 )
  17
  18
  19 class ZDFIE(InfoExtractor):
  20     _VALID_URL = r'(?:zdf:|zdf:video:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/(.*beitrag/(?:video/)?))(?P<id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
  21
  22     _TESTS = [{
  23         'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt',
  24         'info_dict': {
  25             'id': '2037704',
  26             'ext': 'webm',
  27             'title': 'ZDFspezial - Ende des Machtpokers',
  28             '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".',
  29             'duration': 1022,
  30             'uploader': 'spezial',
  31             'uploader_id': '225948',
  32             'upload_date': '20131127',
  33         },
  34         'skip': 'Videos on ZDF.de are depublicised in short order',
  35     }]
  36
  37     def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
  38         param_groups = {}
  39         for param_group in smil.findall(self._xpath_ns('./head/paramGroup', namespace)):
  40             group_id = param_group.attrib.get(self._xpath_ns('id', 'http://www.w3.org/XML/1998/namespace'))
  41             params = {}
  42             for param in param_group:
  43                 params[param.get('name')] = param.get('value')
  44             param_groups[group_id] = params
  45
  46         formats = []
  47         for video in smil.findall(self._xpath_ns('.//video', namespace)):
  48             src = video.get('src')
  49             if not src:
  50                 continue
  51             bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
  52             group_id = video.get('paramGroup')
  53             param_group = param_groups[group_id]
  54             for proto in param_group['protocols'].split(','):
  55                 formats.append({
  56                     'url': '%s://%s' % (proto, param_group['host']),
  57                     'app': param_group['app'],
  58                     'play_path': src,
  59                     'ext': 'flv',
  60                     'format_id': '%s-%d' % (proto, bitrate),
  61                     'tbr': bitrate,
  62                     'protocol': proto,
  63                 })
  64         self._sort_formats(formats)
  65         return formats
  66
  67     def extract_from_xml_url(self, video_id, xml_url):
  68         doc = self._download_xml(
  69             xml_url, video_id,
  70             note='Downloading video info',
  71             errnote='Failed to download video info')
  72
  73         title = doc.find('.//information/title').text
  74         description = xpath_text(doc, './/information/detail', 'description')
  75         duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration'))
  76         uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader')
  77         uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id')
  78         upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date'))
  79
  80         def xml_to_thumbnails(fnode):
  81             thumbnails = []
  82             for node in fnode:
  83                 thumbnail_url = node.text
  84                 if not thumbnail_url:
  85                     continue
  86                 thumbnail = {
  87                     'url': thumbnail_url,
  88                 }
  89                 if 'key' in node.attrib:
  90                     m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key'])
  91                     if m:
  92                         thumbnail['width'] = int(m.group(1))
  93                         thumbnail['height'] = int(m.group(2))
  94                 thumbnails.append(thumbnail)
  95             return thumbnails
  96
  97         thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage'))
  98
  99         format_nodes = doc.findall('.//formitaeten/formitaet')
 100         quality = qualities(['veryhigh', 'high', 'med', 'low'])
 101
 102         def get_quality(elem):
 103             return quality(xpath_text(elem, 'quality'))
 104         format_nodes.sort(key=get_quality)
 105         format_ids = []
 106         formats = []
 107         for fnode in format_nodes:
 108             video_url = fnode.find('url').text
 109             is_available = 'http://www.metafilegenerator' not in video_url
 110             if not is_available:
 111                 continue
 112             format_id = fnode.attrib['basetype']
 113             quality = xpath_text(fnode, './quality', 'quality')
 114             format_m = re.match(r'''(?x)
 115                 (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
 116                 (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
 117             ''', format_id)
 118
 119             ext = determine_ext(video_url, None) or format_m.group('container')
 120             if ext not in ('smil', 'f4m', 'm3u8'):
 121                 format_id = format_id + '-' + quality
 122             if format_id in format_ids:
 123                 continue
 124
 125             if ext == 'meta':
 126                 continue
 127             elif ext == 'smil':
 128                 formats.extend(self._extract_smil_formats(
 129                     video_url, video_id, fatal=False))
 130             elif ext == 'm3u8':
 131                 formats.extend(self._extract_m3u8_formats(
 132                     video_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
 133             elif ext == 'f4m':
 134                 formats.extend(self._extract_f4m_formats(
 135                     video_url, video_id, f4m_id='hds', fatal=False))
 136             else:
 137                 proto = format_m.group('proto').lower()
 138
 139                 abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000)
 140                 vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000)
 141
 142                 width = int_or_none(xpath_text(fnode, './width', 'width'))
 143                 height = int_or_none(xpath_text(fnode, './height', 'height'))
 144
 145                 filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize'))
 146
 147                 format_note = ''
 148                 if not format_note:
 149                     format_note = None
 150
 151                 formats.append({
 152                     'format_id': format_id,
 153                     'url': video_url,
 154                     'ext': ext,
 155                     'acodec': format_m.group('acodec'),
 156                     'vcodec': format_m.group('vcodec'),
 157                     'abr': abr,
 158                     'vbr': vbr,
 159                     'width': width,
 160                     'height': height,
 161                     'filesize': filesize,
 162                     'format_note': format_note,
 163                     'protocol': proto,
 164                     '_available': is_available,
 165                 })
 166             format_ids.append(format_id)
 167
 168         self._sort_formats(formats)
 169
 170         return {
 171             'id': video_id,
 172             'title': title,
 173             'description': description,
 174             'duration': duration,
 175             'thumbnails': thumbnails,
 176             'uploader': uploader,
 177             'uploader_id': uploader_id,
 178             'upload_date': upload_date,
 179             'formats': formats,
 180         }
 181
 182     def _real_extract(self, url):
 183         video_id = self._match_id(url)
 184         xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
 185         return self.extract_from_xml_url(video_id, xml_url)
 186
 187
 188 class ZDFChannelIE(InfoExtractor):
 189     _VALID_URL = r'(?:zdf:topic:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/(?:[^/]+/)?)(?P<id>[0-9]+)'
 190     _TESTS = [{
 191         'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic',
 192         'info_dict': {
 193             'id': '1586442',
 194         },
 195         'playlist_count': 3,
 196     }, {
 197         'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/aktuellste/332',
 198         'only_matching': True,
 199     }, {
 200         'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/meist-gesehen/332',
 201         'only_matching': True,
 202     }, {
 203         'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/_/1798716?bc=nrt;nrm?flash=off',
 204         'only_matching': True,
 205     }]
 206     _PAGE_SIZE = 50
 207
 208     def _fetch_page(self, channel_id, page):
 209         offset = page * self._PAGE_SIZE
 210         xml_url = (
 211             'http://www.zdf.de/ZDFmediathek/xmlservice/web/aktuellste?ak=web&offset=%d&maxLength=%d&id=%s'
 212             % (offset, self._PAGE_SIZE, channel_id))
 213         doc = self._download_xml(
 214             xml_url, channel_id,
 215             note='Downloading channel info',
 216             errnote='Failed to download channel info')
 217
 218         title = doc.find('.//information/title').text
 219         description = doc.find('.//information/detail').text
 220         for asset in doc.findall('.//teasers/teaser'):
 221             a_type = asset.find('./type').text
 222             a_id = asset.find('./details/assetId').text
 223             if a_type not in ('video', 'topic'):
 224                 continue
 225             yield {
 226                 '_type': 'url',
 227                 'playlist_title': title,
 228                 'playlist_description': description,
 229                 'url': 'zdf:%s:%s' % (a_type, a_id),
 230             }
 231
 232     def _real_extract(self, url):
 233         channel_id = self._match_id(url)
 234         entries = OnDemandPagedList(
 235             functools.partial(self._fetch_page, channel_id), self._PAGE_SIZE)
 236
 237         return {
 238             '_type': 'playlist',
 239             'id': channel_id,
 240             'entries': entries,
 241         }