_ 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     ExtractorError,
  17 )
  18
  19
  20 class ZDFIE(InfoExtractor):
  21     _VALID_URL = r'(?:zdf:|zdf:video:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/(.*beitrag/(?:video/)?))(?P<id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
  22
  23     _TESTS = [{
  24         'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt',
  25         'info_dict': {
  26             'id': '2037704',
  27             'ext': 'webm',
  28             'title': 'ZDFspezial - Ende des Machtpokers',
  29             '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".',
  30             'duration': 1022,
  31             'uploader': 'spezial',
  32             'uploader_id': '225948',
  33             'upload_date': '20131127',
  34         },
  35         'skip': 'Videos on ZDF.de are depublicised in short order',
  36     }]
  37
  38     def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
  39         param_groups = {}
  40         for param_group in smil.findall(self._xpath_ns('./head/paramGroup', namespace)):
  41             group_id = param_group.attrib.get(self._xpath_ns('id', 'http://www.w3.org/XML/1998/namespace'))
  42             params = {}
  43             for param in param_group:
  44                 params[param.get('name')] = param.get('value')
  45             param_groups[group_id] = params
  46
  47         formats = []
  48         for video in smil.findall(self._xpath_ns('.//video', namespace)):
  49             src = video.get('src')
  50             if not src:
  51                 continue
  52             bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
  53             group_id = video.get('paramGroup')
  54             param_group = param_groups[group_id]
  55             for proto in param_group['protocols'].split(','):
  56                 formats.append({
  57                     'url': '%s://%s' % (proto, param_group['host']),
  58                     'app': param_group['app'],
  59                     'play_path': src,
  60                     'ext': 'flv',
  61                     'format_id': '%s-%d' % (proto, bitrate),
  62                     'tbr': bitrate,
  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         status_code = doc.find('./status/statuscode')
  74         if status_code is not None and status_code.text != 'ok':
  75             code = status_code.text
  76             if code == 'notVisibleAnymore':
  77                 message = 'Video %s is not available' % video_id
  78             else:
  79                 message = '%s returned error: %s' % (self.IE_NAME, code)
  80             raise ExtractorError(message, expected=True)
  81
  82         title = doc.find('.//information/title').text
  83         description = xpath_text(doc, './/information/detail', 'description')
  84         duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration'))
  85         uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader')
  86         uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id')
  87         upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date'))
  88
  89         def xml_to_thumbnails(fnode):
  90             thumbnails = []
  91             for node in fnode:
  92                 thumbnail_url = node.text
  93                 if not thumbnail_url:
  94                     continue
  95                 thumbnail = {
  96                     'url': thumbnail_url,
  97                 }
  98                 if 'key' in node.attrib:
  99                     m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key'])
 100                     if m:
 101                         thumbnail['width'] = int(m.group(1))
 102                         thumbnail['height'] = int(m.group(2))
 103                 thumbnails.append(thumbnail)
 104             return thumbnails
 105
 106         thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage'))
 107
 108         format_nodes = doc.findall('.//formitaeten/formitaet')
 109         quality = qualities(['veryhigh', 'high', 'med', 'low'])
 110
 111         def get_quality(elem):
 112             return quality(xpath_text(elem, 'quality'))
 113         format_nodes.sort(key=get_quality)
 114         format_ids = []
 115         formats = []
 116         for fnode in format_nodes:
 117             video_url = fnode.find('url').text
 118             is_available = 'http://www.metafilegenerator' not in video_url
 119             if not is_available:
 120                 continue
 121             format_id = fnode.attrib['basetype']
 122             quality = xpath_text(fnode, './quality', 'quality')
 123             format_m = re.match(r'''(?x)
 124                 (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
 125                 (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
 126             ''', format_id)
 127
 128             ext = determine_ext(video_url, None) or format_m.group('container')
 129             if ext not in ('smil', 'f4m', 'm3u8'):
 130                 format_id = format_id + '-' + quality
 131             if format_id in format_ids:
 132                 continue
 133
 134             if ext == 'meta':
 135                 continue
 136             elif ext == 'smil':
 137                 formats.extend(self._extract_smil_formats(
 138                     video_url, video_id, fatal=False))
 139             elif ext == 'm3u8':
 140                 # the certificates are misconfigured (see
 141                 # https://github.com/rg3/youtube-dl/issues/8665)
 142                 if video_url.startswith('https://'):
 143                     continue
 144                 formats.extend(self._extract_m3u8_formats(
 145                     video_url, video_id, 'mp4', m3u8_id=format_id, fatal=False))
 146             elif ext == 'f4m':
 147                 formats.extend(self._extract_f4m_formats(
 148                     video_url, video_id, f4m_id=format_id, fatal=False))
 149             else:
 150                 proto = format_m.group('proto').lower()
 151
 152                 abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000)
 153                 vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000)
 154
 155                 width = int_or_none(xpath_text(fnode, './width', 'width'))
 156                 height = int_or_none(xpath_text(fnode, './height', 'height'))
 157
 158                 filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize'))
 159
 160                 format_note = ''
 161                 if not format_note:
 162                     format_note = None
 163
 164                 formats.append({
 165                     'format_id': format_id,
 166                     'url': video_url,
 167                     'ext': ext,
 168                     'acodec': format_m.group('acodec'),
 169                     'vcodec': format_m.group('vcodec'),
 170                     'abr': abr,
 171                     'vbr': vbr,
 172                     'width': width,
 173                     'height': height,
 174                     'filesize': filesize,
 175                     'format_note': format_note,
 176                     'protocol': proto,
 177                     '_available': is_available,
 178                 })
 179             format_ids.append(format_id)
 180
 181         self._sort_formats(formats)
 182
 183         return {
 184             'id': video_id,
 185             'title': title,
 186             'description': description,
 187             'duration': duration,
 188             'thumbnails': thumbnails,
 189             'uploader': uploader,
 190             'uploader_id': uploader_id,
 191             'upload_date': upload_date,
 192             'formats': formats,
 193         }
 194
 195     def _real_extract(self, url):
 196         video_id = self._match_id(url)
 197         xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
 198         return self.extract_from_xml_url(video_id, xml_url)
 199
 200
 201 class ZDFChannelIE(InfoExtractor):
 202     _VALID_URL = r'(?:zdf:topic:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/(?:[^/]+/)?)(?P<id>[0-9]+)'
 203     _TESTS = [{
 204         'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic',
 205         'info_dict': {
 206             'id': '1586442',
 207         },
 208         'playlist_count': 3,
 209     }, {
 210         'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/aktuellste/332',
 211         'only_matching': True,
 212     }, {
 213         'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/meist-gesehen/332',
 214         'only_matching': True,
 215     }, {
 216         'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/_/1798716?bc=nrt;nrm?flash=off',
 217         'only_matching': True,
 218     }]
 219     _PAGE_SIZE = 50
 220
 221     def _fetch_page(self, channel_id, page):
 222         offset = page * self._PAGE_SIZE
 223         xml_url = (
 224             'http://www.zdf.de/ZDFmediathek/xmlservice/web/aktuellste?ak=web&offset=%d&maxLength=%d&id=%s'
 225             % (offset, self._PAGE_SIZE, channel_id))
 226         doc = self._download_xml(
 227             xml_url, channel_id,
 228             note='Downloading channel info',
 229             errnote='Failed to download channel info')
 230
 231         title = doc.find('.//information/title').text
 232         description = doc.find('.//information/detail').text
 233         for asset in doc.findall('.//teasers/teaser'):
 234             a_type = asset.find('./type').text
 235             a_id = asset.find('./details/assetId').text
 236             if a_type not in ('video', 'topic'):
 237                 continue
 238             yield {
 239                 '_type': 'url',
 240                 'playlist_title': title,
 241                 'playlist_description': description,
 242                 'url': 'zdf:%s:%s' % (a_type, a_id),
 243             }
 244
 245     def _real_extract(self, url):
 246         channel_id = self._match_id(url)
 247         entries = OnDemandPagedList(
 248             functools.partial(self._fetch_page, channel_id), self._PAGE_SIZE)
 249
 250         return {
 251             '_type': 'playlist',
 252             'id': channel_id,
 253             'entries': entries,
 254         }