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     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         subtitles = {}
  89         captions_url = doc.find('.//caption/url')
  90         if captions_url is not None:
  91             subtitles['de'] = [{
  92                 'url': captions_url.text,
  93                 'ext': 'ttml',
  94             }]
  95
  96         def xml_to_thumbnails(fnode):
  97             thumbnails = []
  98             for node in fnode:
  99                 thumbnail_url = node.text
 100                 if not thumbnail_url:
 101                     continue
 102                 thumbnail = {
 103                     'url': thumbnail_url,
 104                 }
 105                 if 'key' in node.attrib:
 106                     m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key'])
 107                     if m:
 108                         thumbnail['width'] = int(m.group(1))
 109                         thumbnail['height'] = int(m.group(2))
 110                 thumbnails.append(thumbnail)
 111             return thumbnails
 112
 113         thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage'))
 114
 115         format_nodes = doc.findall('.//formitaeten/formitaet')
 116         quality = qualities(['veryhigh', 'high', 'med', 'low'])
 117
 118         def get_quality(elem):
 119             return quality(xpath_text(elem, 'quality'))
 120         format_nodes.sort(key=get_quality)
 121         format_ids = []
 122         formats = []
 123         for fnode in format_nodes:
 124             video_url = fnode.find('url').text
 125             is_available = 'http://www.metafilegenerator' not in video_url
 126             if not is_available:
 127                 continue
 128             format_id = fnode.attrib['basetype']
 129             quality = xpath_text(fnode, './quality', 'quality')
 130             format_m = re.match(r'''(?x)
 131                 (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
 132                 (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
 133             ''', format_id)
 134
 135             ext = determine_ext(video_url, None) or format_m.group('container')
 136             if ext not in ('smil', 'f4m', 'm3u8'):
 137                 format_id = format_id + '-' + quality
 138             if format_id in format_ids:
 139                 continue
 140
 141             if ext == 'meta':
 142                 continue
 143             elif ext == 'smil':
 144                 formats.extend(self._extract_smil_formats(
 145                     video_url, video_id, fatal=False))
 146             elif ext == 'm3u8':
 147                 # the certificates are misconfigured (see
 148                 # https://github.com/rg3/youtube-dl/issues/8665)
 149                 if video_url.startswith('https://'):
 150                     continue
 151                 formats.extend(self._extract_m3u8_formats(
 152                     video_url, video_id, 'mp4', m3u8_id=format_id, fatal=False))
 153             elif ext == 'f4m':
 154                 formats.extend(self._extract_f4m_formats(
 155                     video_url, video_id, f4m_id=format_id, fatal=False))
 156             else:
 157                 proto = format_m.group('proto').lower()
 158
 159                 abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000)
 160                 vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000)
 161
 162                 width = int_or_none(xpath_text(fnode, './width', 'width'))
 163                 height = int_or_none(xpath_text(fnode, './height', 'height'))
 164
 165                 filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize'))
 166
 167                 format_note = ''
 168                 if not format_note:
 169                     format_note = None
 170
 171                 formats.append({
 172                     'format_id': format_id,
 173                     'url': video_url,
 174                     'ext': ext,
 175                     'acodec': format_m.group('acodec'),
 176                     'vcodec': format_m.group('vcodec'),
 177                     'abr': abr,
 178                     'vbr': vbr,
 179                     'width': width,
 180                     'height': height,
 181                     'filesize': filesize,
 182                     'format_note': format_note,
 183                     'protocol': proto,
 184                     '_available': is_available,
 185                 })
 186             format_ids.append(format_id)
 187
 188         self._sort_formats(formats)
 189
 190         return {
 191             'id': video_id,
 192             'title': title,
 193             'description': description,
 194             'duration': duration,
 195             'thumbnails': thumbnails,
 196             'uploader': uploader,
 197             'uploader_id': uploader_id,
 198             'upload_date': upload_date,
 199             'formats': formats,
 200             'subtitles': subtitles,
 201         }
 202
 203     def _real_extract(self, url):
 204         video_id = self._match_id(url)
 205         xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
 206         return self.extract_from_xml_url(video_id, xml_url)
 207
 208
 209 class ZDFChannelIE(InfoExtractor):
 210     _VALID_URL = r'(?:zdf:topic:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/(?:[^/]+/)?)(?P<id>[0-9]+)'
 211     _TESTS = [{
 212         'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic',
 213         'info_dict': {
 214             'id': '1586442',
 215         },
 216         'playlist_count': 3,
 217     }, {
 218         'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/aktuellste/332',
 219         'only_matching': True,
 220     }, {
 221         'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/meist-gesehen/332',
 222         'only_matching': True,
 223     }, {
 224         'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/_/1798716?bc=nrt;nrm?flash=off',
 225         'only_matching': True,
 226     }]
 227     _PAGE_SIZE = 50
 228
 229     def _fetch_page(self, channel_id, page):
 230         offset = page * self._PAGE_SIZE
 231         xml_url = (
 232             'http://www.zdf.de/ZDFmediathek/xmlservice/web/aktuellste?ak=web&offset=%d&maxLength=%d&id=%s'
 233             % (offset, self._PAGE_SIZE, channel_id))
 234         doc = self._download_xml(
 235             xml_url, channel_id,
 236             note='Downloading channel info',
 237             errnote='Failed to download channel info')
 238
 239         title = doc.find('.//information/title').text
 240         description = doc.find('.//information/detail').text
 241         for asset in doc.findall('.//teasers/teaser'):
 242             a_type = asset.find('./type').text
 243             a_id = asset.find('./details/assetId').text
 244             if a_type not in ('video', 'topic'):
 245                 continue
 246             yield {
 247                 '_type': 'url',
 248                 'playlist_title': title,
 249                 'playlist_description': description,
 250                 'url': 'zdf:%s:%s' % (a_type, a_id),
 251             }
 252
 253     def _real_extract(self, url):
 254         channel_id = self._match_id(url)
 255         entries = OnDemandPagedList(
 256             functools.partial(self._fetch_page, channel_id), self._PAGE_SIZE)
 257
 258         return {
 259             '_type': 'playlist',
 260             'id': channel_id,
 261             'entries': entries,
 262         }