yt_dlp/extractor/br.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10     int_or_none,
  11     parse_duration,
  12     parse_iso8601,
  13     xpath_element,
  14     xpath_text,
  15 )
  16
  17
  18 class BRIE(InfoExtractor):
  19     IE_DESC = 'Bayerischer Rundfunk'
  20     _VALID_URL = r'(?P<base_url>https?://(?:www\.)?br(?:-klassik)?\.de)/(?:[a-z0-9\-_]+/)+(?P<id>[a-z0-9\-_]+)\.html'
  21
  22     _TESTS = [
  23         {
  24             'url': 'http://www.br.de/mediathek/video/sendungen/abendschau/betriebliche-altersvorsorge-104.html',
  25             'md5': '83a0477cf0b8451027eb566d88b51106',
  26             'info_dict': {
  27                 'id': '48f656ef-287e-486f-be86-459122db22cc',
  28                 'ext': 'mp4',
  29                 'title': 'Die böse Überraschung',
  30                 'description': 'md5:ce9ac81b466ce775b8018f6801b48ac9',
  31                 'duration': 180,
  32                 'uploader': 'Reinhard Weber',
  33                 'upload_date': '20150422',
  34             },
  35             'skip': '404 not found',
  36         },
  37         {
  38             'url': 'http://www.br.de/nachrichten/oberbayern/inhalt/muenchner-polizeipraesident-schreiber-gestorben-100.html',
  39             'md5': 'af3a3a4aa43ff0ce6a89504c67f427ef',
  40             'info_dict': {
  41                 'id': 'a4b83e34-123d-4b81-9f4e-c0d3121a4e05',
  42                 'ext': 'flv',
  43                 'title': 'Manfred Schreiber ist tot',
  44                 'description': 'md5:b454d867f2a9fc524ebe88c3f5092d97',
  45                 'duration': 26,
  46             },
  47             'skip': '404 not found',
  48         },
  49         {
  50             'url': 'https://www.br-klassik.de/audio/peeping-tom-premierenkritik-dance-festival-muenchen-100.html',
  51             'md5': '8b5b27c0b090f3b35eac4ab3f7a73d3d',
  52             'info_dict': {
  53                 'id': '74c603c9-26d3-48bb-b85b-079aeed66e0b',
  54                 'ext': 'aac',
  55                 'title': 'Kurzweilig und sehr bewegend',
  56                 'description': 'md5:0351996e3283d64adeb38ede91fac54e',
  57                 'duration': 296,
  58             },
  59             'skip': '404 not found',
  60         },
  61         {
  62             'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html',
  63             'md5': 'dbab0aef2e047060ea7a21fc1ce1078a',
  64             'info_dict': {
  65                 'id': '6ba73750-d405-45d3-861d-1ce8c524e059',
  66                 'ext': 'mp4',
  67                 'title': 'Umweltbewusster Häuslebauer',
  68                 'description': 'md5:d52dae9792d00226348c1dbb13c9bae2',
  69                 'duration': 116,
  70             }
  71         },
  72         {
  73             'url': 'http://www.br.de/fernsehen/br-alpha/sendungen/kant-fuer-anfaenger/kritik-der-reinen-vernunft/kant-kritik-01-metaphysik100.html',
  74             'md5': '23bca295f1650d698f94fc570977dae3',
  75             'info_dict': {
  76                 'id': 'd982c9ce-8648-4753-b358-98abb8aec43d',
  77                 'ext': 'mp4',
  78                 'title': 'Folge 1 - Metaphysik',
  79                 'description': 'md5:bb659990e9e59905c3d41e369db1fbe3',
  80                 'duration': 893,
  81                 'uploader': 'Eva Maria Steimle',
  82                 'upload_date': '20170208',
  83             }
  84         },
  85     ]
  86
  87     def _real_extract(self, url):
  88         base_url, display_id = self._match_valid_url(url).groups()
  89         page = self._download_webpage(url, display_id)
  90         xml_url = self._search_regex(
  91             r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
  92         xml = self._download_xml(base_url + xml_url, display_id)
  93
  94         medias = []
  95
  96         for xml_media in xml.findall('video') + xml.findall('audio'):
  97             media_id = xml_media.get('externalId')
  98             media = {
  99                 'id': media_id,
 100                 'title': xpath_text(xml_media, 'title', 'title', True),
 101                 'duration': parse_duration(xpath_text(xml_media, 'duration')),
 102                 'formats': self._extract_formats(xpath_element(
 103                     xml_media, 'assets'), media_id),
 104                 'thumbnails': self._extract_thumbnails(xpath_element(
 105                     xml_media, 'teaserImage/variants'), base_url),
 106                 'description': xpath_text(xml_media, 'desc'),
 107                 'webpage_url': xpath_text(xml_media, 'permalink'),
 108                 'uploader': xpath_text(xml_media, 'author'),
 109             }
 110             broadcast_date = xpath_text(xml_media, 'broadcastDate')
 111             if broadcast_date:
 112                 media['upload_date'] = ''.join(reversed(broadcast_date.split('.')))
 113             medias.append(media)
 114
 115         if len(medias) > 1:
 116             self.report_warning(
 117                 'found multiple medias; please '
 118                 'report this with the video URL to http://yt-dl.org/bug')
 119         if not medias:
 120             raise ExtractorError('No media entries found')
 121         return medias[0]
 122
 123     def _extract_formats(self, assets, media_id):
 124         formats = []
 125         for asset in assets.findall('asset'):
 126             format_url = xpath_text(asset, ['downloadUrl', 'url'])
 127             asset_type = asset.get('type')
 128             if asset_type.startswith('HDS'):
 129                 formats.extend(self._extract_f4m_formats(
 130                     format_url + '?hdcore=3.2.0', media_id, f4m_id='hds', fatal=False))
 131             elif asset_type.startswith('HLS'):
 132                 formats.extend(self._extract_m3u8_formats(
 133                     format_url, media_id, 'mp4', 'm3u8_native', m3u8_id='hds', fatal=False))
 134             else:
 135                 format_info = {
 136                     'ext': xpath_text(asset, 'mediaType'),
 137                     'width': int_or_none(xpath_text(asset, 'frameWidth')),
 138                     'height': int_or_none(xpath_text(asset, 'frameHeight')),
 139                     'tbr': int_or_none(xpath_text(asset, 'bitrateVideo')),
 140                     'abr': int_or_none(xpath_text(asset, 'bitrateAudio')),
 141                     'vcodec': xpath_text(asset, 'codecVideo'),
 142                     'acodec': xpath_text(asset, 'codecAudio'),
 143                     'container': xpath_text(asset, 'mediaType'),
 144                     'filesize': int_or_none(xpath_text(asset, 'size')),
 145                 }
 146                 format_url = self._proto_relative_url(format_url)
 147                 if format_url:
 148                     http_format_info = format_info.copy()
 149                     http_format_info.update({
 150                         'url': format_url,
 151                         'format_id': 'http-%s' % asset_type,
 152                     })
 153                     formats.append(http_format_info)
 154                 server_prefix = xpath_text(asset, 'serverPrefix')
 155                 if server_prefix:
 156                     rtmp_format_info = format_info.copy()
 157                     rtmp_format_info.update({
 158                         'url': server_prefix,
 159                         'play_path': xpath_text(asset, 'fileName'),
 160                         'format_id': 'rtmp-%s' % asset_type,
 161                     })
 162                     formats.append(rtmp_format_info)
 163         self._sort_formats(formats)
 164         return formats
 165
 166     def _extract_thumbnails(self, variants, base_url):
 167         thumbnails = [{
 168             'url': base_url + xpath_text(variant, 'url'),
 169             'width': int_or_none(xpath_text(variant, 'width')),
 170             'height': int_or_none(xpath_text(variant, 'height')),
 171         } for variant in variants.findall('variant') if xpath_text(variant, 'url')]
 172         thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
 173         return thumbnails
 174
 175
 176 class BRMediathekIE(InfoExtractor):
 177     IE_DESC = 'Bayerischer Rundfunk Mediathek'
 178     _VALID_URL = r'https?://(?:www\.)?br\.de/mediathek/video/[^/?&#]*?-(?P<id>av:[0-9a-f]{24})'
 179
 180     _TESTS = [{
 181         'url': 'https://www.br.de/mediathek/video/gesundheit-die-sendung-vom-28112017-av:5a1e6a6e8fce6d001871cc8e',
 182         'md5': 'fdc3d485835966d1622587d08ba632ec',
 183         'info_dict': {
 184             'id': 'av:5a1e6a6e8fce6d001871cc8e',
 185             'ext': 'mp4',
 186             'title': 'Die Sendung vom 28.11.2017',
 187             'description': 'md5:6000cdca5912ab2277e5b7339f201ccc',
 188             'timestamp': 1511942766,
 189             'upload_date': '20171129',
 190         }
 191     }]
 192
 193     def _real_extract(self, url):
 194         clip_id = self._match_id(url)
 195
 196         clip = self._download_json(
 197             'https://proxy-base.master.mango.express/graphql',
 198             clip_id, data=json.dumps({
 199                 "query": """{
 200   viewer {
 201     clip(id: "%s") {
 202       title
 203       description
 204       duration
 205       createdAt
 206       ageRestriction
 207       videoFiles {
 208         edges {
 209           node {
 210             publicLocation
 211             fileSize
 212             videoProfile {
 213               width
 214               height
 215               bitrate
 216               encoding
 217             }
 218           }
 219         }
 220       }
 221       captionFiles {
 222         edges {
 223           node {
 224             publicLocation
 225           }
 226         }
 227       }
 228       teaserImages {
 229         edges {
 230           node {
 231             imageFiles {
 232               edges {
 233                 node {
 234                   publicLocation
 235                   width
 236                   height
 237                 }
 238               }
 239             }
 240           }
 241         }
 242       }
 243     }
 244   }
 245 }""" % clip_id}).encode(), headers={
 246                 'Content-Type': 'application/json',
 247             })['data']['viewer']['clip']
 248         title = clip['title']
 249
 250         formats = []
 251         for edge in clip.get('videoFiles', {}).get('edges', []):
 252             node = edge.get('node', {})
 253             n_url = node.get('publicLocation')
 254             if not n_url:
 255                 continue
 256             ext = determine_ext(n_url)
 257             if ext == 'm3u8':
 258                 formats.extend(self._extract_m3u8_formats(
 259                     n_url, clip_id, 'mp4', 'm3u8_native',
 260                     m3u8_id='hls', fatal=False))
 261             else:
 262                 video_profile = node.get('videoProfile', {})
 263                 tbr = int_or_none(video_profile.get('bitrate'))
 264                 format_id = 'http'
 265                 if tbr:
 266                     format_id += '-%d' % tbr
 267                 formats.append({
 268                     'format_id': format_id,
 269                     'url': n_url,
 270                     'width': int_or_none(video_profile.get('width')),
 271                     'height': int_or_none(video_profile.get('height')),
 272                     'tbr': tbr,
 273                     'filesize': int_or_none(node.get('fileSize')),
 274                 })
 275         self._sort_formats(formats)
 276
 277         subtitles = {}
 278         for edge in clip.get('captionFiles', {}).get('edges', []):
 279             node = edge.get('node', {})
 280             n_url = node.get('publicLocation')
 281             if not n_url:
 282                 continue
 283             subtitles.setdefault('de', []).append({
 284                 'url': n_url,
 285             })
 286
 287         thumbnails = []
 288         for edge in clip.get('teaserImages', {}).get('edges', []):
 289             for image_edge in edge.get('node', {}).get('imageFiles', {}).get('edges', []):
 290                 node = image_edge.get('node', {})
 291                 n_url = node.get('publicLocation')
 292                 if not n_url:
 293                     continue
 294                 thumbnails.append({
 295                     'url': n_url,
 296                     'width': int_or_none(node.get('width')),
 297                     'height': int_or_none(node.get('height')),
 298                 })
 299
 300         return {
 301             'id': clip_id,
 302             'title': title,
 303             'description': clip.get('description'),
 304             'duration': int_or_none(clip.get('duration')),
 305             'timestamp': parse_iso8601(clip.get('createdAt')),
 306             'age_limit': int_or_none(clip.get('ageRestriction')),
 307             'formats': formats,
 308             'subtitles': subtitles,
 309             'thumbnails': thumbnails,
 310         }