yt_dlp/extractor/adobetv.py

   1 import functools
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ISO639Utils,
   7     OnDemandPagedList,
   8     float_or_none,
   9     int_or_none,
  10     join_nonempty,
  11     parse_duration,
  12     str_or_none,
  13     str_to_int,
  14     unified_strdate,
  15 )
  16
  17
  18 class AdobeTVBaseIE(InfoExtractor):
  19     def _call_api(self, path, video_id, query, note=None):
  20         return self._download_json(
  21             'http://tv.adobe.com/api/v4/' + path,
  22             video_id, note, query=query)['data']
  23
  24     def _parse_subtitles(self, video_data, url_key):
  25         subtitles = {}
  26         for translation in video_data.get('translations', []):
  27             vtt_path = translation.get(url_key)
  28             if not vtt_path:
  29                 continue
  30             lang = translation.get('language_w3c') or ISO639Utils.long2short(translation['language_medium'])
  31             subtitles.setdefault(lang, []).append({
  32                 'ext': 'vtt',
  33                 'url': vtt_path,
  34             })
  35         return subtitles
  36
  37     def _parse_video_data(self, video_data):
  38         video_id = str(video_data['id'])
  39         title = video_data['title']
  40
  41         s3_extracted = False
  42         formats = []
  43         for source in video_data.get('videos', []):
  44             source_url = source.get('url')
  45             if not source_url:
  46                 continue
  47             f = {
  48                 'format_id': source.get('quality_level'),
  49                 'fps': int_or_none(source.get('frame_rate')),
  50                 'height': int_or_none(source.get('height')),
  51                 'tbr': int_or_none(source.get('video_data_rate')),
  52                 'width': int_or_none(source.get('width')),
  53                 'url': source_url,
  54             }
  55             original_filename = source.get('original_filename')
  56             if original_filename:
  57                 if not (f.get('height') and f.get('width')):
  58                     mobj = re.search(r'_(\d+)x(\d+)', original_filename)
  59                     if mobj:
  60                         f.update({
  61                             'height': int(mobj.group(2)),
  62                             'width': int(mobj.group(1)),
  63                         })
  64                 if original_filename.startswith('s3://') and not s3_extracted:
  65                     formats.append({
  66                         'format_id': 'original',
  67                         'quality': 1,
  68                         'url': original_filename.replace('s3://', 'https://s3.amazonaws.com/'),
  69                     })
  70                     s3_extracted = True
  71             formats.append(f)
  72
  73         return {
  74             'id': video_id,
  75             'title': title,
  76             'description': video_data.get('description'),
  77             'thumbnail': video_data.get('thumbnail'),
  78             'upload_date': unified_strdate(video_data.get('start_date')),
  79             'duration': parse_duration(video_data.get('duration')),
  80             'view_count': str_to_int(video_data.get('playcount')),
  81             'formats': formats,
  82             'subtitles': self._parse_subtitles(video_data, 'vtt'),
  83         }
  84
  85
  86 class AdobeTVEmbedIE(AdobeTVBaseIE):
  87     IE_NAME = 'adobetv:embed'
  88     _VALID_URL = r'https?://tv\.adobe\.com/embed/\d+/(?P<id>\d+)'
  89     _TEST = {
  90         'url': 'https://tv.adobe.com/embed/22/4153',
  91         'md5': 'c8c0461bf04d54574fc2b4d07ac6783a',
  92         'info_dict': {
  93             'id': '4153',
  94             'ext': 'flv',
  95             'title': 'Creating Graphics Optimized for BlackBerry',
  96             'description': 'md5:eac6e8dced38bdaae51cd94447927459',
  97             'thumbnail': r're:https?://.*\.jpg$',
  98             'upload_date': '20091109',
  99             'duration': 377,
 100             'view_count': int,
 101         },
 102     }
 103
 104     def _real_extract(self, url):
 105         video_id = self._match_id(url)
 106
 107         video_data = self._call_api(
 108             'episode/' + video_id, video_id, {'disclosure': 'standard'})[0]
 109         return self._parse_video_data(video_data)
 110
 111
 112 class AdobeTVIE(AdobeTVBaseIE):
 113     IE_NAME = 'adobetv'
 114     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?watch/(?P<show_urlname>[^/]+)/(?P<id>[^/]+)'
 115
 116     _TEST = {
 117         'url': 'http://tv.adobe.com/watch/the-complete-picture-with-julieanne-kost/quick-tip-how-to-draw-a-circle-around-an-object-in-photoshop/',
 118         'md5': '9bc5727bcdd55251f35ad311ca74fa1e',
 119         'info_dict': {
 120             'id': '10981',
 121             'ext': 'mp4',
 122             'title': 'Quick Tip - How to Draw a Circle Around an Object in Photoshop',
 123             'description': 'md5:99ec318dc909d7ba2a1f2b038f7d2311',
 124             'thumbnail': r're:https?://.*\.jpg$',
 125             'upload_date': '20110914',
 126             'duration': 60,
 127             'view_count': int,
 128         },
 129     }
 130
 131     def _real_extract(self, url):
 132         language, show_urlname, urlname = self._match_valid_url(url).groups()
 133         if not language:
 134             language = 'en'
 135
 136         video_data = self._call_api(
 137             'episode/get', urlname, {
 138                 'disclosure': 'standard',
 139                 'language': language,
 140                 'show_urlname': show_urlname,
 141                 'urlname': urlname,
 142             })[0]
 143         return self._parse_video_data(video_data)
 144
 145
 146 class AdobeTVPlaylistBaseIE(AdobeTVBaseIE):
 147     _PAGE_SIZE = 25
 148
 149     def _fetch_page(self, display_id, query, page):
 150         page += 1
 151         query['page'] = page
 152         for element_data in self._call_api(
 153                 self._RESOURCE, display_id, query, f'Download Page {page}'):
 154             yield self._process_data(element_data)
 155
 156     def _extract_playlist_entries(self, display_id, query):
 157         return OnDemandPagedList(functools.partial(
 158             self._fetch_page, display_id, query), self._PAGE_SIZE)
 159
 160
 161 class AdobeTVShowIE(AdobeTVPlaylistBaseIE):
 162     IE_NAME = 'adobetv:show'
 163     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?show/(?P<id>[^/]+)'
 164
 165     _TEST = {
 166         'url': 'http://tv.adobe.com/show/the-complete-picture-with-julieanne-kost',
 167         'info_dict': {
 168             'id': '36',
 169             'title': 'The Complete Picture with Julieanne Kost',
 170             'description': 'md5:fa50867102dcd1aa0ddf2ab039311b27',
 171         },
 172         'playlist_mincount': 136,
 173     }
 174     _RESOURCE = 'episode'
 175     _process_data = AdobeTVBaseIE._parse_video_data
 176
 177     def _real_extract(self, url):
 178         language, show_urlname = self._match_valid_url(url).groups()
 179         if not language:
 180             language = 'en'
 181         query = {
 182             'disclosure': 'standard',
 183             'language': language,
 184             'show_urlname': show_urlname,
 185         }
 186
 187         show_data = self._call_api(
 188             'show/get', show_urlname, query)[0]
 189
 190         return self.playlist_result(
 191             self._extract_playlist_entries(show_urlname, query),
 192             str_or_none(show_data.get('id')),
 193             show_data.get('show_name'),
 194             show_data.get('show_description'))
 195
 196
 197 class AdobeTVChannelIE(AdobeTVPlaylistBaseIE):
 198     IE_NAME = 'adobetv:channel'
 199     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?channel/(?P<id>[^/]+)(?:/(?P<category_urlname>[^/]+))?'
 200
 201     _TEST = {
 202         'url': 'http://tv.adobe.com/channel/development',
 203         'info_dict': {
 204             'id': 'development',
 205         },
 206         'playlist_mincount': 96,
 207     }
 208     _RESOURCE = 'show'
 209
 210     def _process_data(self, show_data):
 211         return self.url_result(
 212             show_data['url'], 'AdobeTVShow', str_or_none(show_data.get('id')))
 213
 214     def _real_extract(self, url):
 215         language, channel_urlname, category_urlname = self._match_valid_url(url).groups()
 216         if not language:
 217             language = 'en'
 218         query = {
 219             'channel_urlname': channel_urlname,
 220             'language': language,
 221         }
 222         if category_urlname:
 223             query['category_urlname'] = category_urlname
 224
 225         return self.playlist_result(
 226             self._extract_playlist_entries(channel_urlname, query),
 227             channel_urlname)
 228
 229
 230 class AdobeTVVideoIE(AdobeTVBaseIE):
 231     IE_NAME = 'adobetv:video'
 232     _VALID_URL = r'https?://video\.tv\.adobe\.com/v/(?P<id>\d+)'
 233     _EMBED_REGEX = [r'<iframe[^>]+src=[\'"](?P<url>(?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]']
 234
 235     _TEST = {
 236         # From https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners
 237         'url': 'https://video.tv.adobe.com/v/2456/',
 238         'md5': '43662b577c018ad707a63766462b1e87',
 239         'info_dict': {
 240             'id': '2456',
 241             'ext': 'mp4',
 242             'title': 'New experience with Acrobat DC',
 243             'description': 'New experience with Acrobat DC',
 244             'duration': 248.667,
 245         },
 246     }
 247
 248     def _real_extract(self, url):
 249         video_id = self._match_id(url)
 250         webpage = self._download_webpage(url, video_id)
 251
 252         video_data = self._parse_json(self._search_regex(
 253             r'var\s+bridge\s*=\s*([^;]+);', webpage, 'bridged data'), video_id)
 254         title = video_data['title']
 255
 256         formats = []
 257         sources = video_data.get('sources') or []
 258         for source in sources:
 259             source_src = source.get('src')
 260             if not source_src:
 261                 continue
 262             formats.append({
 263                 'filesize': int_or_none(source.get('kilobytes') or None, invscale=1000),
 264                 'format_id': join_nonempty(source.get('format'), source.get('label')),
 265                 'height': int_or_none(source.get('height') or None),
 266                 'tbr': int_or_none(source.get('bitrate') or None),
 267                 'width': int_or_none(source.get('width') or None),
 268                 'url': source_src,
 269             })
 270
 271         # For both metadata and downloaded files the duration varies among
 272         # formats. I just pick the max one
 273         duration = max(filter(None, [
 274             float_or_none(source.get('duration'), scale=1000)
 275             for source in sources]))
 276
 277         return {
 278             'id': video_id,
 279             'formats': formats,
 280             'title': title,
 281             'description': video_data.get('description'),
 282             'thumbnail': video_data.get('video', {}).get('poster'),
 283             'duration': duration,
 284             'subtitles': self._parse_subtitles(video_data, 'vttPath'),
 285         }