yt_dlp/extractor/adobetv.py

   1 from __future__ import unicode_literals
   2
   3 import functools
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     float_or_none,
  10     int_or_none,
  11     ISO639Utils,
  12     join_nonempty,
  13     OnDemandPagedList,
  14     parse_duration,
  15     str_or_none,
  16     str_to_int,
  17     unified_strdate,
  18 )
  19
  20
  21 class AdobeTVBaseIE(InfoExtractor):
  22     def _call_api(self, path, video_id, query, note=None):
  23         return self._download_json(
  24             'http://tv.adobe.com/api/v4/' + path,
  25             video_id, note, query=query)['data']
  26
  27     def _parse_subtitles(self, video_data, url_key):
  28         subtitles = {}
  29         for translation in video_data.get('translations', []):
  30             vtt_path = translation.get(url_key)
  31             if not vtt_path:
  32                 continue
  33             lang = translation.get('language_w3c') or ISO639Utils.long2short(translation['language_medium'])
  34             subtitles.setdefault(lang, []).append({
  35                 'ext': 'vtt',
  36                 'url': vtt_path,
  37             })
  38         return subtitles
  39
  40     def _parse_video_data(self, video_data):
  41         video_id = compat_str(video_data['id'])
  42         title = video_data['title']
  43
  44         s3_extracted = False
  45         formats = []
  46         for source in video_data.get('videos', []):
  47             source_url = source.get('url')
  48             if not source_url:
  49                 continue
  50             f = {
  51                 'format_id': source.get('quality_level'),
  52                 'fps': int_or_none(source.get('frame_rate')),
  53                 'height': int_or_none(source.get('height')),
  54                 'tbr': int_or_none(source.get('video_data_rate')),
  55                 'width': int_or_none(source.get('width')),
  56                 'url': source_url,
  57             }
  58             original_filename = source.get('original_filename')
  59             if original_filename:
  60                 if not (f.get('height') and f.get('width')):
  61                     mobj = re.search(r'_(\d+)x(\d+)', original_filename)
  62                     if mobj:
  63                         f.update({
  64                             'height': int(mobj.group(2)),
  65                             'width': int(mobj.group(1)),
  66                         })
  67                 if original_filename.startswith('s3://') and not s3_extracted:
  68                     formats.append({
  69                         'format_id': 'original',
  70                         'quality': 1,
  71                         'url': original_filename.replace('s3://', 'https://s3.amazonaws.com/'),
  72                     })
  73                     s3_extracted = True
  74             formats.append(f)
  75         self._sort_formats(formats)
  76
  77         return {
  78             'id': video_id,
  79             'title': title,
  80             'description': video_data.get('description'),
  81             'thumbnail': video_data.get('thumbnail'),
  82             'upload_date': unified_strdate(video_data.get('start_date')),
  83             'duration': parse_duration(video_data.get('duration')),
  84             'view_count': str_to_int(video_data.get('playcount')),
  85             'formats': formats,
  86             'subtitles': self._parse_subtitles(video_data, 'vtt'),
  87         }
  88
  89
  90 class AdobeTVEmbedIE(AdobeTVBaseIE):
  91     IE_NAME = 'adobetv:embed'
  92     _VALID_URL = r'https?://tv\.adobe\.com/embed/\d+/(?P<id>\d+)'
  93     _TEST = {
  94         'url': 'https://tv.adobe.com/embed/22/4153',
  95         'md5': 'c8c0461bf04d54574fc2b4d07ac6783a',
  96         'info_dict': {
  97             'id': '4153',
  98             'ext': 'flv',
  99             'title': 'Creating Graphics Optimized for BlackBerry',
 100             'description': 'md5:eac6e8dced38bdaae51cd94447927459',
 101             'thumbnail': r're:https?://.*\.jpg$',
 102             'upload_date': '20091109',
 103             'duration': 377,
 104             'view_count': int,
 105         },
 106     }
 107
 108     def _real_extract(self, url):
 109         video_id = self._match_id(url)
 110
 111         video_data = self._call_api(
 112             'episode/' + video_id, video_id, {'disclosure': 'standard'})[0]
 113         return self._parse_video_data(video_data)
 114
 115
 116 class AdobeTVIE(AdobeTVBaseIE):
 117     IE_NAME = 'adobetv'
 118     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?watch/(?P<show_urlname>[^/]+)/(?P<id>[^/]+)'
 119
 120     _TEST = {
 121         'url': 'http://tv.adobe.com/watch/the-complete-picture-with-julieanne-kost/quick-tip-how-to-draw-a-circle-around-an-object-in-photoshop/',
 122         'md5': '9bc5727bcdd55251f35ad311ca74fa1e',
 123         'info_dict': {
 124             'id': '10981',
 125             'ext': 'mp4',
 126             'title': 'Quick Tip - How to Draw a Circle Around an Object in Photoshop',
 127             'description': 'md5:99ec318dc909d7ba2a1f2b038f7d2311',
 128             'thumbnail': r're:https?://.*\.jpg$',
 129             'upload_date': '20110914',
 130             'duration': 60,
 131             'view_count': int,
 132         },
 133     }
 134
 135     def _real_extract(self, url):
 136         language, show_urlname, urlname = self._match_valid_url(url).groups()
 137         if not language:
 138             language = 'en'
 139
 140         video_data = self._call_api(
 141             'episode/get', urlname, {
 142                 'disclosure': 'standard',
 143                 'language': language,
 144                 'show_urlname': show_urlname,
 145                 'urlname': urlname,
 146             })[0]
 147         return self._parse_video_data(video_data)
 148
 149
 150 class AdobeTVPlaylistBaseIE(AdobeTVBaseIE):
 151     _PAGE_SIZE = 25
 152
 153     def _fetch_page(self, display_id, query, page):
 154         page += 1
 155         query['page'] = page
 156         for element_data in self._call_api(
 157                 self._RESOURCE, display_id, query, 'Download Page %d' % page):
 158             yield self._process_data(element_data)
 159
 160     def _extract_playlist_entries(self, display_id, query):
 161         return OnDemandPagedList(functools.partial(
 162             self._fetch_page, display_id, query), self._PAGE_SIZE)
 163
 164
 165 class AdobeTVShowIE(AdobeTVPlaylistBaseIE):
 166     IE_NAME = 'adobetv:show'
 167     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?show/(?P<id>[^/]+)'
 168
 169     _TEST = {
 170         'url': 'http://tv.adobe.com/show/the-complete-picture-with-julieanne-kost',
 171         'info_dict': {
 172             'id': '36',
 173             'title': 'The Complete Picture with Julieanne Kost',
 174             'description': 'md5:fa50867102dcd1aa0ddf2ab039311b27',
 175         },
 176         'playlist_mincount': 136,
 177     }
 178     _RESOURCE = 'episode'
 179     _process_data = AdobeTVBaseIE._parse_video_data
 180
 181     def _real_extract(self, url):
 182         language, show_urlname = self._match_valid_url(url).groups()
 183         if not language:
 184             language = 'en'
 185         query = {
 186             'disclosure': 'standard',
 187             'language': language,
 188             'show_urlname': show_urlname,
 189         }
 190
 191         show_data = self._call_api(
 192             'show/get', show_urlname, query)[0]
 193
 194         return self.playlist_result(
 195             self._extract_playlist_entries(show_urlname, query),
 196             str_or_none(show_data.get('id')),
 197             show_data.get('show_name'),
 198             show_data.get('show_description'))
 199
 200
 201 class AdobeTVChannelIE(AdobeTVPlaylistBaseIE):
 202     IE_NAME = 'adobetv:channel'
 203     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?channel/(?P<id>[^/]+)(?:/(?P<category_urlname>[^/]+))?'
 204
 205     _TEST = {
 206         'url': 'http://tv.adobe.com/channel/development',
 207         'info_dict': {
 208             'id': 'development',
 209         },
 210         'playlist_mincount': 96,
 211     }
 212     _RESOURCE = 'show'
 213
 214     def _process_data(self, show_data):
 215         return self.url_result(
 216             show_data['url'], 'AdobeTVShow', str_or_none(show_data.get('id')))
 217
 218     def _real_extract(self, url):
 219         language, channel_urlname, category_urlname = self._match_valid_url(url).groups()
 220         if not language:
 221             language = 'en'
 222         query = {
 223             'channel_urlname': channel_urlname,
 224             'language': language,
 225         }
 226         if category_urlname:
 227             query['category_urlname'] = category_urlname
 228
 229         return self.playlist_result(
 230             self._extract_playlist_entries(channel_urlname, query),
 231             channel_urlname)
 232
 233
 234 class AdobeTVVideoIE(AdobeTVBaseIE):
 235     IE_NAME = 'adobetv:video'
 236     _VALID_URL = r'https?://video\.tv\.adobe\.com/v/(?P<id>\d+)'
 237
 238     _TEST = {
 239         # From https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners
 240         'url': 'https://video.tv.adobe.com/v/2456/',
 241         'md5': '43662b577c018ad707a63766462b1e87',
 242         'info_dict': {
 243             'id': '2456',
 244             'ext': 'mp4',
 245             'title': 'New experience with Acrobat DC',
 246             'description': 'New experience with Acrobat DC',
 247             'duration': 248.667,
 248         },
 249     }
 250
 251     def _real_extract(self, url):
 252         video_id = self._match_id(url)
 253         webpage = self._download_webpage(url, video_id)
 254
 255         video_data = self._parse_json(self._search_regex(
 256             r'var\s+bridge\s*=\s*([^;]+);', webpage, 'bridged data'), video_id)
 257         title = video_data['title']
 258
 259         formats = []
 260         sources = video_data.get('sources') or []
 261         for source in sources:
 262             source_src = source.get('src')
 263             if not source_src:
 264                 continue
 265             formats.append({
 266                 'filesize': int_or_none(source.get('kilobytes') or None, invscale=1000),
 267                 'format_id': join_nonempty(source.get('format'), source.get('label')),
 268                 'height': int_or_none(source.get('height') or None),
 269                 'tbr': int_or_none(source.get('bitrate') or None),
 270                 'width': int_or_none(source.get('width') or None),
 271                 'url': source_src,
 272             })
 273         self._sort_formats(formats)
 274
 275         # For both metadata and downloaded files the duration varies among
 276         # formats. I just pick the max one
 277         duration = max(filter(None, [
 278             float_or_none(source.get('duration'), scale=1000)
 279             for source in sources]))
 280
 281         return {
 282             'id': video_id,
 283             'formats': formats,
 284             'title': title,
 285             'description': video_data.get('description'),
 286             'thumbnail': video_data.get('video', {}).get('poster'),
 287             'duration': duration,
 288             'subtitles': self._parse_subtitles(video_data, 'vttPath'),
 289         }