yt_dlp/extractor/microsoftembed.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     int_or_none,
   6     parse_iso8601,
   7     traverse_obj,
   8     unified_timestamp,
   9     url_basename,
  10     url_or_none,
  11 )
  12
  13
  14 class MicrosoftEmbedIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?microsoft\.com/(?:[^/]+/)?videoplayer/embed/(?P<id>[a-z0-9A-Z]+)'
  16
  17     _TESTS = [{
  18         'url': 'https://www.microsoft.com/en-us/videoplayer/embed/RWL07e',
  19         'md5': 'eb0ae9007f9b305f9acd0a03e74cb1a9',
  20         'info_dict': {
  21             'id': 'RWL07e',
  22             'title': 'Microsoft for Public Health and Social Services',
  23             'ext': 'mp4',
  24             'thumbnail': 'http://img-prod-cms-rt-microsoft-com.akamaized.net/cms/api/am/imageFileData/RWL7Ju?ver=cae5',
  25             'age_limit': 0,
  26             'timestamp': 1631658316,
  27             'upload_date': '20210914',
  28         },
  29         'expected_warnings': ['Failed to parse XML: syntax error: line 1, column 0'],
  30     }]
  31     _API_URL = 'https://prod-video-cms-rt-microsoft-com.akamaized.net/vhs/api/videos/'
  32
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35         metadata = self._download_json(self._API_URL + video_id, video_id)
  36
  37         formats = []
  38         for source_type, source in metadata['streams'].items():
  39             if source_type == 'smooth_Streaming':
  40                 formats.extend(self._extract_ism_formats(source['url'], video_id, 'mss', fatal=False))
  41             elif source_type == 'apple_HTTP_Live_Streaming':
  42                 formats.extend(self._extract_m3u8_formats(source['url'], video_id, 'mp4', fatal=False))
  43             elif source_type == 'mPEG_DASH':
  44                 formats.extend(self._extract_mpd_formats(source['url'], video_id, fatal=False))
  45             else:
  46                 formats.append({
  47                     'format_id': source_type,
  48                     'url': source['url'],
  49                     'height': source.get('heightPixels'),
  50                     'width': source.get('widthPixels'),
  51                 })
  52
  53         subtitles = {
  54             lang: [{
  55                 'url': data.get('url'),
  56                 'ext': 'vtt',
  57             }] for lang, data in traverse_obj(metadata, 'captions', default={}).items()
  58         }
  59
  60         thumbnails = [{
  61             'url': thumb.get('url'),
  62             'width': thumb.get('width') or None,
  63             'height': thumb.get('height') or None,
  64         } for thumb in traverse_obj(metadata, ('snippet', 'thumbnails', ...))]
  65         self._remove_duplicate_formats(thumbnails)
  66
  67         return {
  68             'id': video_id,
  69             'title': traverse_obj(metadata, ('snippet', 'title')),
  70             'timestamp': unified_timestamp(traverse_obj(metadata, ('snippet', 'activeStartDate'))),
  71             'age_limit': int_or_none(traverse_obj(metadata, ('snippet', 'minimumAge'))) or 0,
  72             'formats': formats,
  73             'subtitles': subtitles,
  74             'thumbnails': thumbnails,
  75         }
  76
  77
  78 class MicrosoftMediusBaseIE(InfoExtractor):
  79     @staticmethod
  80     def _sub_to_dict(subtitle_list):
  81         subtitles = {}
  82         for sub in subtitle_list:
  83             subtitles.setdefault(sub.pop('tag', 'und'), []).append(sub)
  84         return subtitles
  85
  86     def _extract_ism(self, ism_url, video_id):
  87         formats = self._extract_ism_formats(ism_url, video_id)
  88         for fmt in formats:
  89             if fmt['language'] != 'eng' and 'English' not in fmt['format_id']:
  90                 fmt['language_preference'] = -10
  91         return formats
  92
  93
  94 class MicrosoftMediusIE(MicrosoftMediusBaseIE):
  95     _VALID_URL = r'https?://medius\.microsoft\.com/Embed/(?:Video\?id=|video-nc/|VideoDetails/)(?P<id>[\da-f-]+)'
  96
  97     _TESTS = [{
  98         'url': 'https://medius.microsoft.com/Embed/video-nc/9640d86c-f513-4889-959e-5dace86e7d2b',
  99         'info_dict': {
 100             'id': '9640d86c-f513-4889-959e-5dace86e7d2b',
 101             'ext': 'ismv',
 102             'title': 'Rapidly code, test and ship from secure cloud developer environments',
 103             'description': 'md5:33c8e4facadc438613476eea24165f71',
 104             'thumbnail': r're:https://mediusimg\.event\.microsoft\.com/video-\d+/thumbnail\.jpg.*',
 105             'subtitles': 'count:30',
 106         },
 107     }, {
 108         'url': 'https://medius.microsoft.com/Embed/video-nc/81215af5-c813-4dcd-aede-94f4e1a7daa3',
 109         'info_dict': {
 110             'id': '81215af5-c813-4dcd-aede-94f4e1a7daa3',
 111             'ext': 'ismv',
 112             'title': 'Microsoft Build opening',
 113             'description': 'md5:43455096141077a1f23144cab8cec1cb',
 114             'thumbnail': r're:https://mediusimg\.event\.microsoft\.com/video-\d+/thumbnail\.jpg.*',
 115             'subtitles': 'count:31',
 116         },
 117     }, {
 118         'url': 'https://medius.microsoft.com/Embed/VideoDetails/78493569-9b3b-4a85-a409-ee76e789e25c',
 119         'info_dict': {
 120             'id': '78493569-9b3b-4a85-a409-ee76e789e25c',
 121             'ext': 'ismv',
 122             'title': ' Anomaly Detection & Root cause at Edge',
 123             'description': 'md5:f8f1ad93d7918649bfb97fa081b03b83',
 124             'thumbnail': r're:https://mediusdownload.event.microsoft.com/asset.*\.jpg.*',
 125             'subtitles': 'count:17',
 126         },
 127     }, {
 128         'url': 'https://medius.microsoft.com/Embed/Video?id=0dc69bda-079b-4070-a7db-a8da1a06a9c7',
 129         'only_matching': True,
 130     }, {
 131         'url': 'https://medius.microsoft.com/Embed/video-nc/fe823a91-959c-465b-96d4-8f4db624f72c',
 132         'only_matching': True,
 133     }]
 134
 135     def _extract_subtitle(self, webpage, video_id):
 136         captions = traverse_obj(
 137             self._search_json(r'const\s+captionsConfiguration\s*=', webpage, 'captions', video_id, default=None),
 138             ('languageList', lambda _, v: url_or_none(v['src']), {
 139                 'url': 'src',
 140                 'tag': ('srclang', {str}),
 141                 'name': ('kind', {str}),
 142             })) or [{'url': url, 'tag': url_basename(url).split('.vtt')[0].split('_')[-1]}
 143                     for url in re.findall(r'var\s+file\s+=\s+\{[^}]+\'(https://[^\']+\.vtt\?[^\']+)', webpage)]
 144
 145         return self._sub_to_dict(captions)
 146
 147     def _real_extract(self, url):
 148         video_id = self._match_id(url)
 149         webpage = self._download_webpage(f'https://medius.microsoft.com/Embed/video-nc/{video_id}', video_id)
 150
 151         return {
 152             'id': video_id,
 153             'title': self._og_search_title(webpage),
 154             'description': self._og_search_description(webpage),
 155             'formats': self._extract_ism(
 156                 self._search_regex(r'StreamUrl\s*=\s*"([^"]+manifest)"', webpage, 'ism url'), video_id),
 157             'thumbnail': self._og_search_thumbnail(webpage),
 158             'subtitles': self._extract_subtitle(webpage, video_id),
 159         }
 160
 161
 162 class MicrosoftLearnPlaylistIE(InfoExtractor):
 163     _VALID_URL = r'https?://learn\.microsoft\.com/(?:[\w-]+/)?(?P<type>shows|events)/(?P<id>[\w-]+)/?(?:[?#]|$)'
 164     _TESTS = [{
 165         'url': 'https://learn.microsoft.com/en-us/shows/bash-for-beginners',
 166         'info_dict': {
 167             'id': 'bash-for-beginners',
 168             'title': 'Bash for Beginners',
 169             'description': 'md5:16a91c07222117d1e00912f0dbc02c2c',
 170         },
 171         'playlist_count': 20,
 172     }, {
 173         'url': 'https://learn.microsoft.com/en-us/events/build-2022',
 174         'info_dict': {
 175             'id': 'build-2022',
 176             'title': 'Microsoft Build 2022 - Events',
 177             'description': 'md5:c16b43848027df837b22c6fbac7648d3',
 178         },
 179         'playlist_count': 201,
 180     }]
 181
 182     def _entries(self, url_base, video_id):
 183         skip = 0
 184         while True:
 185             playlist_info = self._download_json(url_base, video_id, f'Downloading entries {skip}', query={
 186                 'locale': 'en-us',
 187                 '$skip': skip,
 188             })
 189             url_paths = traverse_obj(playlist_info, ('results', ..., 'url', {str}))
 190             for url_path in url_paths:
 191                 yield self.url_result(f'https://learn.microsoft.com/en-us{url_path}')
 192             skip += len(url_paths)
 193             if skip >= playlist_info.get('count', 0) or not url_paths:
 194                 break
 195
 196     def _real_extract(self, url):
 197         playlist_id, playlist_type = self._match_valid_url(url).group('id', 'type')
 198         webpage = self._download_webpage(url, playlist_id)
 199
 200         metainfo = {
 201             'title': self._og_search_title(webpage),
 202             'description': self._og_search_description(webpage),
 203         }
 204         sub_type = 'episodes' if playlist_type == 'shows' else 'sessions'
 205
 206         url_base = f'https://learn.microsoft.com/api/contentbrowser/search/{playlist_type}/{playlist_id}/{sub_type}'
 207         return self.playlist_result(self._entries(url_base, playlist_id), playlist_id, **metainfo)
 208
 209
 210 class MicrosoftLearnEpisodeIE(MicrosoftMediusBaseIE):
 211     _VALID_URL = r'https?://learn\.microsoft\.com/(?:[\w-]+/)?shows/[\w-]+/(?P<id>[^?#/]+)'
 212     _TESTS = [{
 213         'url': 'https://learn.microsoft.com/en-us/shows/bash-for-beginners/what-is-the-difference-between-a-terminal-and-a-shell-2-of-20-bash-for-beginners/',
 214         'info_dict': {
 215             'id': 'd44e1a03-a0e5-45c2-9496-5c9fa08dc94c',
 216             'ext': 'ismv',
 217             'title': 'What is the Difference Between a Terminal and a Shell? (Part 2 of 20)',
 218             'description': 'md5:7bbbfb593d21c2cf2babc3715ade6b88',
 219             'timestamp': 1676339547,
 220             'upload_date': '20230214',
 221             'thumbnail': r're:https://learn\.microsoft\.com/video/media/.*\.png',
 222             'subtitles': 'count:14',
 223         },
 224     }]
 225
 226     def _real_extract(self, url):
 227         video_id = self._match_id(url)
 228         webpage = self._download_webpage(url, video_id)
 229
 230         entry_id = self._html_search_meta('entryId', webpage, 'entryId', fatal=True)
 231         video_info = self._download_json(
 232             f'https://learn.microsoft.com/api/video/public/v1/entries/{entry_id}', video_id)
 233         return {
 234             'id': entry_id,
 235             'formats': self._extract_ism(video_info['publicVideo']['adaptiveVideoUrl'], video_id),
 236             'subtitles': self._sub_to_dict(traverse_obj(video_info, (
 237                 'publicVideo', 'captions', lambda _, v: url_or_none(v['url']), {
 238                     'tag': ('language', {str}),
 239                     'url': 'url',
 240                 }))),
 241             'title': self._og_search_title(webpage),
 242             'description': self._og_search_description(webpage),
 243             **traverse_obj(video_info, {
 244                 'timestamp': ('createTime', {parse_iso8601}),
 245                 'thumbnails': ('publicVideo', 'thumbnailOtherSizes', ..., {'url': {url_or_none}}),
 246             }),
 247         }
 248
 249
 250 class MicrosoftLearnSessionIE(InfoExtractor):
 251     _VALID_URL = r'https?://learn\.microsoft\.com/(?:[\w-]+/)?events/[\w-]+/(?P<id>[^?#/]+)'
 252     _TESTS = [{
 253         'url': 'https://learn.microsoft.com/en-us/events/build-2022/ts01-rapidly-code-test-ship-from-secure-cloud-developer-environments',
 254         'info_dict': {
 255             'id': '9640d86c-f513-4889-959e-5dace86e7d2b',
 256             'ext': 'ismv',
 257             'title': 'Rapidly code, test and ship from secure cloud developer environments - Events',
 258             'description': 'md5:f26c1a85d41c1cffd27a0279254a25c3',
 259             'timestamp': 1653408600,
 260             'upload_date': '20220524',
 261             'thumbnail': r're:https://mediusimg\.event\.microsoft\.com/video-\d+/thumbnail\.jpg.*',
 262         },
 263     }]
 264
 265     def _real_extract(self, url):
 266         video_id = self._match_id(url)
 267         webpage = self._download_webpage(url, video_id)
 268
 269         metainfo = {
 270             'title': self._og_search_title(webpage),
 271             'description': self._og_search_description(webpage),
 272             'timestamp': parse_iso8601(self._html_search_meta('startDate', webpage, 'startDate')),
 273         }
 274
 275         return self.url_result(
 276             self._html_search_meta('externalVideoUrl', webpage, 'videoUrl', fatal=True),
 277             url_transparent=True, ie=MicrosoftMediusIE, **metainfo)
 278
 279
 280 class MicrosoftBuildIE(InfoExtractor):
 281     _VALID_URL = [
 282         r'https?://build\.microsoft\.com/[\w-]+/sessions/(?P<id>[\da-f-]+)',
 283         r'https?://build\.microsoft\.com/[\w-]+/(?P<id>sessions)/?(?:[?#]|$)',
 284     ]
 285
 286     _TESTS = [{
 287         'url': 'https://build.microsoft.com/en-US/sessions/b49feb31-afcd-4217-a538-d3ca1d171198?source=sessions',
 288         'info_dict': {
 289             'id': 'aee55fb5-fcf9-4b38-b764-a3527cb57554',
 290             'ext': 'ismv',
 291             'title': 'Microsoft Build opening keynote',
 292             'description': 'md5:d38338f336ef4b6ef9ad2a7466a76655',
 293             'timestamp': 1716307200,
 294             'upload_date': '20240521',
 295             'thumbnail': r're:https://mediusimg\.event\.microsoft\.com/video-\d+/thumbnail\.jpg.*',
 296         },
 297     }, {
 298         'url': 'https://build.microsoft.com/en-US/sessions',
 299         'info_dict': {
 300             'id': 'sessions',
 301         },
 302         'playlist_mincount': 418,
 303     }]
 304
 305     def _real_extract(self, url):
 306         video_id = self._match_id(url)
 307
 308         entries = [
 309             self.url_result(
 310                 video_info['onDemand'], ie=MicrosoftMediusIE, url_transparent=True, **traverse_obj(video_info, {
 311                     'id': ('sessionId', {str}),
 312                     'title': ('title', {str}),
 313                     'description': ('description', {str}),
 314                     'timestamp': ('startDateTime', {parse_iso8601}),
 315                 }))
 316             for video_info in self._download_json(
 317                 'https://api-v2.build.microsoft.com/api/session/all/en-US', video_id, 'Downloading video info')
 318         ]
 319         if video_id == 'sessions':
 320             return self.playlist_result(entries, video_id)
 321         else:
 322             return traverse_obj(entries, (lambda _, v: v['id'] == video_id), get_all=False)