yt_dlp/extractor/channel9.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     clean_html,
   8     int_or_none,
   9     parse_iso8601,
  10     qualities,
  11     unescapeHTML,
  12 )
  13
  14
  15 class Channel9IE(InfoExtractor):
  16     IE_DESC = 'Channel 9'
  17     IE_NAME = 'channel9'
  18     _VALID_URL = r'https?://(?:www\.)?(?:channel9\.msdn\.com|s\.ch9\.ms)/(?P<contentpath>.+?)(?P<rss>/RSS)?/?(?:[?#&]|$)'
  19
  20     _TESTS = [{
  21         'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
  22         'md5': '32083d4eaf1946db6d454313f44510ca',
  23         'info_dict': {
  24             'id': '6c413323-383a-49dc-88f9-a22800cab024',
  25             'ext': 'wmv',
  26             'title': 'Developer Kick-Off Session: Stuff We Love',
  27             'description': 'md5:b80bf9355a503c193aff7ec6cd5a7731',
  28             'duration': 4576,
  29             'thumbnail': r're:https?://.*\.jpg',
  30             'timestamp': 1377717420,
  31             'upload_date': '20130828',
  32             'session_code': 'KOS002',
  33             'session_room': 'Arena 1A',
  34             'session_speakers': 'count:5',
  35         },
  36     }, {
  37         'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
  38         'md5': 'dcf983ee6acd2088e7188c3cf79b46bc',
  39         'info_dict': {
  40             'id': 'fe8e435f-bb93-4e01-8e97-a28c01887024',
  41             'ext': 'wmv',
  42             'title': 'Self-service BI with Power BI - nuclear testing',
  43             'description': 'md5:2d17fec927fc91e9e17783b3ecc88f54',
  44             'duration': 1540,
  45             'thumbnail': r're:https?://.*\.jpg',
  46             'timestamp': 1386381991,
  47             'upload_date': '20131207',
  48             'authors': ['Mike Wilmot'],
  49         },
  50     }, {
  51         # low quality mp4 is best
  52         'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
  53         'info_dict': {
  54             'id': '33ad69d2-6a4e-4172-83a1-a523013dec76',
  55             'ext': 'mp4',
  56             'title': 'Ranges for the Standard Library',
  57             'description': 'md5:9895e0a9fd80822d2f01c454b8f4a372',
  58             'duration': 5646,
  59             'thumbnail': r're:https?://.*\.jpg',
  60             'upload_date': '20150930',
  61             'timestamp': 1443640735,
  62         },
  63         'params': {
  64             'skip_download': True,
  65         },
  66     }, {
  67         'url': 'https://channel9.msdn.com/Events/DEVintersection/DEVintersection-2016/RSS',
  68         'info_dict': {
  69             'id': 'Events/DEVintersection/DEVintersection-2016',
  70             'title': 'DEVintersection 2016 Orlando Sessions',
  71         },
  72         'playlist_mincount': 14,
  73     }, {
  74         'url': 'https://channel9.msdn.com/Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b/RSS',
  75         'only_matching': True,
  76     }, {
  77         'url': 'https://channel9.msdn.com/Events/Speakers/scott-hanselman/RSS?UrlSafeName=scott-hanselman',
  78         'only_matching': True,
  79     }]
  80
  81     _RSS_URL = 'http://channel9.msdn.com/%s/RSS'
  82
  83     @staticmethod
  84     def _extract_urls(webpage):
  85         return re.findall(
  86             r'<iframe[^>]+src=["\'](https?://channel9\.msdn\.com/(?:[^/]+/)+)player\b',
  87             webpage)
  88
  89     def _extract_list(self, video_id, rss_url=None):
  90         if not rss_url:
  91             rss_url = self._RSS_URL % video_id
  92         rss = self._download_xml(rss_url, video_id, 'Downloading RSS')
  93         entries = [self.url_result(session_url.text, 'Channel9')
  94                    for session_url in rss.findall('./channel/item/link')]
  95         title_text = rss.find('./channel/title').text
  96         return self.playlist_result(entries, video_id, title_text)
  97
  98     def _real_extract(self, url):
  99         content_path, rss = re.match(self._VALID_URL, url).groups()
 100
 101         if rss:
 102             return self._extract_list(content_path, url)
 103
 104         webpage = self._download_webpage(
 105             url, content_path, 'Downloading web page')
 106
 107         episode_data = self._search_regex(
 108             r"data-episode='([^']+)'", webpage, 'episode data', default=None)
 109         if episode_data:
 110             episode_data = self._parse_json(unescapeHTML(
 111                 episode_data), content_path)
 112             content_id = episode_data['contentId']
 113             is_session = '/Sessions(' in episode_data['api']
 114             content_url = 'https://channel9.msdn.com/odata' + episode_data['api'] + '?$select=Captions,CommentCount,MediaLengthInSeconds,PublishedDate,Rating,RatingCount,Title,VideoMP4High,VideoMP4Low,VideoMP4Medium,VideoPlayerPreviewImage,VideoWMV,VideoWMVHQ,Views,'
 115             if is_session:
 116                 content_url += 'Code,Description,Room,Slides,Speakers,ZipFile&$expand=Speakers'
 117             else:
 118                 content_url += 'Authors,Body&$expand=Authors'
 119             content_data = self._download_json(content_url, content_id)
 120             title = content_data['Title']
 121
 122             QUALITIES = (
 123                 'mp3',
 124                 'wmv', 'mp4',
 125                 'wmv-low', 'mp4-low',
 126                 'wmv-mid', 'mp4-mid',
 127                 'wmv-high', 'mp4-high',
 128             )
 129
 130             quality_key = qualities(QUALITIES)
 131
 132             def quality(quality_id, format_url):
 133                 return (len(QUALITIES) if '_Source.' in format_url
 134                         else quality_key(quality_id))
 135
 136             formats = []
 137             urls = set()
 138
 139             SITE_QUALITIES = {
 140                 'MP3': 'mp3',
 141                 'MP4': 'mp4',
 142                 'Low Quality WMV': 'wmv-low',
 143                 'Low Quality MP4': 'mp4-low',
 144                 'Mid Quality WMV': 'wmv-mid',
 145                 'Mid Quality MP4': 'mp4-mid',
 146                 'High Quality WMV': 'wmv-high',
 147                 'High Quality MP4': 'mp4-high',
 148             }
 149
 150             formats_select = self._search_regex(
 151                 r'(?s)<select[^>]+name=["\']format[^>]+>(.+?)</select', webpage,
 152                 'formats select', default=None)
 153             if formats_select:
 154                 for mobj in re.finditer(
 155                         r'<option\b[^>]+\bvalue=(["\'])(?P<url>(?:(?!\1).)+)\1[^>]*>\s*(?P<format>[^<]+?)\s*<',
 156                         formats_select):
 157                     format_url = mobj.group('url')
 158                     if format_url in urls:
 159                         continue
 160                     urls.add(format_url)
 161                     format_id = mobj.group('format')
 162                     quality_id = SITE_QUALITIES.get(format_id, format_id)
 163                     formats.append({
 164                         'url': format_url,
 165                         'format_id': quality_id,
 166                         'quality': quality(quality_id, format_url),
 167                         'vcodec': 'none' if quality_id == 'mp3' else None,
 168                     })
 169
 170             API_QUALITIES = {
 171                 'VideoMP4Low': 'mp4-low',
 172                 'VideoWMV': 'wmv-mid',
 173                 'VideoMP4Medium': 'mp4-mid',
 174                 'VideoMP4High': 'mp4-high',
 175                 'VideoWMVHQ': 'wmv-hq',
 176             }
 177
 178             for format_id, q in API_QUALITIES.items():
 179                 q_url = content_data.get(format_id)
 180                 if not q_url or q_url in urls:
 181                     continue
 182                 urls.add(q_url)
 183                 formats.append({
 184                     'url': q_url,
 185                     'format_id': q,
 186                     'quality': quality(q, q_url),
 187                 })
 188
 189             slides = content_data.get('Slides')
 190             zip_file = content_data.get('ZipFile')
 191
 192             if not formats and not slides and not zip_file:
 193                 self.raise_no_formats(
 194                     'None of recording, slides or zip are available for %s' % content_path)
 195             self._sort_formats(formats)
 196
 197             subtitles = {}
 198             for caption in content_data.get('Captions', []):
 199                 caption_url = caption.get('Url')
 200                 if not caption_url:
 201                     continue
 202                 subtitles.setdefault(caption.get('Language', 'en'), []).append({
 203                     'url': caption_url,
 204                     'ext': 'vtt',
 205                 })
 206
 207             common = {
 208                 'id': content_id,
 209                 'title': title,
 210                 'description': clean_html(content_data.get('Description') or content_data.get('Body')),
 211                 'thumbnail': content_data.get('VideoPlayerPreviewImage'),
 212                 'duration': int_or_none(content_data.get('MediaLengthInSeconds')),
 213                 'timestamp': parse_iso8601(content_data.get('PublishedDate')),
 214                 'avg_rating': int_or_none(content_data.get('Rating')),
 215                 'rating_count': int_or_none(content_data.get('RatingCount')),
 216                 'view_count': int_or_none(content_data.get('Views')),
 217                 'comment_count': int_or_none(content_data.get('CommentCount')),
 218                 'subtitles': subtitles,
 219             }
 220             if is_session:
 221                 speakers = []
 222                 for s in content_data.get('Speakers', []):
 223                     speaker_name = s.get('FullName')
 224                     if not speaker_name:
 225                         continue
 226                     speakers.append(speaker_name)
 227
 228                 common.update({
 229                     'session_code': content_data.get('Code'),
 230                     'session_room': content_data.get('Room'),
 231                     'session_speakers': speakers,
 232                 })
 233             else:
 234                 authors = []
 235                 for a in content_data.get('Authors', []):
 236                     author_name = a.get('DisplayName')
 237                     if not author_name:
 238                         continue
 239                     authors.append(author_name)
 240                 common['authors'] = authors
 241
 242             contents = []
 243
 244             if slides:
 245                 d = common.copy()
 246                 d.update({'title': title + '-Slides', 'url': slides})
 247                 contents.append(d)
 248
 249             if zip_file:
 250                 d = common.copy()
 251                 d.update({'title': title + '-Zip', 'url': zip_file})
 252                 contents.append(d)
 253
 254             if formats:
 255                 d = common.copy()
 256                 d.update({'title': title, 'formats': formats})
 257                 contents.append(d)
 258             return self.playlist_result(contents)
 259         else:
 260             return self._extract_list(content_path)