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