_ Git - youtube-dl/blob - 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     def _extract_list(self, video_id, rss_url=None):
  85         if not rss_url:
  86             rss_url = self._RSS_URL % video_id
  87         rss = self._download_xml(rss_url, video_id, 'Downloading RSS')
  88         entries = [self.url_result(session_url.text, 'Channel9')
  89                    for session_url in rss.findall('./channel/item/link')]
  90         title_text = rss.find('./channel/title').text
  91         return self.playlist_result(entries, video_id, title_text)
  92
  93     def _real_extract(self, url):
  94         content_path, rss = re.match(self._VALID_URL, url).groups()
  95
  96         if rss:
  97             return self._extract_list(content_path, url)
  98
  99         webpage = self._download_webpage(
 100             url, content_path, 'Downloading web page')
 101
 102         episode_data = self._search_regex(
 103             r"data-episode='([^']+)'", webpage, 'episode data', default=None)
 104         if episode_data:
 105             episode_data = self._parse_json(unescapeHTML(
 106                 episode_data), content_path)
 107             content_id = episode_data['contentId']
 108             is_session = '/Sessions(' in episode_data['api']
 109             content_url = 'https://channel9.msdn.com/odata' + episode_data['api']
 110             if is_session:
 111                 content_url += '?$expand=Speakers'
 112             else:
 113                 content_url += '?$expand=Authors'
 114             content_data = self._download_json(content_url, content_id)
 115             title = content_data['Title']
 116
 117             QUALITIES = (
 118                 'mp3',
 119                 'wmv', 'mp4',
 120                 'wmv-low', 'mp4-low',
 121                 'wmv-mid', 'mp4-mid',
 122                 'wmv-high', 'mp4-high',
 123             )
 124
 125             quality_key = qualities(QUALITIES)
 126
 127             def quality(quality_id, format_url):
 128                 return (len(QUALITIES) if '_Source.' in format_url
 129                         else quality_key(quality_id))
 130
 131             formats = []
 132             urls = set()
 133
 134             SITE_QUALITIES = {
 135                 'MP3': 'mp3',
 136                 'MP4': 'mp4',
 137                 'Low Quality WMV': 'wmv-low',
 138                 'Low Quality MP4': 'mp4-low',
 139                 'Mid Quality WMV': 'wmv-mid',
 140                 'Mid Quality MP4': 'mp4-mid',
 141                 'High Quality WMV': 'wmv-high',
 142                 'High Quality MP4': 'mp4-high',
 143             }
 144
 145             formats_select = self._search_regex(
 146                 r'(?s)<select[^>]+name=["\']format[^>]+>(.+?)</select', webpage,
 147                 'formats select', default=None)
 148             if formats_select:
 149                 for mobj in re.finditer(
 150                         r'<option\b[^>]+\bvalue=(["\'])(?P<url>(?:(?!\1).)+)\1[^>]*>\s*(?P<format>[^<]+?)\s*<',
 151                         formats_select):
 152                     format_url = mobj.group('url')
 153                     if format_url in urls:
 154                         continue
 155                     urls.add(format_url)
 156                     format_id = mobj.group('format')
 157                     quality_id = SITE_QUALITIES.get(format_id, format_id)
 158                     formats.append({
 159                         'url': format_url,
 160                         'format_id': quality_id,
 161                         'quality': quality(quality_id, format_url),
 162                         'vcodec': 'none' if quality_id == 'mp3' else None,
 163                     })
 164
 165             API_QUALITIES = {
 166                 'VideoMP4Low': 'mp4-low',
 167                 'VideoWMV': 'wmv-mid',
 168                 'VideoMP4Medium': 'mp4-mid',
 169                 'VideoMP4High': 'mp4-high',
 170                 'VideoWMVHQ': 'wmv-hq',
 171             }
 172
 173             for format_id, q in API_QUALITIES.items():
 174                 q_url = content_data.get(format_id)
 175                 if not q_url or q_url in urls:
 176                     continue
 177                 urls.add(q_url)
 178                 formats.append({
 179                     'url': q_url,
 180                     'format_id': q,
 181                     'quality': quality(q, q_url),
 182                 })
 183
 184             self._sort_formats(formats)
 185
 186             slides = content_data.get('Slides')
 187             zip_file = content_data.get('ZipFile')
 188
 189             if not formats and not slides and not zip_file:
 190                 raise ExtractorError(
 191                     'None of recording, slides or zip are available for %s' % content_path)
 192
 193             subtitles = {}
 194             for caption in content_data.get('Captions', []):
 195                 caption_url = caption.get('Url')
 196                 if not caption_url:
 197                     continue
 198                 subtitles.setdefault(caption.get('Language', 'en'), []).append({
 199                     'url': caption_url,
 200                     'ext': 'vtt',
 201                 })
 202
 203             common = {
 204                 'id': content_id,
 205                 'title': title,
 206                 'description': clean_html(content_data.get('Description') or content_data.get('Body')),
 207                 'thumbnail': content_data.get('Thumbnail') or content_data.get('VideoPlayerPreviewImage'),
 208                 'duration': int_or_none(content_data.get('MediaLengthInSeconds')),
 209                 'timestamp': parse_iso8601(content_data.get('PublishedDate')),
 210                 'avg_rating': int_or_none(content_data.get('Rating')),
 211                 'rating_count': int_or_none(content_data.get('RatingCount')),
 212                 'view_count': int_or_none(content_data.get('Views')),
 213                 'comment_count': int_or_none(content_data.get('CommentCount')),
 214                 'subtitles': subtitles,
 215             }
 216             if is_session:
 217                 speakers = []
 218                 for s in content_data.get('Speakers', []):
 219                     speaker_name = s.get('FullName')
 220                     if not speaker_name:
 221                         continue
 222                     speakers.append(speaker_name)
 223
 224                 common.update({
 225                     'session_code': content_data.get('Code'),
 226                     'session_room': content_data.get('Room'),
 227                     'session_speakers': speakers,
 228                 })
 229             else:
 230                 authors = []
 231                 for a in content_data.get('Authors', []):
 232                     author_name = a.get('DisplayName')
 233                     if not author_name:
 234                         continue
 235                     authors.append(author_name)
 236                 common['authors'] = authors
 237
 238             contents = []
 239
 240             if slides:
 241                 d = common.copy()
 242                 d.update({'title': title + '-Slides', 'url': slides})
 243                 contents.append(d)
 244
 245             if zip_file:
 246                 d = common.copy()
 247                 d.update({'title': title + '-Zip', 'url': zip_file})
 248                 contents.append(d)
 249
 250             if formats:
 251                 d = common.copy()
 252                 d.update({'title': title, 'formats': formats})
 253                 contents.append(d)
 254             return self.playlist_result(contents)
 255         else:
 256             return self._extract_list(content_path)