youtube_dl/extractor/viidea.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urlparse,
   8     compat_str,
   9 )
  10 from ..utils import (
  11     parse_duration,
  12     js_to_json,
  13     parse_iso8601,
  14 )
  15
  16
  17 class ViideaIE(InfoExtractor):
  18     _VALID_URL = r'''(?x)http://(?:www\.)?(?:
  19             videolectures\.net|
  20             flexilearn\.viidea\.net|
  21             presentations\.ocwconsortium\.org|
  22             video\.travel-zoom\.si|
  23             video\.pomp-forum\.si|
  24             tv\.nil\.si|
  25             video\.hekovnik.com|
  26             video\.szko\.si|
  27             kpk\.viidea\.com|
  28             inside\.viidea\.net|
  29             video\.kiberpipa\.org|
  30             bvvideo\.si|
  31             kongres\.viidea\.net|
  32             edemokracija\.viidea\.com
  33         )(?:/lecture)?/(?P<id>[^/]+)(?:/video/(?P<part>\d+))?/*(?:[#?].*)?$'''
  34
  35     _TESTS = [{
  36         'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/',
  37         'info_dict': {
  38             'id': '20171',
  39             'display_id': 'promogram_igor_mekjavic_eng',
  40             'ext': 'mp4',
  41             'title': 'Automatics, robotics and biocybernetics',
  42             'description': 'md5:815fc1deb6b3a2bff99de2d5325be482',
  43             'thumbnail': 're:http://.*\.jpg',
  44             'timestamp': 1372349289,
  45             'upload_date': '20130627',
  46             'duration': 565,
  47         },
  48     }, {
  49         # video with invalid direct format links (HTTP 403)
  50         'url': 'http://videolectures.net/russir2010_filippova_nlp/',
  51         'info_dict': {
  52             'id': '14891',
  53             'display_id': 'russir2010_filippova_nlp',
  54             'ext': 'flv',
  55             'title': 'NLP at Google',
  56             'description': 'md5:fc7a6d9bf0302d7cc0e53f7ca23747b3',
  57             'thumbnail': 're:http://.*\.jpg',
  58             'timestamp': 1284375600,
  59             'upload_date': '20100913',
  60             'duration': 5352,
  61         },
  62         'params': {
  63             # rtmp download
  64             'skip_download': True,
  65         },
  66     }, {
  67         # event playlist
  68         'url': 'http://videolectures.net/deeplearning2015_montreal/',
  69         'info_dict': {
  70             'id': '23181',
  71             'title': 'Deep Learning Summer School, Montreal 2015',
  72             'description': 'md5:0533a85e4bd918df52a01f0e1ebe87b7',
  73             'thumbnail': 're:http://.*\.jpg',
  74             'timestamp': 1438560000,
  75         },
  76         'playlist_count': 30,
  77     }, {
  78         # multi part lecture
  79         'url': 'http://videolectures.net/mlss09uk_bishop_ibi/',
  80         'info_dict': {
  81             'id': '9737',
  82             'display_id': 'mlss09uk_bishop_ibi',
  83             'title': 'Introduction To Bayesian Inference',
  84             'thumbnail': 're:http://.*\.jpg',
  85             'timestamp': 1251622800,
  86         },
  87         'playlist': [{
  88             'info_dict': {
  89                 'id': '9737_part1',
  90                 'display_id': 'mlss09uk_bishop_ibi_part1',
  91                 'ext': 'wmv',
  92                 'title': 'Introduction To Bayesian Inference (Part 1)',
  93                 'thumbnail': 're:http://.*\.jpg',
  94                 'duration': 4622,
  95                 'timestamp': 1251622800,
  96                 'upload_date': '20090830',
  97             },
  98         }, {
  99             'info_dict': {
 100                 'id': '9737_part2',
 101                 'display_id': 'mlss09uk_bishop_ibi_part2',
 102                 'ext': 'wmv',
 103                 'title': 'Introduction To Bayesian Inference (Part 2)',
 104                 'thumbnail': 're:http://.*\.jpg',
 105                 'duration': 5641,
 106                 'timestamp': 1251622800,
 107                 'upload_date': '20090830',
 108             },
 109         }],
 110         'playlist_count': 2,
 111     }]
 112
 113     def _real_extract(self, url):
 114         lecture_slug, explicit_part_id = re.match(self._VALID_URL, url).groups()
 115
 116         webpage = self._download_webpage(url, lecture_slug)
 117
 118         cfg = self._parse_json(self._search_regex(
 119             [r'cfg\s*:\s*({.+?})\s*,\s*[\da-zA-Z_]+\s*:\s*\(?\s*function',
 120              r'cfg\s*:\s*({[^}]+})'],
 121             webpage, 'cfg'), lecture_slug, js_to_json)
 122
 123         lecture_id = compat_str(cfg['obj_id'])
 124
 125         base_url = self._proto_relative_url(cfg['livepipe'], 'http:')
 126
 127         lecture_data = self._download_json(
 128             '%s/site/api/lecture/%s?format=json' % (base_url, lecture_id),
 129             lecture_id)['lecture'][0]
 130
 131         lecture_info = {
 132             'id': lecture_id,
 133             'display_id': lecture_slug,
 134             'title': lecture_data['title'],
 135             'timestamp': parse_iso8601(lecture_data.get('time')),
 136             'description': lecture_data.get('description_wiki'),
 137             'thumbnail': lecture_data.get('thumb'),
 138         }
 139
 140         playlist_entries = []
 141         lecture_type = lecture_data.get('type')
 142         parts = [compat_str(video) for video in cfg.get('videos', [])]
 143         if parts:
 144             multipart = len(parts) > 1
 145
 146             def extract_part(part_id):
 147                 smil_url = '%s/%s/video/%s/smil.xml' % (base_url, lecture_slug, part_id)
 148                 smil = self._download_smil(smil_url, lecture_id)
 149                 info = self._parse_smil(smil, smil_url, lecture_id)
 150                 info['id'] = lecture_id if not multipart else '%s_part%s' % (lecture_id, part_id)
 151                 info['display_id'] = lecture_slug if not multipart else '%s_part%s' % (lecture_slug, part_id)
 152                 if multipart:
 153                     info['title'] += ' (Part %s)' % part_id
 154                 switch = smil.find('.//switch')
 155                 if switch is not None:
 156                     info['duration'] = parse_duration(switch.attrib.get('dur'))
 157                 item_info = lecture_info.copy()
 158                 item_info.update(info)
 159                 return item_info
 160
 161             if explicit_part_id or not multipart:
 162                 result = extract_part(explicit_part_id or parts[0])
 163             else:
 164                 result = {
 165                     '_type': 'multi_video',
 166                     'entries': [extract_part(part) for part in parts],
 167                 }
 168                 result.update(lecture_info)
 169
 170             # Immediately return explicitly requested part or non event item
 171             if explicit_part_id or lecture_type != 'evt':
 172                 return result
 173
 174             playlist_entries.append(result)
 175
 176         # It's probably a playlist
 177         if not parts or lecture_type == 'evt':
 178             playlist_webpage = self._download_webpage(
 179                 '%s/site/ajax/drilldown/?id=%s' % (base_url, lecture_id), lecture_id)
 180             entries = [
 181                 self.url_result(compat_urlparse.urljoin(url, video_url), 'Viidea')
 182                 for _, video_url in re.findall(
 183                     r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', playlist_webpage)]
 184             playlist_entries.extend(entries)
 185
 186         playlist = self.playlist_result(playlist_entries, lecture_id)
 187         playlist.update(lecture_info)
 188         return playlist