Raphaël G. Git Repositories - youtubedl/blob - 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)https?://(?: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         'params': {
  49             # m3u8 download
  50             'skip_download': True,
  51         },
  52     }, {
  53         # video with invalid direct format links (HTTP 403)
  54         'url': 'http://videolectures.net/russir2010_filippova_nlp/',
  55         'info_dict': {
  56             'id': '14891',
  57             'display_id': 'russir2010_filippova_nlp',
  58             'ext': 'flv',
  59             'title': 'NLP at Google',
  60             'description': 'md5:fc7a6d9bf0302d7cc0e53f7ca23747b3',
  61             'thumbnail': 're:http://.*\.jpg',
  62             'timestamp': 1284375600,
  63             'upload_date': '20100913',
  64             'duration': 5352,
  65         },
  66         'params': {
  67             # rtmp download
  68             'skip_download': True,
  69         },
  70     }, {
  71         # event playlist
  72         'url': 'http://videolectures.net/deeplearning2015_montreal/',
  73         'info_dict': {
  74             'id': '23181',
  75             'title': 'Deep Learning Summer School, Montreal 2015',
  76             'description': 'md5:0533a85e4bd918df52a01f0e1ebe87b7',
  77             'thumbnail': 're:http://.*\.jpg',
  78             'timestamp': 1438560000,
  79         },
  80         'playlist_count': 30,
  81     }, {
  82         # multi part lecture
  83         'url': 'http://videolectures.net/mlss09uk_bishop_ibi/',
  84         'info_dict': {
  85             'id': '9737',
  86             'display_id': 'mlss09uk_bishop_ibi',
  87             'title': 'Introduction To Bayesian Inference',
  88             'thumbnail': 're:http://.*\.jpg',
  89             'timestamp': 1251622800,
  90         },
  91         'playlist': [{
  92             'info_dict': {
  93                 'id': '9737_part1',
  94                 'display_id': 'mlss09uk_bishop_ibi_part1',
  95                 'ext': 'wmv',
  96                 'title': 'Introduction To Bayesian Inference (Part 1)',
  97                 'thumbnail': 're:http://.*\.jpg',
  98                 'duration': 4622,
  99                 'timestamp': 1251622800,
 100                 'upload_date': '20090830',
 101             },
 102         }, {
 103             'info_dict': {
 104                 'id': '9737_part2',
 105                 'display_id': 'mlss09uk_bishop_ibi_part2',
 106                 'ext': 'wmv',
 107                 'title': 'Introduction To Bayesian Inference (Part 2)',
 108                 'thumbnail': 're:http://.*\.jpg',
 109                 'duration': 5641,
 110                 'timestamp': 1251622800,
 111                 'upload_date': '20090830',
 112             },
 113         }],
 114         'playlist_count': 2,
 115     }]
 116
 117     def _real_extract(self, url):
 118         lecture_slug, explicit_part_id = re.match(self._VALID_URL, url).groups()
 119
 120         webpage = self._download_webpage(url, lecture_slug)
 121
 122         cfg = self._parse_json(self._search_regex(
 123             [r'cfg\s*:\s*({.+?})\s*,\s*[\da-zA-Z_]+\s*:\s*\(?\s*function',
 124              r'cfg\s*:\s*({[^}]+})'],
 125             webpage, 'cfg'), lecture_slug, js_to_json)
 126
 127         lecture_id = compat_str(cfg['obj_id'])
 128
 129         base_url = self._proto_relative_url(cfg['livepipe'], 'http:')
 130
 131         lecture_data = self._download_json(
 132             '%s/site/api/lecture/%s?format=json' % (base_url, lecture_id),
 133             lecture_id)['lecture'][0]
 134
 135         lecture_info = {
 136             'id': lecture_id,
 137             'display_id': lecture_slug,
 138             'title': lecture_data['title'],
 139             'timestamp': parse_iso8601(lecture_data.get('time')),
 140             'description': lecture_data.get('description_wiki'),
 141             'thumbnail': lecture_data.get('thumb'),
 142         }
 143
 144         playlist_entries = []
 145         lecture_type = lecture_data.get('type')
 146         parts = [compat_str(video) for video in cfg.get('videos', [])]
 147         if parts:
 148             multipart = len(parts) > 1
 149
 150             def extract_part(part_id):
 151                 smil_url = '%s/%s/video/%s/smil.xml' % (base_url, lecture_slug, part_id)
 152                 smil = self._download_smil(smil_url, lecture_id)
 153                 info = self._parse_smil(smil, smil_url, lecture_id)
 154                 self._sort_formats(info['formats'])
 155                 info['id'] = lecture_id if not multipart else '%s_part%s' % (lecture_id, part_id)
 156                 info['display_id'] = lecture_slug if not multipart else '%s_part%s' % (lecture_slug, part_id)
 157                 if multipart:
 158                     info['title'] += ' (Part %s)' % part_id
 159                 switch = smil.find('.//switch')
 160                 if switch is not None:
 161                     info['duration'] = parse_duration(switch.attrib.get('dur'))
 162                 item_info = lecture_info.copy()
 163                 item_info.update(info)
 164                 return item_info
 165
 166             if explicit_part_id or not multipart:
 167                 result = extract_part(explicit_part_id or parts[0])
 168             else:
 169                 result = {
 170                     '_type': 'multi_video',
 171                     'entries': [extract_part(part) for part in parts],
 172                 }
 173                 result.update(lecture_info)
 174
 175             # Immediately return explicitly requested part or non event item
 176             if explicit_part_id or lecture_type != 'evt':
 177                 return result
 178
 179             playlist_entries.append(result)
 180
 181         # It's probably a playlist
 182         if not parts or lecture_type == 'evt':
 183             playlist_webpage = self._download_webpage(
 184                 '%s/site/ajax/drilldown/?id=%s' % (base_url, lecture_id), lecture_id)
 185             entries = [
 186                 self.url_result(compat_urlparse.urljoin(url, video_url), 'Viidea')
 187                 for _, video_url in re.findall(
 188                     r'<a[^>]+href=(["\'])(.+?)\1[^>]+id=["\']lec=\d+', playlist_webpage)]
 189             playlist_entries.extend(entries)
 190
 191         playlist = self.playlist_result(playlist_entries, lecture_id)
 192         playlist.update(lecture_info)
 193         return playlist