youtube_dl/extractor/canalc2.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import parse_duration
   8
   9
  10 class Canalc2IE(InfoExtractor):
  11     IE_NAME = 'canalc2.tv'
  12     _VALID_URL = r'https?://(?:(?:www\.)?canalc2\.tv/video/|archives-canalc2\.u-strasbg\.fr/video\.asp\?.*\bidVideo=)(?P<id>\d+)'
  13
  14     _TESTS = [{
  15         'url': 'http://www.canalc2.tv/video/12163',
  16         'md5': '060158428b650f896c542dfbb3d6487f',
  17         'info_dict': {
  18             'id': '12163',
  19             'ext': 'mp4',
  20             'title': 'Terrasses du Numérique',
  21             'duration': 122,
  22         },
  23     }, {
  24         'url': 'http://archives-canalc2.u-strasbg.fr/video.asp?idVideo=11427&voir=oui',
  25         'only_matching': True,
  26     }]
  27
  28     def _real_extract(self, url):
  29         video_id = self._match_id(url)
  30
  31         webpage = self._download_webpage(
  32             'http://www.canalc2.tv/video/%s' % video_id, video_id)
  33
  34         title = self._html_search_regex(
  35             r'(?s)class="[^"]*col_description[^"]*">.*?<h3>(.+?)</h3>',
  36             webpage, 'title')
  37
  38         formats = []
  39         for _, video_url in re.findall(r'file\s*=\s*(["\'])(.+?)\1', webpage):
  40             if video_url.startswith('rtmp://'):
  41                 rtmp = re.search(
  42                     r'^(?P<url>rtmp://[^/]+/(?P<app>.+/))(?P<play_path>mp4:.+)$', video_url)
  43                 formats.append({
  44                     'url': rtmp.group('url'),
  45                     'format_id': 'rtmp',
  46                     'ext': 'flv',
  47                     'app': rtmp.group('app'),
  48                     'play_path': rtmp.group('play_path'),
  49                     'page_url': url,
  50                 })
  51             else:
  52                 formats.append({
  53                     'url': video_url,
  54                     'format_id': 'http',
  55                 })
  56
  57         if formats:
  58             info = {
  59                 'formats': formats,
  60             }
  61         else:
  62             info = self._parse_html5_media_entries(url, webpage, url)[0]
  63
  64         self._sort_formats(info['formats'])
  65
  66         info.update({
  67             'id': video_id,
  68             'title': title,
  69             'duration': parse_duration(self._search_regex(
  70                 r'id=["\']video_duree["\'][^>]*>([^<]+)',
  71                 webpage, 'duration', fatal=False)),
  72         })
  73         return info