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': 'flv',
  20             'title': 'Terrasses du Numérique',
  21             'duration': 122,
  22         },
  23         'params': {
  24             'skip_download': True,  # Requires rtmpdump
  25         }
  26     }, {
  27         'url': 'http://archives-canalc2.u-strasbg.fr/video.asp?idVideo=11427&voir=oui',
  28         'only_matching': True,
  29     }]
  30
  31     def _real_extract(self, url):
  32         video_id = self._match_id(url)
  33
  34         webpage = self._download_webpage(
  35             'http://www.canalc2.tv/video/%s' % video_id, video_id)
  36
  37         formats = []
  38         for _, video_url in re.findall(r'file\s*=\s*(["\'])(.+?)\1', webpage):
  39             if video_url.startswith('rtmp://'):
  40                 rtmp = re.search(
  41                     r'^(?P<url>rtmp://[^/]+/(?P<app>.+/))(?P<play_path>mp4:.+)$', video_url)
  42                 formats.append({
  43                     'url': rtmp.group('url'),
  44                     'format_id': 'rtmp',
  45                     'ext': 'flv',
  46                     'app': rtmp.group('app'),
  47                     'play_path': rtmp.group('play_path'),
  48                     'page_url': url,
  49                 })
  50             else:
  51                 formats.append({
  52                     'url': video_url,
  53                     'format_id': 'http',
  54                 })
  55         self._sort_formats(formats)
  56
  57         title = self._html_search_regex(
  58             r'(?s)class="[^"]*col_description[^"]*">.*?<h3>(.*?)</h3>', webpage, 'title')
  59         duration = parse_duration(self._search_regex(
  60             r'id=["\']video_duree["\'][^>]*>([^<]+)',
  61             webpage, 'duration', fatal=False))
  62
  63         return {
  64             'id': video_id,
  65             'title': title,
  66             'duration': duration,
  67             'formats': formats,
  68         }