yt_dlp/extractor/canalc2.py

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