youtube_dl/extractor/pyvideo.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import os
   5
   6 from .common import InfoExtractor
   7
   8
   9 class PyvideoIE(InfoExtractor):
  10     _VALID_URL = r'http://(?:www\.)?pyvideo\.org/video/(?P<id>\d+)/(.*)'
  11
  12     _TESTS = [
  13         {
  14             'url': 'http://pyvideo.org/video/1737/become-a-logging-expert-in-30-minutes',
  15             'md5': 'de317418c8bc76b1fd8633e4f32acbc6',
  16             'info_dict': {
  17                 'id': '24_4WWkSmNo',
  18                 'ext': 'mp4',
  19                 'title': 'Become a logging expert in 30 minutes',
  20                 'description': 'md5:9665350d466c67fb5b1598de379021f7',
  21                 'upload_date': '20130320',
  22                 'uploader': 'NextDayVideo',
  23                 'uploader_id': 'NextDayVideo',
  24             },
  25             'add_ie': ['Youtube'],
  26         },
  27         {
  28             'url': 'http://pyvideo.org/video/2542/gloriajw-spotifywitherikbernhardsson182m4v',
  29             'md5': '5fe1c7e0a8aa5570330784c847ff6d12',
  30             'info_dict': {
  31                 'id': '2542',
  32                 'ext': 'm4v',
  33                 'title': 'Gloriajw-SpotifyWithErikBernhardsson182',
  34             },
  35         },
  36     ]
  37
  38     def _real_extract(self, url):
  39         mobj = re.match(self._VALID_URL, url)
  40         video_id = mobj.group('id')
  41
  42         webpage = self._download_webpage(url, video_id)
  43
  44         m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', webpage)
  45         if m_youtube is not None:
  46             return self.url_result(m_youtube.group(1), 'Youtube')
  47
  48         title = self._html_search_regex(
  49             r'<div class="section">\s*<h3(?:\s+class="[^"]*"[^>]*)?>([^>]+?)</h3>',
  50             webpage, 'title', flags=re.DOTALL)
  51         video_url = self._search_regex(
  52             [r'<source src="(.*?)"', r'<dt>Download</dt>.*?<a href="(.+?)"'],
  53             webpage, 'video url', flags=re.DOTALL)
  54
  55         return {
  56             'id': video_id,
  57             'title': os.path.splitext(title)[0],
  58             'url': video_url,
  59         }