youtube_dl/extractor/veehd.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_urlparse,
   9     get_element_by_id,
  10     clean_html,
  11 )
  12
  13
  14 class VeeHDIE(InfoExtractor):
  15     _VALID_URL = r'https?://veehd\.com/video/(?P<id>\d+)'
  16
  17     _TEST = {
  18         'url': 'http://veehd.com/video/4686958',
  19         'file': '4686958.mp4',
  20         'info_dict': {
  21             'title': 'Time Lapse View from Space ( ISS)',
  22             'uploader_id': 'spotted',
  23             'description': 'md5:f0094c4cf3a72e22bc4e4239ef767ad7',
  24         },
  25     }
  26
  27     def _real_extract(self, url):
  28         mobj = re.match(self._VALID_URL, url)
  29         video_id = mobj.group('id')
  30
  31         # VeeHD seems to send garbage on the first request.
  32         # See https://github.com/rg3/youtube-dl/issues/2102
  33         self._download_webpage(url, video_id, 'Requesting webpage')
  34         webpage = self._download_webpage(url, video_id)
  35         player_path = self._search_regex(
  36             r'\$\("#playeriframe"\).attr\({src : "(.+?)"',
  37             webpage, 'player path')
  38         player_url = compat_urlparse.urljoin(url, player_path)
  39
  40         self._download_webpage(player_url, video_id, 'Requesting player page')
  41         player_page = self._download_webpage(
  42             player_url, video_id, 'Downloading player page')
  43         config_json = self._search_regex(
  44             r'value=\'config=({.+?})\'', player_page, 'config json')
  45         config = json.loads(config_json)
  46
  47         video_url = compat_urlparse.unquote(config['clip']['url'])
  48         title = clean_html(get_element_by_id('videoName', webpage).rpartition('|')[0])
  49         uploader_id = self._html_search_regex(r'<a href="/profile/\d+">(.+?)</a>',
  50             webpage, 'uploader')
  51         thumbnail = self._search_regex(r'<img id="veehdpreview" src="(.+?)"',
  52             webpage, 'thumbnail')
  53         description = self._html_search_regex(r'<td class="infodropdown".*?<div>(.*?)<ul',
  54             webpage, 'description', flags=re.DOTALL)
  55
  56         return {
  57             '_type': 'video',
  58             'id': video_id,
  59             'title': title,
  60             'url': video_url,
  61             'ext': 'mp4',
  62             'uploader_id': uploader_id,
  63             'thumbnail': thumbnail,
  64             'description': description,
  65         }