_ Git - youtube-dl/blob - 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         'info_dict': {
  20             'id': '4686958',
  21             'ext': 'mp4',
  22             'title': 'Time Lapse View from Space ( ISS)',
  23             'uploader_id': 'spotted',
  24             'description': 'md5:f0094c4cf3a72e22bc4e4239ef767ad7',
  25         },
  26     }
  27
  28     def _real_extract(self, url):
  29         mobj = re.match(self._VALID_URL, url)
  30         video_id = mobj.group('id')
  31
  32         # VeeHD seems to send garbage on the first request.
  33         # See https://github.com/rg3/youtube-dl/issues/2102
  34         self._download_webpage(url, video_id, 'Requesting webpage')
  35         webpage = self._download_webpage(url, video_id)
  36         player_path = self._search_regex(
  37             r'\$\("#playeriframe"\).attr\({src : "(.+?)"',
  38             webpage, 'player path')
  39         player_url = compat_urlparse.urljoin(url, player_path)
  40
  41         self._download_webpage(player_url, video_id, 'Requesting player page')
  42         player_page = self._download_webpage(
  43             player_url, video_id, 'Downloading player page')
  44         config_json = self._search_regex(
  45             r'value=\'config=({.+?})\'', player_page, 'config json')
  46         config = json.loads(config_json)
  47
  48         video_url = compat_urlparse.unquote(config['clip']['url'])
  49         title = clean_html(get_element_by_id('videoName', webpage).rpartition('|')[0])
  50         uploader_id = self._html_search_regex(r'<a href="/profile/\d+">(.+?)</a>',
  51                                               webpage, 'uploader')
  52         thumbnail = self._search_regex(r'<img id="veehdpreview" src="(.+?)"',
  53                                        webpage, 'thumbnail')
  54         description = self._html_search_regex(r'<td class="infodropdown".*?<div>(.*?)<ul',
  55                                               webpage, 'description', flags=re.DOTALL)
  56
  57         return {
  58             '_type': 'video',
  59             'id': video_id,
  60             'title': title,
  61             'url': video_url,
  62             'ext': 'mp4',
  63             'uploader_id': uploader_id,
  64             'thumbnail': thumbnail,
  65             'description': description,
  66         }