_ Git - youtube-dl/blob - youtube_dl/extractor/crooksandliars.py

   1 from __future__ import unicode_literals
   2
   3 import json
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     mimetype2ext,
   8 )
   9
  10
  11 class CrooksAndLiarsIE(InfoExtractor):
  12     _VALID_URL = r'(?:https?:)?//embed.crooksandliars.com/embed/(?P<id>[A-Za-z0-9]+)(?:$|[?#])'
  13
  14     _TESTS = [{
  15         'url': 'https://embed.crooksandliars.com/embed/8RUoRhRi',
  16         'info_dict': {
  17             'id': 'https://embed.crooksandliars.com/embed/8RUoRhRi',
  18             'title': "Fox & Friends Says Protecting Atheists From Discrimination Is Anti-Christian!",
  19             'description': "Fox News, Fox & Friends Weekend, April 4, 2015. Read more... http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists",
  20             'timestamp': 1428207000,
  21             'thumbnail': '//crooksandliars.com/files/mediaposters/2015/04/31235.jpg?ts=1428207050',
  22             'uploader': "Heather",
  23         }
  24     }]
  25
  26     def _real_extract(self, url):
  27         video_id = self._match_id(url)
  28         webpage = self._download_webpage(url, video_id)
  29
  30         manifest = json.loads(self._html_search_regex(r'var manifest = ({.*?})\n', webpage, 'manifest JSON'))
  31
  32         formats = []
  33         for item in manifest['flavors']:
  34             if not item['mime'].startswith('video/'): # XXX: or item['exclude']?
  35                 continue
  36             formats.append({
  37                 'format_id': item['type'],
  38                 'ext': mimetype2ext(item['mime']),
  39                 'url': item['url'],
  40             })
  41
  42         # XXX: manifest['url']?
  43         return {
  44             'url': url,
  45             'id': video_id,
  46             'uploader': manifest['author'],
  47             'title': manifest['title'],
  48             'description': manifest['description'],
  49             'thumbnail': manifest['poster'],
  50             'duration': manifest['duration'],
  51             'timestamp': int(manifest['created']),
  52             'formats': formats,
  53         }
  54
  55 class CrooksAndLiarsArticleIE(InfoExtractor):
  56     _VALID_URL = r'(?:https?:)?//crooksandliars.com/\d+/\d+/(?P<id>[a-z\-]+)(?:/|$)'
  57
  58     _TESTS = [{
  59         'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
  60         'only_matching': True,
  61     }]
  62
  63     def _real_extract(self, url):
  64         video_id = self._match_id(url)
  65         webpage = self._download_webpage(url, video_id)
  66         player_url = self._proto_relative_url(self._html_search_regex(r'<iframe src="(//embed.crooksandliars.com/.*)"', webpage, 'embedded player'))
  67
  68         return {
  69             '_type': 'url',
  70             'url': player_url
  71         }