Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/funnyordie.py

   1 from __future__ import unicode_literals
   2
   3 import json
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import ExtractorError
   8
   9
  10 class FunnyOrDieIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?funnyordie\.com/(?P<type>embed|articles|videos)/(?P<id>[0-9a-f]+)(?:$|[?#/])'
  12     _TESTS = [{
  13         'url': 'http://www.funnyordie.com/videos/0732f586d7/heart-shaped-box-literal-video-version',
  14         'md5': 'bcd81e0c4f26189ee09be362ad6e6ba9',
  15         'info_dict': {
  16             'id': '0732f586d7',
  17             'ext': 'mp4',
  18             'title': 'Heart-Shaped Box: Literal Video Version',
  19             'description': 'md5:ea09a01bc9a1c46d9ab696c01747c338',
  20             'thumbnail': 're:^http:.*\.jpg$',
  21         },
  22     }, {
  23         'url': 'http://www.funnyordie.com/embed/e402820827',
  24         'info_dict': {
  25             'id': 'e402820827',
  26             'ext': 'mp4',
  27             'title': 'Please Use This Song (Jon Lajoie)',
  28             'description': 'Please use this to sell something.  www.jonlajoie.com',
  29             'thumbnail': 're:^http:.*\.jpg$',
  30         },
  31     }, {
  32         'url': 'http://www.funnyordie.com/articles/ebf5e34fc8/10-hours-of-walking-in-nyc-as-a-man',
  33         'only_matching': True,
  34     }]
  35
  36     def _real_extract(self, url):
  37         mobj = re.match(self._VALID_URL, url)
  38
  39         video_id = mobj.group('id')
  40         webpage = self._download_webpage(url, video_id)
  41
  42         links = re.findall(r'<source src="([^"]+/v)[^"]+\.([^"]+)" type=\'video', webpage)
  43         if not links:
  44             raise ExtractorError('No media links available for %s' % video_id)
  45
  46         links.sort(key=lambda link: 1 if link[1] == 'mp4' else 0)
  47
  48         bitrates = self._html_search_regex(r'<source src="[^"]+/v,((?:\d+,)+)\.mp4\.csmil', webpage, 'video bitrates')
  49         bitrates = [int(b) for b in bitrates.rstrip(',').split(',')]
  50         bitrates.sort()
  51
  52         formats = []
  53         for bitrate in bitrates:
  54             for link in links:
  55                 formats.append({
  56                     'url': '%s%d.%s' % (link[0], bitrate, link[1]),
  57                     'format_id': '%s-%d' % (link[1], bitrate),
  58                     'vbr': bitrate,
  59                 })
  60
  61         subtitles = {}
  62         for src, src_lang in re.findall(r'<track kind="captions" src="([^"]+)" srclang="([^"]+)"', webpage):
  63             subtitles[src_lang] = [{
  64                 'ext': src.split('/')[-1],
  65                 'url': 'http://www.funnyordie.com%s' % src,
  66             }]
  67
  68         post_json = self._search_regex(
  69             r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details')
  70         post = json.loads(post_json)
  71
  72         return {
  73             'id': video_id,
  74             'title': post['name'],
  75             'description': post.get('description'),
  76             'thumbnail': post.get('picture'),
  77             'formats': formats,
  78             'subtitles': subtitles,
  79         }