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         'params': {
  32             'skip_download': True,
  33         },
  34     }, {
  35         'url': 'http://www.funnyordie.com/articles/ebf5e34fc8/10-hours-of-walking-in-nyc-as-a-man',
  36         'only_matching': True,
  37     }]
  38
  39     def _real_extract(self, url):
  40         mobj = re.match(self._VALID_URL, url)
  41
  42         video_id = mobj.group('id')
  43         webpage = self._download_webpage(url, video_id)
  44
  45         links = re.findall(r'<source src="([^"]+/v)[^"]+\.([^"]+)" type=\'video', webpage)
  46         if not links:
  47             raise ExtractorError('No media links available for %s' % video_id)
  48
  49         links.sort(key=lambda link: 1 if link[1] == 'mp4' else 0)
  50
  51         m3u8_url = self._search_regex(
  52             r'<source[^>]+src=(["\'])(?P<url>.+?/master\.m3u8[^"\']*)\1',
  53             webpage, 'm3u8 url', group='url')
  54
  55         formats = []
  56
  57         m3u8_formats = self._extract_m3u8_formats(
  58             m3u8_url, video_id, 'mp4', 'm3u8_native',
  59             m3u8_id='hls', fatal=False)
  60         source_formats = list(filter(
  61             lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
  62             m3u8_formats))
  63
  64         bitrates = [int(bitrate) for bitrate in re.findall(r'[,/]v(\d+)(?=[,/])', m3u8_url)]
  65         bitrates.sort()
  66
  67         if source_formats:
  68             self._sort_formats(source_formats)
  69
  70         for bitrate, f in zip(bitrates, source_formats or [{}] * len(bitrates)):
  71             for path, ext in links:
  72                 ff = f.copy()
  73                 if ff:
  74                     if ext != 'mp4':
  75                         ff = dict(
  76                             [(k, v) for k, v in ff.items()
  77                              if k in ('height', 'width', 'format_id')])
  78                     ff.update({
  79                         'format_id': ff['format_id'].replace('hls', ext),
  80                         'ext': ext,
  81                         'protocol': 'http',
  82                     })
  83                 else:
  84                     ff.update({
  85                         'format_id': '%s-%d' % (ext, bitrate),
  86                         'vbr': bitrate,
  87                     })
  88                 ff['url'] = self._proto_relative_url(
  89                     '%s%d.%s' % (path, bitrate, ext))
  90                 formats.append(ff)
  91         self._check_formats(formats, video_id)
  92
  93         formats.extend(m3u8_formats)
  94         self._sort_formats(
  95             formats, field_preference=('height', 'width', 'tbr', 'format_id'))
  96
  97         subtitles = {}
  98         for src, src_lang in re.findall(r'<track kind="captions" src="([^"]+)" srclang="([^"]+)"', webpage):
  99             subtitles[src_lang] = [{
 100                 'ext': src.split('/')[-1],
 101                 'url': 'http://www.funnyordie.com%s' % src,
 102             }]
 103
 104         post_json = self._search_regex(
 105             r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details')
 106         post = json.loads(post_json)
 107
 108         return {
 109             'id': video_id,
 110             'title': post['name'],
 111             'description': post.get('description'),
 112             'thumbnail': post.get('picture'),
 113             'formats': formats,
 114             'subtitles': subtitles,
 115         }