]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/funnyordie.py
debian/rules: Adapt to new upstream build system.
[youtubedl] / 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 m3u8_url = self._search_regex(
49 r'<source[^>]+src=(["\'])(?P<url>.+?/master\.m3u8[^"\']*)\1',
50 webpage, 'm3u8 url', group='url')
51
52 formats = []
53
54 formats.extend(self._extract_m3u8_formats(
55 m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
56
57 bitrates = [int(bitrate) for bitrate in re.findall(r'[,/]v(\d+)[,/]', m3u8_url)]
58 bitrates.sort()
59
60 for bitrate in bitrates:
61 for link in links:
62 formats.append({
63 'url': self._proto_relative_url('%s%d.%s' % (link[0], bitrate, link[1])),
64 'format_id': '%s-%d' % (link[1], bitrate),
65 'vbr': bitrate,
66 })
67
68 subtitles = {}
69 for src, src_lang in re.findall(r'<track kind="captions" src="([^"]+)" srclang="([^"]+)"', webpage):
70 subtitles[src_lang] = [{
71 'ext': src.split('/')[-1],
72 'url': 'http://www.funnyordie.com%s' % src,
73 }]
74
75 post_json = self._search_regex(
76 r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details')
77 post = json.loads(post_json)
78
79 return {
80 'id': video_id,
81 'title': post['name'],
82 'description': post.get('description'),
83 'thumbnail': post.get('picture'),
84 'formats': formats,
85 'subtitles': subtitles,
86 }