]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/fourtube.py
debian/copyright: Fix lintian's dep5-copyright-license-name-not-unique.
[youtubedl] / youtube_dl / extractor / fourtube.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..compat import (
7 compat_urllib_request,
8 )
9 from ..utils import (
10 parse_duration,
11 parse_iso8601,
12 str_to_int,
13 )
14
15
16 class FourTubeIE(InfoExtractor):
17 IE_NAME = '4tube'
18 _VALID_URL = r'https?://(?:www\.)?4tube\.com/videos/(?P<id>\d+)'
19
20 _TEST = {
21 'url': 'http://www.4tube.com/videos/209733/hot-babe-holly-michaels-gets-her-ass-stuffed-by-black',
22 'md5': '6516c8ac63b03de06bc8eac14362db4f',
23 'info_dict': {
24 'id': '209733',
25 'ext': 'mp4',
26 'title': 'Hot Babe Holly Michaels gets her ass stuffed by black',
27 'uploader': 'WCP Club',
28 'uploader_id': 'wcp-club',
29 'upload_date': '20131031',
30 'timestamp': 1383263892,
31 'duration': 583,
32 'view_count': int,
33 'like_count': int,
34 'categories': list,
35 }
36 }
37
38 def _real_extract(self, url):
39 video_id = self._match_id(url)
40 webpage = self._download_webpage(url, video_id)
41
42 title = self._html_search_meta('name', webpage)
43 timestamp = parse_iso8601(self._html_search_meta(
44 'uploadDate', webpage))
45 thumbnail = self._html_search_meta('thumbnailUrl', webpage)
46 uploader_id = self._html_search_regex(
47 r'<a class="img-avatar" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">',
48 webpage, 'uploader id')
49 uploader = self._html_search_regex(
50 r'<a class="img-avatar" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">',
51 webpage, 'uploader')
52
53 categories_html = self._search_regex(
54 r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>',
55 webpage, 'categories', fatal=False)
56 categories = None
57 if categories_html:
58 categories = [
59 c.strip() for c in re.findall(
60 r'(?s)<li><a.*?>(.*?)</a>', categories_html)]
61
62 view_count = str_to_int(self._search_regex(
63 r'<meta itemprop="interactionCount" content="UserPlays:([0-9,]+)">',
64 webpage, 'view count', fatal=False))
65 like_count = str_to_int(self._search_regex(
66 r'<meta itemprop="interactionCount" content="UserLikes:([0-9,]+)">',
67 webpage, 'like count', fatal=False))
68 duration = parse_duration(self._html_search_meta('duration', webpage))
69
70 params_js = self._search_regex(
71 r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)',
72 webpage, 'initialization parameters'
73 )
74 params = self._parse_json('[%s]' % params_js, video_id)
75 media_id = params[0]
76 sources = ['%s' % p for p in params[2]]
77
78 token_url = 'http://tkn.4tube.com/{0}/desktop/{1}'.format(
79 media_id, '+'.join(sources))
80 headers = {
81 b'Content-Type': b'application/x-www-form-urlencoded',
82 b'Origin': b'http://www.4tube.com',
83 }
84 token_req = compat_urllib_request.Request(token_url, b'{}', headers)
85 tokens = self._download_json(token_req, video_id)
86 formats = [{
87 'url': tokens[format]['token'],
88 'format_id': format + 'p',
89 'resolution': format + 'p',
90 'quality': int(format),
91 } for format in sources]
92 self._sort_formats(formats)
93
94 return {
95 'id': video_id,
96 'title': title,
97 'formats': formats,
98 'categories': categories,
99 'thumbnail': thumbnail,
100 'uploader': uploader,
101 'uploader_id': uploader_id,
102 'timestamp': timestamp,
103 'like_count': like_count,
104 'view_count': view_count,
105 'duration': duration,
106 'age_limit': 18,
107 }