]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/liveleak.py
Tidy up cleanup rule and keep lintian happy.
[youtubedl] / youtube_dl / extractor / liveleak.py
1 from __future__ import unicode_literals
2
3 import json
4 import re
5
6 from .common import InfoExtractor
7 from ..utils import int_or_none
8
9
10 class LiveLeakIE(InfoExtractor):
11 _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<id>[\w_]+)(?:.*)'
12 _TESTS = [{
13 'url': 'http://www.liveleak.com/view?i=757_1364311680',
14 'md5': '50f79e05ba149149c1b4ea961223d5b3',
15 'info_dict': {
16 'id': '757_1364311680',
17 'ext': 'flv',
18 'description': 'extremely bad day for this guy..!',
19 'uploader': 'ljfriel2',
20 'title': 'Most unlucky car accident',
21 'thumbnail': 're:^https?://.*\.jpg$'
22 }
23 }, {
24 'url': 'http://www.liveleak.com/view?i=f93_1390833151',
25 'md5': 'b13a29626183c9d33944e6a04f41aafc',
26 'info_dict': {
27 'id': 'f93_1390833151',
28 'ext': 'mp4',
29 'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.',
30 'uploader': 'ARD_Stinkt',
31 'title': 'German Television does first Edward Snowden Interview (ENGLISH)',
32 'thumbnail': 're:^https?://.*\.jpg$'
33 }
34 }, {
35 'url': 'http://www.liveleak.com/view?i=4f7_1392687779',
36 'md5': '42c6d97d54f1db107958760788c5f48f',
37 'info_dict': {
38 'id': '4f7_1392687779',
39 'ext': 'mp4',
40 'description': "The guy with the cigarette seems amazingly nonchalant about the whole thing... I really hope my friends' reactions would be a bit stronger.\r\n\r\nAction-go to 0:55.",
41 'uploader': 'CapObveus',
42 'title': 'Man is Fatally Struck by Reckless Car While Packing up a Moving Truck',
43 'age_limit': 18,
44 }
45 }, {
46 # Covers https://github.com/rg3/youtube-dl/pull/5983
47 'url': 'http://www.liveleak.com/view?i=801_1409392012',
48 'md5': '0b3bec2d888c20728ca2ad3642f0ef15',
49 'info_dict': {
50 'id': '801_1409392012',
51 'ext': 'mp4',
52 'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.',
53 'uploader': 'bony333',
54 'title': 'Crazy Hungarian tourist films close call waterspout in Croatia',
55 'thumbnail': 're:^https?://.*\.jpg$'
56 }
57 }]
58
59 @staticmethod
60 def _extract_url(webpage):
61 mobj = re.search(
62 r'<iframe[^>]+src="https?://(?:\w+\.)?liveleak\.com/ll_embed\?(?:.*?)i=(?P<id>[\w_]+)(?:.*)',
63 webpage)
64 if mobj:
65 return 'http://www.liveleak.com/view?i=%s' % mobj.group('id')
66
67 def _real_extract(self, url):
68 video_id = self._match_id(url)
69 webpage = self._download_webpage(url, video_id)
70
71 video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
72 video_description = self._og_search_description(webpage)
73 video_uploader = self._html_search_regex(
74 r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
75 age_limit = int_or_none(self._search_regex(
76 r'you confirm that you are ([0-9]+) years and over.',
77 webpage, 'age limit', default=None))
78 video_thumbnail = self._og_search_thumbnail(webpage)
79
80 sources_raw = self._search_regex(
81 r'(?s)sources:\s*(\[.*?\]),', webpage, 'video URLs', default=None)
82 if sources_raw is None:
83 alt_source = self._search_regex(
84 r'(file: ".*?"),', webpage, 'video URL', default=None)
85 if alt_source:
86 sources_raw = '[{ %s}]' % alt_source
87 else:
88 # Maybe an embed?
89 embed_url = self._search_regex(
90 r'<iframe[^>]+src="(http://www.prochan.com/embed\?[^"]+)"',
91 webpage, 'embed URL')
92 return {
93 '_type': 'url_transparent',
94 'url': embed_url,
95 'id': video_id,
96 'title': video_title,
97 'description': video_description,
98 'uploader': video_uploader,
99 'age_limit': age_limit,
100 }
101
102 sources_json = re.sub(r'\s([a-z]+):\s', r'"\1": ', sources_raw)
103 sources = json.loads(sources_json)
104
105 formats = [{
106 'format_id': '%s' % i,
107 'format_note': s.get('label'),
108 'url': s['file'],
109 } for i, s in enumerate(sources)]
110 for i, s in enumerate(sources):
111 # Removing '.h264_*.mp4' gives the raw video, which is essentially
112 # the same video without the LiveLeak logo at the top (see
113 # https://github.com/rg3/youtube-dl/pull/4768)
114 orig_url = re.sub(r'\.h264_.+?\.mp4', '', s['file'])
115 if s['file'] != orig_url:
116 formats.append({
117 'format_id': 'original-%s' % i,
118 'format_note': s.get('label'),
119 'url': orig_url,
120 'preference': 1,
121 })
122 self._sort_formats(formats)
123
124 return {
125 'id': video_id,
126 'title': video_title,
127 'description': video_description,
128 'uploader': video_uploader,
129 'formats': formats,
130 'age_limit': age_limit,
131 'thumbnail': video_thumbnail,
132 }