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         }
  22     }, {
  23         'url': 'http://www.liveleak.com/view?i=f93_1390833151',
  24         'md5': 'b13a29626183c9d33944e6a04f41aafc',
  25         'info_dict': {
  26             'id': 'f93_1390833151',
  27             'ext': 'mp4',
  28             '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.',
  29             'uploader': 'ARD_Stinkt',
  30             'title': 'German Television does first Edward Snowden Interview (ENGLISH)',
  31         }
  32     }, {
  33         'url': 'http://www.liveleak.com/view?i=4f7_1392687779',
  34         'md5': '42c6d97d54f1db107958760788c5f48f',
  35         'info_dict': {
  36             'id': '4f7_1392687779',
  37             'ext': 'mp4',
  38             '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.",
  39             'uploader': 'CapObveus',
  40             'title': 'Man is Fatally Struck by Reckless Car While Packing up a Moving Truck',
  41             'age_limit': 18,
  42         }
  43     }, {
  44         'url': 'http://www.liveleak.com/view?i=801_1409392012',
  45         'md5': '0b3bec2d888c20728ca2ad3642f0ef15',
  46         'info_dict': {
  47             'id': '801_1409392012',
  48             'ext': 'mp4',
  49             'description': "Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.",
  50             'uploader': 'bony333',
  51             'title': 'Crazy Hungarian tourist films close call waterspout in Croatia'
  52         }
  53     }]
  54
  55     def _real_extract(self, url):
  56         video_id = self._match_id(url)
  57         webpage = self._download_webpage(url, video_id)
  58
  59         video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
  60         video_description = self._og_search_description(webpage)
  61         video_uploader = self._html_search_regex(
  62             r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
  63         age_limit = int_or_none(self._search_regex(
  64             r'you confirm that you are ([0-9]+) years and over.',
  65             webpage, 'age limit', default=None))
  66
  67         sources_raw = self._search_regex(
  68             r'(?s)sources:\s*(\[.*?\]),', webpage, 'video URLs', default=None)
  69         if sources_raw is None:
  70             alt_source = self._search_regex(
  71                 r'(file: ".*?"),', webpage, 'video URL', default=None)
  72             if alt_source:
  73                 sources_raw = '[{ %s}]' % alt_source
  74             else:
  75                 # Maybe an embed?
  76                 embed_url = self._search_regex(
  77                     r'<iframe[^>]+src="(http://www.prochan.com/embed\?[^"]+)"',
  78                     webpage, 'embed URL')
  79                 return {
  80                     '_type': 'url_transparent',
  81                     'url': embed_url,
  82                     'id': video_id,
  83                     'title': video_title,
  84                     'description': video_description,
  85                     'uploader': video_uploader,
  86                     'age_limit': age_limit,
  87                 }
  88
  89         sources_json = re.sub(r'\s([a-z]+):\s', r'"\1": ', sources_raw)
  90         sources = json.loads(sources_json)
  91
  92         formats = [{
  93             'format_id': '%s' % i,
  94             'format_note': s.get('label'),
  95             'url': s['file'],
  96         } for i, s in enumerate(sources)]
  97         for i, s in enumerate(sources):
  98             # Removing '.h264_*.mp4' gives the raw video, which is essentially
  99             # the same video without the LiveLeak logo at the top (see
 100             # https://github.com/rg3/youtube-dl/pull/4768)
 101             orig_url = re.sub(r'\.h264_.+?\.mp4', '', s['file'])
 102             if s['file'] != orig_url:
 103                 formats.append({
 104                     'format_id': 'original-%s' % i,
 105                     'format_note': s.get('label'),
 106                     'url': orig_url,
 107                     'preference': 1,
 108                 })
 109         self._sort_formats(formats)
 110
 111         return {
 112             'id': video_id,
 113             'title': video_title,
 114             'description': video_description,
 115             'uploader': video_uploader,
 116             'formats': formats,
 117             'age_limit': age_limit,
 118         }