Raphaël G. Git Repositories - youtubedl/blob - 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'^(?:http://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_id>[\w_]+)(?:.*)'
  12     _TESTS = [{
  13         'url': 'http://www.liveleak.com/view?i=757_1364311680',
  14         'md5': '0813c2430bea7a46bf13acf3406992f4',
  15         'info_dict': {
  16             'id': '757_1364311680',
  17             'ext': 'mp4',
  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': 'd3f1367d14cc3c15bf24fbfbe04b9abf',
  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
  45     def _real_extract(self, url):
  46         mobj = re.match(self._VALID_URL, url)
  47         video_id = mobj.group('video_id')
  48         webpage = self._download_webpage(url, video_id)
  49
  50         video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
  51         video_description = self._og_search_description(webpage)
  52         video_uploader = self._html_search_regex(
  53             r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
  54         age_limit = int_or_none(self._search_regex(
  55             r'you confirm that you are ([0-9]+) years and over.',
  56             webpage, 'age limit', default=None))
  57
  58         sources_raw = self._search_regex(
  59             r'(?s)sources:\s*(\[.*?\]),', webpage, 'video URLs', default=None)
  60         if sources_raw is None:
  61             alt_source = self._search_regex(
  62                 r'(file: ".*?"),', webpage, 'video URL', default=None)
  63             if alt_source:
  64                 sources_raw = '[{ %s}]' % alt_source
  65             else:
  66                 # Maybe an embed?
  67                 embed_url = self._search_regex(
  68                     r'<iframe[^>]+src="(http://www.prochan.com/embed\?[^"]+)"',
  69                     webpage, 'embed URL')
  70                 return {
  71                     '_type': 'url_transparent',
  72                     'url': embed_url,
  73                     'id': video_id,
  74                     'title': video_title,
  75                     'description': video_description,
  76                     'uploader': video_uploader,
  77                     'age_limit': age_limit,
  78                 }
  79
  80         sources_json = re.sub(r'\s([a-z]+):\s', r'"\1": ', sources_raw)
  81         sources = json.loads(sources_json)
  82
  83         formats = [{
  84             'format_note': s.get('label'),
  85             'url': s['file'],
  86         } for s in sources]
  87         self._sort_formats(formats)
  88
  89         return {
  90             'id': video_id,
  91             'title': video_title,
  92             'description': video_description,
  93             'uploader': video_uploader,
  94             'formats': formats,
  95             'age_limit': age_limit,
  96         }