Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/liveleak.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import int_or_none
   7
   8
   9 class LiveLeakIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?.*?\b[it]=(?P<id>[\w_]+)'
  11     _TESTS = [{
  12         'url': 'http://www.liveleak.com/view?i=757_1364311680',
  13         'md5': '0813c2430bea7a46bf13acf3406992f4',
  14         'info_dict': {
  15             'id': '757_1364311680',
  16             'ext': 'mp4',
  17             'description': 'extremely bad day for this guy..!',
  18             'uploader': 'ljfriel2',
  19             'title': 'Most unlucky car accident',
  20             'thumbnail': r're:^https?://.*\.jpg$'
  21         }
  22     }, {
  23         'url': 'http://www.liveleak.com/view?i=f93_1390833151',
  24         'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf',
  25         'info_dict': {
  26             'id': 'f93_1390833151',
  27             'ext': 'mp4',
  28             'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.',
  29             'uploader': 'ARD_Stinkt',
  30             'title': 'German Television does first Edward Snowden Interview (ENGLISH)',
  31             'thumbnail': r're:^https?://.*\.jpg$'
  32         }
  33     }, {
  34         # Prochan embed
  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         'skip': 'Video is dead',
  46     }, {
  47         # Covers https://github.com/rg3/youtube-dl/pull/5983
  48         # Multiple resolutions
  49         'url': 'http://www.liveleak.com/view?i=801_1409392012',
  50         'md5': 'c3a449dbaca5c0d1825caecd52a57d7b',
  51         'info_dict': {
  52             'id': '801_1409392012',
  53             'ext': 'mp4',
  54             'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.',
  55             'uploader': 'bony333',
  56             'title': 'Crazy Hungarian tourist films close call waterspout in Croatia',
  57             'thumbnail': r're:^https?://.*\.jpg$'
  58         }
  59     }, {
  60         # Covers https://github.com/rg3/youtube-dl/pull/10664#issuecomment-247439521
  61         'url': 'http://m.liveleak.com/view?i=763_1473349649',
  62         'add_ie': ['Youtube'],
  63         'info_dict': {
  64             'id': '763_1473349649',
  65             'ext': 'mp4',
  66             'title': 'Reporters and public officials ignore epidemic of black on asian violence in Sacramento | Colin Flaherty',
  67             'description': 'Colin being the warrior he is and showing the injustice Asians in Sacramento are being subjected to.',
  68             'uploader': 'Ziz',
  69             'upload_date': '20160908',
  70             'uploader_id': 'UCEbta5E_jqlZmEJsriTEtnw'
  71         },
  72         'params': {
  73             'skip_download': True,
  74         },
  75     }, {
  76         'url': 'https://www.liveleak.com/view?i=677_1439397581',
  77         'info_dict': {
  78             'id': '677_1439397581',
  79             'title': 'Fuel Depot in China Explosion caught on video',
  80         },
  81         'playlist_count': 3,
  82     }, {
  83         'url': 'https://www.liveleak.com/view?t=HvHi_1523016227',
  84         'only_matching': True,
  85     }]
  86
  87     @staticmethod
  88     def _extract_urls(webpage):
  89         return re.findall(
  90             r'<iframe[^>]+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[ift]=[\w_]+[^"]+)"',
  91             webpage)
  92
  93     def _real_extract(self, url):
  94         video_id = self._match_id(url)
  95         webpage = self._download_webpage(url, video_id)
  96
  97         video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
  98         video_description = self._og_search_description(webpage)
  99         video_uploader = self._html_search_regex(
 100             r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
 101         age_limit = int_or_none(self._search_regex(
 102             r'you confirm that you are ([0-9]+) years and over.',
 103             webpage, 'age limit', default=None))
 104         video_thumbnail = self._og_search_thumbnail(webpage)
 105
 106         entries = self._parse_html5_media_entries(url, webpage, video_id)
 107         if not entries:
 108             # Maybe an embed?
 109             embed_url = self._search_regex(
 110                 r'<iframe[^>]+src="((?:https?:)?//(?:www\.)?(?:prochan|youtube)\.com/embed[^"]+)"',
 111                 webpage, 'embed URL')
 112             return {
 113                 '_type': 'url_transparent',
 114                 'url': embed_url,
 115                 'id': video_id,
 116                 'title': video_title,
 117                 'description': video_description,
 118                 'uploader': video_uploader,
 119                 'age_limit': age_limit,
 120             }
 121
 122         for idx, info_dict in enumerate(entries):
 123             formats = []
 124             for a_format in info_dict['formats']:
 125                 if not a_format.get('height'):
 126                     a_format['height'] = int_or_none(self._search_regex(
 127                         r'([0-9]+)p\.mp4', a_format['url'], 'height label',
 128                         default=None))
 129                 formats.append(a_format)
 130
 131                 # Removing '.*.mp4' gives the raw video, which is essentially
 132                 # the same video without the LiveLeak logo at the top (see
 133                 # https://github.com/rg3/youtube-dl/pull/4768)
 134                 orig_url = re.sub(r'\.mp4\.[^.]+', '', a_format['url'])
 135                 if a_format['url'] != orig_url:
 136                     format_id = a_format.get('format_id')
 137                     formats.append({
 138                         'format_id': 'original' + ('-' + format_id if format_id else ''),
 139                         'url': orig_url,
 140                         'preference': 1,
 141                     })
 142             self._sort_formats(formats)
 143             info_dict['formats'] = formats
 144
 145             # Don't append entry ID for one-video pages to keep backward compatibility
 146             if len(entries) > 1:
 147                 info_dict['id'] = '%s_%s' % (video_id, idx + 1)
 148             else:
 149                 info_dict['id'] = video_id
 150
 151             info_dict.update({
 152                 'title': video_title,
 153                 'description': video_description,
 154                 'uploader': video_uploader,
 155                 'age_limit': age_limit,
 156                 'thumbnail': video_thumbnail,
 157             })
 158
 159         return self.playlist_result(entries, video_id, video_title)
 160
 161
 162 class LiveLeakEmbedIE(InfoExtractor):
 163     _VALID_URL = r'https?://(?:www\.)?liveleak\.com/ll_embed\?.*?\b(?P<kind>[ift])=(?P<id>[\w_]+)'
 164
 165     # See generic.py for actual test cases
 166     _TESTS = [{
 167         'url': 'https://www.liveleak.com/ll_embed?i=874_1459135191',
 168         'only_matching': True,
 169     }, {
 170         'url': 'https://www.liveleak.com/ll_embed?f=ab065df993c1',
 171         'only_matching': True,
 172     }]
 173
 174     def _real_extract(self, url):
 175         kind, video_id = re.match(self._VALID_URL, url).groups()
 176
 177         if kind == 'f':
 178             webpage = self._download_webpage(url, video_id)
 179             liveleak_url = self._search_regex(
 180                 r'(?:logourl\s*:\s*|window\.open\()(?P<q1>[\'"])(?P<url>%s)(?P=q1)' % LiveLeakIE._VALID_URL,
 181                 webpage, 'LiveLeak URL', group='url')
 182         else:
 183             liveleak_url = 'http://www.liveleak.com/view?%s=%s' % (kind, video_id)
 184
 185         return self.url_result(liveleak_url, ie=LiveLeakIE.ie_key())