youtube-dl

Another place where youtube-dl lives on
git clone git://git.oshgnacknak.de/youtube-dl.git
Log | Files | Refs | README | LICENSE

liveleak.py (8015B)


      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/ytdl-org/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/ytdl-org/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         # No original video
     87         'url': 'https://www.liveleak.com/view?t=C26ZZ_1558612804',
     88         'only_matching': True,
     89     }]
     90 
     91     @staticmethod
     92     def _extract_urls(webpage):
     93         return re.findall(
     94             r'<iframe[^>]+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[ift]=[\w_]+[^"]+)"',
     95             webpage)
     96 
     97     def _real_extract(self, url):
     98         video_id = self._match_id(url)
     99         webpage = self._download_webpage(url, video_id)
    100 
    101         video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
    102         video_description = self._og_search_description(webpage)
    103         video_uploader = self._html_search_regex(
    104             r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
    105         age_limit = int_or_none(self._search_regex(
    106             r'you confirm that you are ([0-9]+) years and over.',
    107             webpage, 'age limit', default=None))
    108         video_thumbnail = self._og_search_thumbnail(webpage)
    109 
    110         entries = self._parse_html5_media_entries(url, webpage, video_id)
    111         if not entries:
    112             # Maybe an embed?
    113             embed_url = self._search_regex(
    114                 r'<iframe[^>]+src="((?:https?:)?//(?:www\.)?(?:prochan|youtube)\.com/embed[^"]+)"',
    115                 webpage, 'embed URL')
    116             return {
    117                 '_type': 'url_transparent',
    118                 'url': embed_url,
    119                 'id': video_id,
    120                 'title': video_title,
    121                 'description': video_description,
    122                 'uploader': video_uploader,
    123                 'age_limit': age_limit,
    124             }
    125 
    126         for idx, info_dict in enumerate(entries):
    127             formats = []
    128             for a_format in info_dict['formats']:
    129                 if not a_format.get('height'):
    130                     a_format['height'] = int_or_none(self._search_regex(
    131                         r'([0-9]+)p\.mp4', a_format['url'], 'height label',
    132                         default=None))
    133                 formats.append(a_format)
    134 
    135                 # Removing '.*.mp4' gives the raw video, which is essentially
    136                 # the same video without the LiveLeak logo at the top (see
    137                 # https://github.com/ytdl-org/youtube-dl/pull/4768)
    138                 orig_url = re.sub(r'\.mp4\.[^.]+', '', a_format['url'])
    139                 if a_format['url'] != orig_url:
    140                     format_id = a_format.get('format_id')
    141                     format_id = 'original' + ('-' + format_id if format_id else '')
    142                     if self._is_valid_url(orig_url, video_id, format_id):
    143                         formats.append({
    144                             'format_id': format_id,
    145                             'url': orig_url,
    146                             'preference': 1,
    147                         })
    148             self._sort_formats(formats)
    149             info_dict['formats'] = formats
    150 
    151             # Don't append entry ID for one-video pages to keep backward compatibility
    152             if len(entries) > 1:
    153                 info_dict['id'] = '%s_%s' % (video_id, idx + 1)
    154             else:
    155                 info_dict['id'] = video_id
    156 
    157             info_dict.update({
    158                 'title': video_title,
    159                 'description': video_description,
    160                 'uploader': video_uploader,
    161                 'age_limit': age_limit,
    162                 'thumbnail': video_thumbnail,
    163             })
    164 
    165         return self.playlist_result(entries, video_id, video_title)
    166 
    167 
    168 class LiveLeakEmbedIE(InfoExtractor):
    169     _VALID_URL = r'https?://(?:www\.)?liveleak\.com/ll_embed\?.*?\b(?P<kind>[ift])=(?P<id>[\w_]+)'
    170 
    171     # See generic.py for actual test cases
    172     _TESTS = [{
    173         'url': 'https://www.liveleak.com/ll_embed?i=874_1459135191',
    174         'only_matching': True,
    175     }, {
    176         'url': 'https://www.liveleak.com/ll_embed?f=ab065df993c1',
    177         'only_matching': True,
    178     }]
    179 
    180     def _real_extract(self, url):
    181         kind, video_id = re.match(self._VALID_URL, url).groups()
    182 
    183         if kind == 'f':
    184             webpage = self._download_webpage(url, video_id)
    185             liveleak_url = self._search_regex(
    186                 r'(?:logourl\s*:\s*|window\.open\()(?P<q1>[\'"])(?P<url>%s)(?P=q1)' % LiveLeakIE._VALID_URL,
    187                 webpage, 'LiveLeak URL', group='url')
    188         else:
    189             liveleak_url = 'http://www.liveleak.com/view?%s=%s' % (kind, video_id)
    190 
    191         return self.url_result(liveleak_url, ie=LiveLeakIE.ie_key())