yt-dlc/youtube_dl/extractor/liveleak.py

170 lines
6.8 KiB
Python
Raw Normal View History

2014-01-28 21:57:38 +01:00
from __future__ import unicode_literals
2013-06-23 22:23:19 +02:00
import re
from .common import InfoExtractor
from ..utils import int_or_none
2013-06-23 22:23:19 +02:00
class LiveLeakIE(InfoExtractor):
2015-01-23 17:09:26 +01:00
_VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<id>[\w_]+)(?:.*)'
2014-01-30 06:09:06 +01:00
_TESTS = [{
2014-01-28 21:57:38 +01:00
'url': 'http://www.liveleak.com/view?i=757_1364311680',
2017-05-12 13:15:33 +02:00
'md5': '0813c2430bea7a46bf13acf3406992f4',
2014-01-28 21:57:38 +01:00
'info_dict': {
'id': '757_1364311680',
2017-05-12 13:15:33 +02:00
'ext': 'mp4',
2014-01-28 21:57:38 +01:00
'description': 'extremely bad day for this guy..!',
'uploader': 'ljfriel2',
'title': 'Most unlucky car accident',
'thumbnail': r're:^https?://.*\.jpg$'
2013-06-27 20:46:46 +02:00
}
2014-11-23 21:20:46 +01:00
}, {
2014-01-30 06:09:06 +01:00
'url': 'http://www.liveleak.com/view?i=f93_1390833151',
2017-05-12 13:15:33 +02:00
'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf',
2014-01-30 06:09:06 +01:00
'info_dict': {
'id': 'f93_1390833151',
'ext': 'mp4',
2014-01-30 06:09:06 +01:00
'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.',
'uploader': 'ARD_Stinkt',
'title': 'German Television does first Edward Snowden Interview (ENGLISH)',
'thumbnail': r're:^https?://.*\.jpg$'
2014-01-30 06:09:06 +01:00
}
2014-11-23 21:20:46 +01:00
}, {
2017-05-12 13:15:33 +02:00
# Prochan embed
'url': 'http://www.liveleak.com/view?i=4f7_1392687779',
'md5': '42c6d97d54f1db107958760788c5f48f',
'info_dict': {
'id': '4f7_1392687779',
'ext': 'mp4',
'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.",
'uploader': 'CapObveus',
'title': 'Man is Fatally Struck by Reckless Car While Packing up a Moving Truck',
'age_limit': 18,
2017-05-12 13:15:33 +02:00
},
'skip': 'Video is dead',
}, {
2015-06-14 22:54:49 +02:00
# Covers https://github.com/rg3/youtube-dl/pull/5983
2017-05-12 13:15:33 +02:00
# Multiple resolutions
'url': 'http://www.liveleak.com/view?i=801_1409392012',
2017-05-12 13:15:33 +02:00
'md5': 'c3a449dbaca5c0d1825caecd52a57d7b',
'info_dict': {
'id': '801_1409392012',
'ext': 'mp4',
2016-02-14 10:37:17 +01:00
'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.',
'uploader': 'bony333',
'title': 'Crazy Hungarian tourist films close call waterspout in Croatia',
'thumbnail': r're:^https?://.*\.jpg$'
}
}, {
# Covers https://github.com/rg3/youtube-dl/pull/10664#issuecomment-247439521
2016-11-29 17:42:19 +01:00
'url': 'http://m.liveleak.com/view?i=763_1473349649',
'add_ie': ['Youtube'],
'info_dict': {
'id': '763_1473349649',
'ext': 'mp4',
'title': 'Reporters and public officials ignore epidemic of black on asian violence in Sacramento | Colin Flaherty',
'description': 'Colin being the warrior he is and showing the injustice Asians in Sacramento are being subjected to.',
'uploader': 'Ziz',
'upload_date': '20160908',
'uploader_id': 'UCEbta5E_jqlZmEJsriTEtnw'
2016-11-29 17:42:19 +01:00
},
'params': {
'skip_download': True,
},
}, {
'url': 'https://www.liveleak.com/view?i=677_1439397581',
'info_dict': {
'id': '677_1439397581',
'title': 'Fuel Depot in China Explosion caught on video',
},
'playlist_count': 3,
2014-01-30 06:09:06 +01:00
}]
2013-06-23 22:23:19 +02:00
@staticmethod
def _extract_urls(webpage):
return re.findall(
r'<iframe[^>]+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[if]=[\w_]+[^"]+)"',
webpage)
2013-06-23 22:23:19 +02:00
def _real_extract(self, url):
2015-01-23 17:09:26 +01:00
video_id = self._match_id(url)
2013-06-23 22:23:19 +02:00
webpage = self._download_webpage(url, video_id)
video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
video_description = self._og_search_description(webpage)
video_uploader = self._html_search_regex(
r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
age_limit = int_or_none(self._search_regex(
r'you confirm that you are ([0-9]+) years and over.',
webpage, 'age limit', default=None))
video_thumbnail = self._og_search_thumbnail(webpage)
2017-05-12 13:15:33 +02:00
entries = self._parse_html5_media_entries(url, webpage, video_id)
if not entries:
# Maybe an embed?
embed_url = self._search_regex(
r'<iframe[^>]+src="((?:https?:)?//(?:www\.)?(?:prochan|youtube)\.com/embed[^"]+)"',
webpage, 'embed URL')
return {
'_type': 'url_transparent',
'url': embed_url,
'id': video_id,
'title': video_title,
'description': video_description,
'uploader': video_uploader,
'age_limit': age_limit,
}
2014-01-30 06:09:06 +01:00
for idx, info_dict in enumerate(entries):
for a_format in info_dict['formats']:
if not a_format.get('height'):
a_format['height'] = int_or_none(self._search_regex(
r'([0-9]+)p\.mp4', a_format['url'], 'height label',
default=None))
self._sort_formats(info_dict['formats'])
# Don't append entry ID for one-video pages to keep backward compatibility
if len(entries) > 1:
info_dict['id'] = '%s_%s' % (video_id, idx + 1)
else:
info_dict['id'] = video_id
2013-06-23 22:23:19 +02:00
info_dict.update({
'title': video_title,
'description': video_description,
'uploader': video_uploader,
'age_limit': age_limit,
'thumbnail': video_thumbnail,
})
2017-05-12 13:15:33 +02:00
return self.playlist_result(entries, video_id, video_title)
class LiveLeakEmbedIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?liveleak\.com/ll_embed\?.*?\b(?P<kind>[if])=(?P<id>[\w_]+)'
# See generic.py for actual test cases
_TESTS = [{
'url': 'https://www.liveleak.com/ll_embed?i=874_1459135191',
'only_matching': True,
}, {
'url': 'https://www.liveleak.com/ll_embed?f=ab065df993c1',
'only_matching': True,
}]
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
kind, video_id = mobj.group('kind', 'id')
if kind == 'f':
webpage = self._download_webpage(url, video_id)
liveleak_url = self._search_regex(
r'logourl\s*:\s*(?P<q1>[\'"])(?P<url>%s)(?P=q1)' % LiveLeakIE._VALID_URL,
webpage, 'LiveLeak URL', group='url')
elif kind == 'i':
liveleak_url = 'http://www.liveleak.com/view?i=%s' % video_id
return self.url_result(liveleak_url, ie=LiveLeakIE.ie_key())