youtube-dl/youtube_dl/extractor/liveleak.py

64 lines
2.3 KiB
Python
Raw Normal View History

2014-01-28 21:57:38 +01:00
from __future__ import unicode_literals
import json
2013-06-23 22:23:19 +02:00
import re
from .common import InfoExtractor
class LiveLeakIE(InfoExtractor):
_VALID_URL = r'^(?:http://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_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',
'file': '757_1364311680.mp4',
'md5': '0813c2430bea7a46bf13acf3406992f4',
'info_dict': {
'description': 'extremely bad day for this guy..!',
'uploader': 'ljfriel2',
'title': 'Most unlucky car accident'
2013-06-27 20:46:46 +02:00
}
2014-01-30 06:09:06 +01:00
},
{
'url': 'http://www.liveleak.com/view?i=f93_1390833151',
'file': 'f93_1390833151.mp4',
'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf',
'info_dict': {
'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)',
}
}]
2013-06-23 22:23:19 +02:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('video_id')
webpage = self._download_webpage(url, video_id)
sources_raw = self._search_regex(
2014-01-30 06:09:06 +01:00
r'(?s)sources:\s*(\[.*?\]),', webpage, 'video URLs', default=None)
if sources_raw is None:
sources_raw = '[{ %s}]' % (
self._search_regex(r'(file: ".*?"),', webpage, 'video URL'))
sources_json = re.sub(r'\s([a-z]+):\s', r'"\1": ', sources_raw)
sources = json.loads(sources_json)
formats = [{
'format_note': s.get('label'),
'url': s['file'],
} for s in sources]
self._sort_formats(formats)
2013-06-23 22:23:19 +02:00
video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
video_description = self._og_search_description(webpage)
2014-01-28 21:57:38 +01:00
video_uploader = self._html_search_regex(
r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
2013-06-23 22:23:19 +02:00
2014-01-28 21:57:38 +01:00
return {
'id': video_id,
2013-06-23 22:23:19 +02:00
'title': video_title,
'description': video_description,
'uploader': video_uploader,
'formats': formats,
2013-06-23 22:23:19 +02:00
}