mirror of
https://github.com/ytdl-org/youtube-dl
synced 2024-11-09 14:17:41 +01:00
5e1912cfc1
Previously, other extractor would go call a private(!) helper method. Instead, just hardcode the 5min:video_id format - it's not if that would ever change.
42 lines
1.1 KiB
Python
42 lines
1.1 KiB
Python
from __future__ import unicode_literals
|
|
|
|
import re
|
|
|
|
from .common import InfoExtractor
|
|
from ..utils import (
|
|
url_basename,
|
|
)
|
|
|
|
|
|
class EngadgetIE(InfoExtractor):
|
|
_VALID_URL = r'''(?x)https?://www.engadget.com/
|
|
(?:video/5min/(?P<id>\d+)|
|
|
[\d/]+/.*?)
|
|
'''
|
|
|
|
_TEST = {
|
|
'url': 'http://www.engadget.com/video/5min/518153925/',
|
|
'md5': 'c6820d4828a5064447a4d9fc73f312c9',
|
|
'info_dict': {
|
|
'id': '518153925',
|
|
'ext': 'mp4',
|
|
'title': 'Samsung Galaxy Tab Pro 8.4 Review',
|
|
},
|
|
'add_ie': ['FiveMin'],
|
|
}
|
|
|
|
def _real_extract(self, url):
|
|
video_id = self._match_id(url)
|
|
|
|
if video_id is not None:
|
|
return self.url_result('5min:%s' % video_id)
|
|
else:
|
|
title = url_basename(url)
|
|
webpage = self._download_webpage(url, title)
|
|
ids = re.findall(r'<iframe[^>]+?playList=(\d+)', webpage)
|
|
return {
|
|
'_type': 'playlist',
|
|
'title': title,
|
|
'entries': [self.url_result('5min:%s' % vid) for vid in ids]
|
|
}
|