youtube-dl/youtube_dl/extractor/funnyordie.py

74 lines
2.5 KiB
Python
Raw Normal View History

from __future__ import unicode_literals
import json
2013-06-23 20:05:23 +00:00
import re
from .common import InfoExtractor
2014-05-07 14:02:57 +00:00
from ..utils import ExtractorError
2013-06-23 20:05:23 +00:00
class FunnyOrDieIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?funnyordie\.com/(?P<type>embed|articles|videos)/(?P<id>[0-9a-f]+)(?:$|[?#/])'
_TESTS = [{
'url': 'http://www.funnyordie.com/videos/0732f586d7/heart-shaped-box-literal-video-version',
2014-05-07 14:02:57 +00:00
'md5': 'bcd81e0c4f26189ee09be362ad6e6ba9',
'info_dict': {
'id': '0732f586d7',
'ext': 'mp4',
'title': 'Heart-Shaped Box: Literal Video Version',
'description': 'md5:ea09a01bc9a1c46d9ab696c01747c338',
'thumbnail': 're:^http:.*\.jpg$',
},
}, {
'url': 'http://www.funnyordie.com/embed/e402820827',
2014-09-29 10:20:20 +00:00
'md5': '29f4c5e5a61ca39dfd7e8348a75d0aad',
'info_dict': {
'id': 'e402820827',
'ext': 'mp4',
'title': 'Please Use This Song (Jon Lajoie)',
2014-07-21 11:14:41 +00:00
'description': 'Please use this to sell something. www.jonlajoie.com',
'thumbnail': 're:^http:.*\.jpg$',
},
2014-10-31 12:26:56 +00:00
}, {
'url': 'http://www.funnyordie.com/articles/ebf5e34fc8/10-hours-of-walking-in-nyc-as-a-man',
'only_matching': True,
}]
2013-06-23 20:05:23 +00:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
webpage = self._download_webpage(url, video_id)
links = re.findall(r'<source src="([^"]+/v)[^"]+\.([^"]+)" type=\'video', webpage)
2014-05-07 14:02:57 +00:00
if not links:
raise ExtractorError('No media links available for %s' % video_id)
links.sort(key=lambda link: 1 if link[1] == 'mp4' else 0)
bitrates = self._html_search_regex(r'<source src="[^"]+/v,((?:\d+,)+)\.mp4\.csmil', webpage, 'video bitrates')
bitrates = [int(b) for b in bitrates.rstrip(',').split(',')]
bitrates.sort()
formats = []
for bitrate in bitrates:
for link in links:
formats.append({
'url': '%s%d.%s' % (link[0], bitrate, link[1]),
'format_id': '%s-%d' % (link[1], bitrate),
'vbr': bitrate,
})
2013-06-23 20:05:23 +00:00
post_json = self._search_regex(
r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details')
post = json.loads(post_json)
return {
2013-06-23 20:05:23 +00:00
'id': video_id,
2014-05-06 08:19:13 +00:00
'title': post['name'],
'description': post.get('description'),
'thumbnail': post.get('picture'),
2014-05-07 14:02:57 +00:00
'formats': formats,
2013-06-23 20:05:23 +00:00
}