[generic/funnyordie] Add support for funnyordie embeds (Fixes #2546)

This commit is contained in:
Philipp Hagemeister 2014-03-11 16:51:36 +01:00
parent db95dc13a1
commit 60cc4dc4b4
2 changed files with 33 additions and 3 deletions

View File

@ -1,12 +1,13 @@
from __future__ import unicode_literals
import json
import re
from .common import InfoExtractor
class FunnyOrDieIE(InfoExtractor):
_VALID_URL = r'^(?:https?://)?(?:www\.)?funnyordie\.com/videos/(?P<id>[0-9a-f]+)/.*$'
_VALID_URL = r'https?://(?:www\.)?funnyordie\.com/(?P<type>embed|videos)/(?P<id>[0-9a-f]+)(?:$|[?#/])'
_TEST = {
'url': 'http://www.funnyordie.com/videos/0732f586d7/heart-shaped-box-literal-video-version',
'file': '0732f586d7.mp4',
@ -30,10 +31,20 @@ def _real_extract(self, url):
[r'type="video/mp4" src="(.*?)"', r'src="([^>]*?)" type=\'video/mp4\''],
webpage, 'video URL', flags=re.DOTALL)
if mobj.group('type') == 'embed':
post_json = self._search_regex(
r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details')
post = json.loads(post_json)['attachment']
title = post['name']
description = post.get('description')
else:
title = self._og_search_title(webpage)
description = self._og_search_description(webpage)
return {
'id': video_id,
'url': video_url,
'ext': 'mp4',
'title': self._og_search_title(webpage),
'description': self._og_search_description(webpage),
'title': title,
'description': description,
}

View File

@ -134,6 +134,17 @@ class GenericIE(InfoExtractor):
'skip_download': True,
},
},
# funnyordie embed
{
'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
'md5': '7cf780be104d40fea7bae52eed4a470e',
'info_dict': {
'id': '18e820ec3f',
'ext': 'mp4',
'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
}
},
]
def report_download_webpage(self, video_id):
@ -432,6 +443,14 @@ def _real_extract(self, url):
if mobj is not None:
return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
# Look for funnyordie embed
matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
if matches:
urlrs = [self.url_result(unescapeHTML(eurl), 'FunnyOrDie')
for eurl in matches]
return self.playlist_result(
urlrs, playlist_id=video_id, playlist_title=video_title)
# Start with something easy: JW Player in SWFObject
mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
if mobj is None: