aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorJaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>2013-07-20 19:33:40 +0200
committerJaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>2013-07-20 19:33:40 +0200
commitd7ae0639b4e43e9ac9eebb5d09cfcb77bd0671ce (patch)
tree56eb015bec6df52d9c1875f2f6146597fedc7005
parent038243599072cfc33519ffb8e6cacc8c1605de92 (diff)
downloadyoutube-dl-d7ae0639b4e43e9ac9eebb5d09cfcb77bd0671ce.tar.xz
[youtube] Add an extractor for Youtube recommended videos (":ytrec" keyword) (closes #476)
The new extractor and YoutubeSubscriptionsIE are subclasses of YoutubeFeedsInfoExtractor, which allows to fetch videos from http://www.youtube.com/feed_ajax
-rw-r--r--youtube_dl/extractor/__init__.py1
-rw-r--r--youtube_dl/extractor/youtube.py39
2 files changed, 31 insertions, 9 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index b4a1c20e9..6de48c341 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -92,6 +92,7 @@ from .youtube import (
YoutubeChannelIE,
YoutubeShowIE,
YoutubeSubscriptionsIE,
+ YoutubeRecommendedIE,
)
from .zdf import ZDFIE
diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 12e8fc25d..0d49c0721 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -898,12 +898,12 @@ class YoutubeShowIE(InfoExtractor):
return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
-class YoutubeSubscriptionsIE(YoutubeIE):
- """It's a subclass of YoutubeIE because we need to login"""
- IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
- _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
- IE_NAME = u'youtube:subscriptions'
- _FEED_TEMPLATE = 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=subscriptions&paging=%s'
+class YoutubeFeedsInfoExtractor(YoutubeIE):
+ """
+ Base class for extractors that fetch info from
+ http://www.youtube.com/feed_ajax
+ Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
+ """
_PAGING_STEP = 30
# Overwrite YoutubeIE properties we don't want
@@ -912,18 +912,27 @@ class YoutubeSubscriptionsIE(YoutubeIE):
def suitable(cls, url):
return re.match(cls._VALID_URL, url) is not None
+ @property
+ def _FEED_TEMPLATE(self):
+ return 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=%s&paging=%%s' % self._FEED_NAME
+
+ @property
+ def IE_NAME(self):
+ return u'youtube:%s' % self._FEED_NAME
+
def _real_initialize(self):
(username, password) = self._get_login_info()
if username is None:
raise ExtractorError(u'No login info available, needed for downloading the Youtube subscriptions.', expected=True)
- super(YoutubeSubscriptionsIE, self)._real_initialize()
+ super(YoutubeFeedsInfoExtractor, self)._real_initialize()
def _real_extract(self, url):
feed_entries = []
# The step argument is available only in 2.7 or higher
for i in itertools.count(0):
paging = i*self._PAGING_STEP
- info = self._download_webpage(self._FEED_TEMPLATE % paging, 'feed',
+ info = self._download_webpage(self._FEED_TEMPLATE % paging,
+ u'%s feed' % self._FEED_NAME,
u'Downloading page %s' % i)
info = json.loads(info)
feed_html = info['feed_html']
@@ -932,4 +941,16 @@ class YoutubeSubscriptionsIE(YoutubeIE):
feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
if info['paging'] is None:
break
- return self.playlist_result(feed_entries, playlist_title='Youtube Subscriptions')
+ return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
+
+class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
+ IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
+ _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
+ _FEED_NAME = 'subscriptions'
+ _PLAYLIST_TITLE = u'Youtube Subscriptions'
+
+class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
+ IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
+ _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
+ _FEED_NAME = 'recommended'
+ _PLAYLIST_TITLE = u'Youtube Recommended videos'