[youtube] Fix some issues with the detection of playlist/channel urls (reported in #1374)

They were being caught by YoutubeUserIE, now it only extracts a url if the rest of extractors aren't suitable.
Now the url tests check that the urls can only be extracted with an specific extractor.
This commit is contained in:
Jaime Marquínez Ferrándiz 2013-09-06 16:24:24 +02:00
parent faab1d3836
commit e3ea479087
2 changed files with 22 additions and 13 deletions

View File

@ -21,14 +21,15 @@ class TestAllURLsMatching(unittest.TestCase):
self.assertEqual(self.matching_ies(url), ie_list)
def test_youtube_playlist_matching(self):
self.assertTrue(YoutubePlaylistIE.suitable(u'ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8'))
self.assertTrue(YoutubePlaylistIE.suitable(u'UUBABnxM4Ar9ten8Mdjj1j0Q')) #585
self.assertTrue(YoutubePlaylistIE.suitable(u'PL63F0C78739B09958'))
self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q'))
self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8'))
self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC'))
self.assertTrue(YoutubePlaylistIE.suitable(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
self.assertFalse(YoutubePlaylistIE.suitable(u'PLtS2H6bU1M'))
assertPlaylist = lambda url: self.assertMatch(url, ['youtube:playlist'])
assertPlaylist(u'ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
assertPlaylist(u'UUBABnxM4Ar9ten8Mdjj1j0Q') #585
assertPlaylist(u'PL63F0C78739B09958')
assertPlaylist(u'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
assertPlaylist(u'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
assertPlaylist(u'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
assertPlaylist(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012') #668
self.assertFalse('youtube:playlist' in self.matching_ies(u'PLtS2H6bU1M'))
def test_youtube_matching(self):
self.assertTrue(YoutubeIE.suitable(u'PLtS2H6bU1M'))
@ -37,9 +38,10 @@ class TestAllURLsMatching(unittest.TestCase):
self.assertMatch('http://www.youtube.com/v/BaW_jenozKc', ['youtube'])
def test_youtube_channel_matching(self):
self.assertTrue(YoutubeChannelIE.suitable('https://www.youtube.com/channel/HCtnHdj3df7iM'))
self.assertTrue(YoutubeChannelIE.suitable('https://www.youtube.com/channel/HCtnHdj3df7iM?feature=gb_ch_rec'))
self.assertTrue(YoutubeChannelIE.suitable('https://www.youtube.com/channel/HCtnHdj3df7iM/videos'))
assertChannel = lambda url: self.assertMatch(url, ['youtube:channel'])
assertChannel('https://www.youtube.com/channel/HCtnHdj3df7iM')
assertChannel('https://www.youtube.com/channel/HCtnHdj3df7iM?feature=gb_ch_rec')
assertChannel('https://www.youtube.com/channel/HCtnHdj3df7iM/videos')
def test_youtube_user_matching(self):
self.assertMatch('www.youtube.com/NASAgovVideo/videos', ['youtube:user'])
@ -50,6 +52,9 @@ class TestAllURLsMatching(unittest.TestCase):
self.assertMatch('https://www.youtube.com/feed/recommended', ['youtube:recommended'])
self.assertMatch('https://www.youtube.com/my_favorites', ['youtube:favorites'])
def test_youtube_show_matching(self):
self.assertMatch('http://www.youtube.com/show/airdisasters', ['youtube:show'])
def test_justin_tv_channelid_matching(self):
self.assertTrue(JustinTVIE.suitable(u"justin.tv/vanillatv"))
self.assertTrue(JustinTVIE.suitable(u"twitch.tv/vanillatv"))

View File

@ -386,7 +386,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
@classmethod
def suitable(cls, url):
"""Receives a URL and returns True if suitable for this IE."""
if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False
if YoutubePlaylistIE.suitable(url): return False
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
def report_video_webpage_download(self, video_id):
@ -1021,8 +1021,12 @@ class YoutubeUserIE(InfoExtractor):
_GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
IE_NAME = u'youtube:user'
@classmethod
def suitable(cls, url):
if YoutubeIE.suitable(url) or YoutubeFavouritesIE.suitable(url): return False
# Don't return True if the url can be extracted with other youtube
# extractor, the regex would is too permissive and it would match.
other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
if any(ie.suitable(url) for ie in other_ies): return False
else: return super(YoutubeUserIE, cls).suitable(url)
def _real_extract(self, url):