diff --git a/test/test_playlists.py b/test/test_playlists.py index fda2e0112..e0eb05460 100644 --- a/test/test_playlists.py +++ b/test/test_playlists.py @@ -34,6 +34,7 @@ from youtube_dl.extractor import ( KhanAcademyIE, EveryonesMixtapeIE, RutubeChannelIE, + GoogleSearchIE, GenericIE, ) @@ -240,6 +241,14 @@ class TestPlaylists(unittest.TestCase): self.assertEqual(result['title'], 'Always/Never: A Little-Seen Movie About Nuclear Command and Control : The New Yorker') self.assertEqual(len(result['entries']), 3) + def test_GoogleSearch(self): + dl = FakeYDL() + ie = GoogleSearchIE(dl) + result = ie.extract('gvsearch15:python language') + self.assertIsPlaylist(result) + self.assertEqual(result['id'], 'python language') + self.assertEqual(result['title'], 'python language') + self.assertTrue(len(result['entries']) == 15) if __name__ == '__main__': unittest.main() diff --git a/youtube_dl/extractor/googlesearch.py b/youtube_dl/extractor/googlesearch.py index f9c88e9b5..5c2564270 100644 --- a/youtube_dl/extractor/googlesearch.py +++ b/youtube_dl/extractor/googlesearch.py @@ -1,3 +1,5 @@ +from __future__ import unicode_literals + import itertools import re @@ -8,32 +10,42 @@ from ..utils import ( class GoogleSearchIE(SearchInfoExtractor): - IE_DESC = u'Google Video search' - _MORE_PAGES_INDICATOR = r'id="pnnext" class="pn"' + IE_DESC = 'Google Video search' _MAX_RESULTS = 1000 - IE_NAME = u'video.google:search' + IE_NAME = 'video.google:search' _SEARCH_KEY = 'gvsearch' def _get_n_results(self, query, n): """Get a specified number of results for a query""" + entries = [] res = { '_type': 'playlist', 'id': query, - 'entries': [] + 'title': query, } - for pagenum in itertools.count(1): - result_url = u'http://www.google.com/search?tbm=vid&q=%s&start=%s&hl=en' % (compat_urllib_parse.quote_plus(query), pagenum*10) - webpage = self._download_webpage(result_url, u'gvsearch:' + query, - note='Downloading result page ' + str(pagenum)) + for pagenum in itertools.count(): + result_url = ( + 'http://www.google.com/search?tbm=vid&q=%s&start=%s&hl=en' + % (compat_urllib_parse.quote_plus(query), pagenum * 10)) - for mobj in re.finditer(r'