[ted] Fix playlist extraction and add a test

This commit is contained in:
Jaime Marquínez Ferrándiz 2014-03-05 13:22:10 +01:00
parent 6dadaa9930
commit ca1fee34f2
2 changed files with 28 additions and 19 deletions

View File

@ -36,6 +36,7 @@ from youtube_dl.extractor import (
RutubeChannelIE, RutubeChannelIE,
GoogleSearchIE, GoogleSearchIE,
GenericIE, GenericIE,
TEDIE,
) )
@ -259,5 +260,14 @@ class TestPlaylists(unittest.TestCase):
self.assertEqual(result['title'], 'Zero Punctuation') self.assertEqual(result['title'], 'Zero Punctuation')
self.assertTrue(len(result['entries']) > 10) self.assertTrue(len(result['entries']) > 10)
def test_ted_playlist(self):
dl = FakeYDL()
ie = TEDIE(dl)
result = ie.extract('http://www.ted.com/playlists/who_are_the_hackers')
self.assertIsPlaylist(result)
self.assertEqual(result['id'], '10')
self.assertEqual(result['title'], 'Who are the hackers?')
self.assertTrue(len(result['entries']) >= 6)
if __name__ == '__main__': if __name__ == '__main__':
unittest.main() unittest.main()

View File

@ -6,6 +6,7 @@ import re
from .subtitles import SubtitlesInfoExtractor from .subtitles import SubtitlesInfoExtractor
from ..utils import ( from ..utils import (
compat_str,
RegexNotFoundError, RegexNotFoundError,
) )
@ -13,7 +14,7 @@ from ..utils import (
class TEDIE(SubtitlesInfoExtractor): class TEDIE(SubtitlesInfoExtractor):
_VALID_URL=r'''(?x)http://www\.ted\.com/ _VALID_URL=r'''(?x)http://www\.ted\.com/
( (
((?P<type_playlist>playlists)/(?P<playlist_id>\d+)) # We have a playlist (?P<type_playlist>playlists(?:/\d+)?) # We have a playlist
| |
((?P<type_talk>talks)) # We have a simple talk ((?P<type_talk>talks)) # We have a simple talk
) )
@ -37,35 +38,35 @@ class TEDIE(SubtitlesInfoExtractor):
'high': 3, 'high': 3,
} }
def _extract_info(self, webpage):
info_json = self._search_regex(r'q\("\w+.init",({.+})\)</script>', webpage, 'info json')
return json.loads(info_json)
def _real_extract(self, url): def _real_extract(self, url):
m=re.match(self._VALID_URL, url, re.VERBOSE) m=re.match(self._VALID_URL, url, re.VERBOSE)
if m.group('type_talk'): if m.group('type_talk'):
return self._talk_info(url) return self._talk_info(url)
else : else :
playlist_id=m.group('playlist_id')
name=m.group('name') name=m.group('name')
self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name)) return self._playlist_videos_info(url, name)
return [self._playlist_videos_info(url,name,playlist_id)]
def _playlist_videos_info(self, url, name, playlist_id): def _playlist_videos_info(self, url, name):
'''Returns the videos of the playlist''' '''Returns the videos of the playlist'''
webpage = self._download_webpage( webpage = self._download_webpage(url, name,
url, playlist_id, 'Downloading playlist webpage') 'Downloading playlist webpage')
matches = re.finditer( info = self._extract_info(webpage)
r'<p\s+class="talk-title[^"]*"><a\s+href="(?P<talk_url>/talks/[^"]+\.html)">[^<]*</a></p>', playlist_info = info['playlist']
webpage)
playlist_title = self._html_search_regex(r'div class="headline">\s*?<h1>\s*?<span>(.*?)</span>',
webpage, 'playlist title')
playlist_entries = [ playlist_entries = [
self.url_result(u'http://www.ted.com' + m.group('talk_url'), 'TED') self.url_result(u'http://www.ted.com/talks/' + talk['slug'], self.ie_key())
for m in matches for talk in info['talks']
] ]
return self.playlist_result( return self.playlist_result(
playlist_entries, playlist_id=playlist_id, playlist_title=playlist_title) playlist_entries,
playlist_id=compat_str(playlist_info['id']),
playlist_title=playlist_info['title'])
def _talk_info(self, url, video_id=0): def _talk_info(self, url, video_id=0):
"""Return the video for the talk in the url""" """Return the video for the talk in the url"""
@ -74,9 +75,7 @@ class TEDIE(SubtitlesInfoExtractor):
webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name) webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name)
self.report_extraction(video_name) self.report_extraction(video_name)
info_json = self._search_regex(r'"talkPage.init",({.+})\)</script>', webpage, 'info json') talk_info = self._extract_info(webpage)['talks'][0]
info = json.loads(info_json)
talk_info = info['talks'][0]
formats = [{ formats = [{
'ext': 'mp4', 'ext': 'mp4',