diff options
| author | Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com> | 2013-10-04 14:07:29 +0200 |
|---|---|---|
| committer | Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com> | 2013-10-04 14:07:29 +0200 |
| commit | c3fef636b55f30f58a1bd9c0c2a80e20365eee7a (patch) | |
| tree | c77a3230d474f5a6919fd84e89d36d8871ef519d | |
| parent | 46e28a84caae4820c7352232b048ecb3cbcd4012 (diff) | |
[dailymotion] Fix playlist extraction
The html code has changed, make the video ids extraction more solid.
| -rw-r--r-- | youtube_dl/extractor/dailymotion.py | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 259806f38..7d8353946 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -10,6 +10,7 @@ from ..utils import ( compat_str, get_element_by_attribute, get_element_by_id, + orderedSet, ExtractorError, ) @@ -158,12 +159,12 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): id, u'Downloading page %s' % pagenum) playlist_el = get_element_by_attribute(u'class', u'video_list', webpage) - video_ids.extend(re.findall(r'data-id="(.+?)" data-ext-id', playlist_el)) + video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el)) if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: break return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion') - for video_id in video_ids] + for video_id in orderedSet(video_ids)] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) |