diff options
author | Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com> | 2013-11-13 16:21:24 +0100 |
---|---|---|
committer | Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com> | 2013-11-13 16:26:50 +0100 |
commit | dcbb45803f9b70041ec0ef9c3c6547340bd1ef7a (patch) | |
tree | b72bf97d64f68dac6453ca2920d504a80edae030 /youtube_dl | |
parent | 80b9bbce8687f800b79edb36edf8c193dcf26a78 (diff) |
[youtube:playlist] Don't use the gdata api (closes #1508)
Parse the playlist pages instead
Diffstat (limited to 'youtube_dl')
-rw-r--r-- | youtube_dl/extractor/youtube.py | 43 |
1 files changed, 13 insertions, 30 deletions
diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index c992cba97..d97ea8c83 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -1506,8 +1506,9 @@ class YoutubePlaylistIE(InfoExtractor): | ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,}) )""" - _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none' - _MAX_RESULTS = 50 + _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s' + _MORE_PAGES_INDICATOR = r'data-link-type="next"' + _VIDEO_RE = r'href="/watch\?v=([0-9A-Za-z_-]{11})&' IE_NAME = u'youtube:playlist' @classmethod @@ -1532,41 +1533,23 @@ class YoutubePlaylistIE(InfoExtractor): else: self.to_screen(u'Downloading playlist PL%s - add --no-playlist to just download video %s' % (playlist_id, video_id)) - # Download playlist videos from API - videos = [] + # Extract the video ids from the playlist pages + ids = [] for page_num in itertools.count(1): - start_index = self._MAX_RESULTS * (page_num - 1) + 1 - if start_index >= 1000: - self._downloader.report_warning(u'Max number of results reached') - break - url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index) + url = self._TEMPLATE_URL % (playlist_id, page_num) page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) + # The ids are duplicated + new_ids = orderedSet(re.findall(self._VIDEO_RE, page)) + ids.extend(new_ids) - try: - response = json.loads(page) - except ValueError as err: - raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) - - if 'feed' not in response: - raise ExtractorError(u'Got a malformed response from YouTube API') - playlist_title = response['feed']['title']['$t'] - if 'entry' not in response['feed']: - # Number of videos is a multiple of self._MAX_RESULTS + if re.search(self._MORE_PAGES_INDICATOR, page) is None: break - for entry in response['feed']['entry']: - index = entry['yt$position']['$t'] - if 'media$group' in entry and 'yt$videoid' in entry['media$group']: - videos.append(( - index, - 'https://www.youtube.com/watch?v=' + entry['media$group']['yt$videoid']['$t'] - )) - - videos = [v[1] for v in sorted(videos)] + playlist_title = self._og_search_title(page) - url_results = [self.url_result(vurl, 'Youtube') for vurl in videos] - return [self.playlist_result(url_results, playlist_id, playlist_title)] + url_results = [self.url_result(vid, 'Youtube') for vid in ids] + return self.playlist_result(url_results, playlist_id, playlist_title) class YoutubeChannelIE(InfoExtractor): |