diff options
| author | Yen Chi Hsuan <yan12125@gmail.com> | 2017-02-25 01:40:12 +0800 |
|---|---|---|
| committer | Yen Chi Hsuan <yan12125@gmail.com> | 2017-02-26 00:15:49 +0800 |
| commit | 9d0c08a02c55136221621e8b8dda5860211af8dd (patch) | |
| tree | 30941d71a2e785185b8d15b5b280cc5af196be61 | |
| parent | e498758b9cfa983d05b10a185ecdef480a93cf30 (diff) | |
| download | youtube-dl-9d0c08a02c55136221621e8b8dda5860211af8dd.tar.xz | |
[vevo] Fix videos with the new streams/streamsV3 format (closes #11719)
| -rw-r--r-- | ChangeLog | 2 | ||||
| -rw-r--r-- | youtube_dl/extractor/vevo.py | 22 |
2 files changed, 19 insertions, 5 deletions
@@ -1,6 +1,8 @@ version <unreleased> Extractors +* [vevo] Fix extraction for videos with the new streams/streamsV3 format + (#11719) + [njpwworld] Add new extractor (#11561) diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index c4e37f694..9aa38bc5a 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -17,12 +17,12 @@ from ..utils import ( class VevoBaseIE(InfoExtractor): - def _extract_json(self, webpage, video_id, item): + def _extract_json(self, webpage, video_id): return self._parse_json( self._search_regex( r'window\.__INITIAL_STORE__\s*=\s*({.+?});\s*</script>', webpage, 'initial store'), - video_id)['default'][item] + video_id) class VevoIE(VevoBaseIE): @@ -139,6 +139,11 @@ class VevoIE(VevoBaseIE): # no genres available 'url': 'http://www.vevo.com/watch/INS171400764', 'only_matching': True, + }, { + # Another case available only via the webpage; using streams/streamsV3 formats + # Geo-restricted to Netherlands/Germany + 'url': 'http://www.vevo.com/watch/boostee/pop-corn-clip-officiel/FR1A91600909', + 'only_matching': True, }] _VERSIONS = { 0: 'youtube', # only in AuthenticateVideo videoVersions @@ -193,7 +198,14 @@ class VevoIE(VevoBaseIE): # https://github.com/rg3/youtube-dl/issues/9366) if not video_versions: webpage = self._download_webpage(url, video_id) - video_versions = self._extract_json(webpage, video_id, 'streams')[video_id][0] + json_data = self._extract_json(webpage, video_id) + if 'streams' in json_data.get('default', {}): + video_versions = json_data['default']['streams'][video_id][0] + else: + video_versions = [ + value + for key, value in json_data['apollo']['data'].items() + if key.startswith('%s.streams' % video_id)] uploader = None artist = None @@ -207,7 +219,7 @@ class VevoIE(VevoBaseIE): formats = [] for video_version in video_versions: - version = self._VERSIONS.get(video_version['version']) + version = self._VERSIONS.get(video_version.get('version'), 'generic') version_url = video_version.get('url') if not version_url: continue @@ -339,7 +351,7 @@ class VevoPlaylistIE(VevoBaseIE): if video_id: return self.url_result('vevo:%s' % video_id, VevoIE.ie_key()) - playlists = self._extract_json(webpage, playlist_id, '%ss' % playlist_kind) + playlists = self._extract_json(webpage, playlist_id)['default']['%ss' % playlist_kind] playlist = (list(playlists.values())[0] if playlist_kind == 'playlist' else playlists[playlist_id]) |