[vevo] Fix _call_api
[youtube-dl] / youtube_dl / extractor / vevo.py
index 35fcff1b2c1281b609f7350602be2215cdd4db13..35f974c4e9e6ca6c786b103016b4059fb98f7155 100644 (file)
@@ -3,7 +3,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_etree_fromstring
+from ..compat import (
+    compat_etree_fromstring,
+    compat_urlparse,
+)
 from ..utils import (
     ExtractorError,
     int_or_none,
@@ -18,7 +21,7 @@ class VevoIE(InfoExtractor):
     (currently used by MTVIE and MySpaceIE)
     '''
     _VALID_URL = r'''(?x)
-        (?:https?://www\.vevo\.com/watch/(?:[^/]+/(?:[^/]+/)?)?|
+        (?:https?://www\.vevo\.com/watch/(?!playlist|genre)(?:[^/]+/(?:[^/]+/)?)?|
            https?://cache\.vevo\.com/m/html/embed\.html\?video=|
            https?://videoplayer\.vevo\.com/embed/embedded\?videoId=|
            vevo:)
@@ -146,13 +149,13 @@ class VevoIE(InfoExtractor):
         auth_info = self._parse_json(webpage, video_id)
         self._api_url_template = self.http_scheme() + '//apiv2.vevo.com/%s?token=' + auth_info['access_token']
 
-    def _call_api(self, path, video_id, note, errnote, fatal=True):
-        return self._download_json(self._api_url_template % path, video_id, note, errnote)
+    def _call_api(self, path, *args, **kwargs):
+        return self._download_json(self._api_url_template % path, *args, **kwargs)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        json_url = 'http://videoplayer.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
+        json_url = 'http://api.vevo.com/VideoService/AuthenticateVideo?isrc=%s' % video_id
         response = self._download_json(
             json_url, video_id, 'Downloading video info', 'Unable to download info')
         video_info = response.get('video') or {}
@@ -197,8 +200,14 @@ class VevoIE(InfoExtractor):
                 if not version_url:
                     continue
 
-                if '.mpd' in version_url or '.ism' in version_url:
+                if '.ism' in version_url:
                     continue
+                elif '.mpd' in version_url:
+                    formats.extend(self._extract_mpd_formats(
+                        version_url, video_id, mpd_id='dash-%s' % version,
+                        note='Downloading %s MPD information' % version,
+                        errnote='Failed to download %s MPD information' % version,
+                        fatal=False))
                 elif '.m3u8' in version_url:
                     formats.extend(self._extract_m3u8_formats(
                         version_url, video_id, 'mp4', 'm3u8_native',
@@ -295,3 +304,72 @@ class VevoIE(InfoExtractor):
             'view_count': view_count,
             'age_limit': age_limit,
         }
+
+
+class VevoPlaylistIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.vevo\.com/watch/(?P<kind>playlist|genre)/(?P<id>[^/?#&]+)'
+
+    _TESTS = [{
+        'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29',
+        'info_dict': {
+            'id': 'dadbf4e7-b99f-4184-9670-6f0e547b6a29',
+            'title': 'Best-Of: Birdman',
+        },
+        'playlist_count': 10,
+    }, {
+        'url': 'http://www.vevo.com/watch/genre/rock',
+        'info_dict': {
+            'id': 'rock',
+            'title': 'Rock',
+        },
+        'playlist_count': 20,
+    }, {
+        'url': 'http://www.vevo.com/watch/playlist/dadbf4e7-b99f-4184-9670-6f0e547b6a29?index=0',
+        'md5': '32dcdfddddf9ec6917fc88ca26d36282',
+        'info_dict': {
+            'id': 'USCMV1100073',
+            'ext': 'mp4',
+            'title': 'Y.U. MAD',
+            'timestamp': 1323417600,
+            'upload_date': '20111209',
+            'uploader': 'Birdman',
+        },
+        'expected_warnings': ['Unable to download SMIL file'],
+    }, {
+        'url': 'http://www.vevo.com/watch/genre/rock?index=0',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+        playlist_kind = mobj.group('kind')
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
+        index = qs.get('index', [None])[0]
+
+        if index:
+            video_id = self._search_regex(
+                r'<meta[^>]+content=(["\'])vevo://video/(?P<id>.+?)\1[^>]*>',
+                webpage, 'video id', default=None, group='id')
+            if video_id:
+                return self.url_result('vevo:%s' % video_id, VevoIE.ie_key())
+
+        playlists = self._parse_json(
+            self._search_regex(
+                r'window\.__INITIAL_STORE__\s*=\s*({.+?});\s*</script>',
+                webpage, 'initial store'),
+            playlist_id)['default']['%ss' % playlist_kind]
+
+        playlist = (list(playlists.values())[0]
+                    if playlist_kind == 'playlist' else playlists[playlist_id])
+
+        entries = [
+            self.url_result('vevo:%s' % src, VevoIE.ie_key())
+            for src in playlist['isrcs']]
+
+        return self.playlist_result(
+            entries, playlist.get('playlistId'),
+            playlist.get('name'), playlist.get('description'))