improved thumbs, formats, test result.

* now returns all formats.
 * added support for all types of thumbnails.
 * downloadtest will now succeed.
 * updated geolocking check
 * removed 30 second preview notice.

youtube_dl/extractor/deezer.py

@@ -50,34 +50,78 @@ def getformat(song):
     return 1
 
 
+class Format:
+    def __init__(self, name, fmtid, ext, bitrate):
+        self.name = name
+        self.formatid = fmtid
+        self.extension = ext
+        self.bitrate = bitrate
+
+
+_formatinfo = {
+    'MP3_MISC': Format('MP3_MISC', 0, 'mp3', 128),
+    'MP3_128': Format('MP3_128', 1, 'mp3', 128),
+    # 2 used to be MP3_192
+    'MP3_320': Format('MP3_320', 3, 'mp3', 320),
+    # 4 used to be AAC_96
+    'MP3_256': Format('MP3_256', 5, 'mp3', 256),
+    'AAC_64': Format('AAC_64', 6, 'aac', 64),   # used to be AAC_192
+    'MP3_192': Format('MP3_192', 7, 'mp3', 192),
+    'AAC_96': Format('AAC_96', 8, 'aac', 96),
+    'FLAC': Format('FLAC', 9, 'flac', 0),
+    'MP3_64': Format('MP3_64', 10, 'mp3', 64),
+    'MP3_32': Format('MP3_32', 11, 'mp3', 32),
+}
+
+
+def convert_to_int(x):
+    try:
+        return int(x)
+    except:
+        return 0
+
+
+def songformats(s):
+    """
+    Enumerate all formats for a song
+    """
+    for k, v in s.items():
+        if k.startswith("FILESIZE_") and convert_to_int(v):
+            name = k[9:]
+            if name in _formatinfo:
+                yield _formatinfo[name]
+
+
 class DeezerPlaylistIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?deezer\.com/\w+/(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?deezer\.com/(?P<type>\w+)/(?P<id>[0-9]+)'
     _TEST = {
         'url': 'http://www.deezer.com/playlist/176747451',
         'info_dict': {
             'id': '176747451',
             'title': 'Best!',
-            'uploader': 'Anonymous',
+            'uploader': 'anonymous',
-            'thumbnail': r're:^https?://cdn-images.deezer.com/images/cover/.*\.jpg$',
+            'thumbnail': r're:^https?://\S*cdn-images.deezer.com/images/playlist/.*\.jpg$',
         },
         'playlist_count': 30,
-        'skip': 'Only available in .de',
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        playlist_id = mobj.group('id')
+        list_id = mobj.group('id')
+        list_type = mobj.group('type')
 
-        webpage = self._download_webpage(url, playlist_id)
+        webpage = self._download_webpage(url, list_id)
         geoblocking_msg = self._html_search_regex(
-            r'<p class="soon-txt">(.*?)</p>', webpage, 'geoblocking message',
+            r'<div class="page-soon.*<h2>(.*?)</h2>', webpage, 'geoblocking message',
             default=None)
         if geoblocking_msg is not None:
             raise ExtractorError(
                 'Deezer said: %s' % geoblocking_msg, expected=True)
 
         host_stream_cdn = self._search_regex(
-            r'var HOST_STREAM_CDN = \'(.*?)\'', webpage, 'host stream cdn')
+            r'var\s+HOST_STREAM_CDN\s*=\s*[\'"](.*?)[\'"]', webpage, 'host stream cdn')
+        setting_domain_img = self._search_regex(
+            r'var\s+SETTING_DOMAIN_IMG\s*=\s*[\'"](.*?)[\'"]', webpage, 'setting domain img')
 
         data_json = self._search_regex(
             (r'__DZR_APP_STATE__\s*=\s*({.+?})\s*</script>',
@@ -85,39 +129,48 @@ class DeezerPlaylistIE(InfoExtractor):
             webpage, 'data JSON')
         data = json.loads(data_json)
 
-        playlist_title = data.get('DATA', {}).get('TITLE')
+        list_title = data.get('DATA', {}).get('TITLE')
-        playlist_uploader = data.get('DATA', {}).get('PARENT_USERNAME')
+        list_uploader = data.get('DATA', {}).get('PARENT_USERNAME')
-        # playlist_thumbnail = self._search_regex( r'<img id="naboo_playlist_image".*?src="([^"]+)"', webpage, 'playlist thumbnail')
+        list_thumbnail = setting_domain_img + ("/playlist/" if list_type == 'playlist' else "/cover/")
 
         entries = []
 
         if 'SONGS' in data:
             # playlist, album
             songlist = data['SONGS']['data']
+            if list_type == 'playlist':
+                listimgid = data['DATA']['PLAYLIST_PICTURE']
+            else:
+                listimgid = data['DATA']['ALB_PICTURE']
         elif 'MD5_ORIGIN' in data['DATA']:
             # track
             songlist = [data['DATA']]
+            listimgid = data['DATA']['ALB_PICTURE']
         elif "ALBUMS" in data:
+            # artist
             songlist = []
             for album in data["ALBUMS"]["data"]:
                 songlist.extend(album['SONGS']['data'])
+            listimgid = data['DATA']['ART_PICTURE']
         else:
             raise ExtractorError('Could not find songs')
 
-        # album, playlist
         for s in songlist:
 
-            urlkey = calcurlkey(int(s["SNG_ID"]), str(s["MD5_ORIGIN"]), int(s["MEDIA_VERSION"]), getformat(s))
+            formats = []
-            url = host_stream_cdn.replace('{0}', str(s["MD5_ORIGIN"])[0]) + "/" + urlkey
+            for fmt in songformats(s):
+                urlkey = calcurlkey(int(s["SNG_ID"]), str(s["MD5_ORIGIN"]), int(s["MEDIA_VERSION"]), fmt.formatid)
+                url = host_stream_cdn.replace('{0}', str(s["MD5_ORIGIN"])[0]) + "/" + urlkey
 
-            formats = [{
+                formats.append({
-                'format_id': 'preview',
+                    'format_id': fmt.name,
-                'key': calcblowfishkey(int(s["SNG_ID"])),
+                    'key': calcblowfishkey(int(s["SNG_ID"])),
-                'url': url,
+                    'url': url,
-                'preference': -100,  # Only the first 30 seconds
+                    'preference': fmt.bitrate,
-                'ext': 'mp3',
+                    'ext': fmt.extension,
-                'protocol': 'deezer',
+                    'protocol': 'deezer',
-            }]
+                    'vcodec': 'none',
+                })
             self._sort_formats(formats)
             artists = ', '.join(
                 orderedSet(a['ART_NAME'] for a in s['ARTISTS']))
@@ -127,15 +180,16 @@ class DeezerPlaylistIE(InfoExtractor):
                 'title': '%s - %s' % (artists, s['SNG_TITLE']),
                 'uploader': s['ART_NAME'],
                 'uploader_id': s['ART_ID'],
-                'age_limit': 16 if s.get('EXPLICIT_LYRICS') == '1' else 0,
+                'age_limit': 16 if convert_to_int(s.get('EXPLICIT_LYRICS')) else 0,
                 'formats': formats,
+                'thumbnail': setting_domain_img + '/cover/' + s['ALB_PICTURE'] + '/200x200.jpg'
             })
 
         return {
             '_type': 'playlist',
-            'id': playlist_id,
+            'id': list_id,
-            'title': playlist_title,
+            'title': list_title,
-            'uploader': playlist_uploader,
+            'uploader': list_uploader,
-            # 'thumbnail': playlist_thumbnail,
+            'thumbnail': list_thumbnail + listimgid + '/200x200.jpg',
             'entries': entries,
         }