X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fsoundgasm.py;h=3d78a9d76cce8aa28902111fe3f9018dafbb856a;hb=HEAD;hp=e568ff18cd7002df8c9dfa99c3bc162498833f0e;hpb=93b5071f73738d788c878b38a57f2b6efe0da883;p=youtube-dl

diff --git a/youtube_dl/extractor/soundgasm.py b/youtube_dl/extractor/soundgasm.py
index e568ff18c..3d78a9d76 100644
--- a/youtube_dl/extractor/soundgasm.py
+++ b/youtube_dl/extractor/soundgasm.py
@@ -4,73 +4,74 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import clean_html
 
 
 class SoundgasmIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?soundgasm\.net/u/(?P<user>[0-9a-zA-Z_\-]+)/(?P<title>[0-9a-zA-Z_\-]+)'
+    IE_NAME = 'soundgasm'
+    _VALID_URL = r'https?://(?:www\.)?soundgasm\.net/u/(?P<user>[0-9a-zA-Z_-]+)/(?P<display_id>[0-9a-zA-Z_-]+)'
     _TEST = {
         'url': 'http://soundgasm.net/u/ytdl/Piano-sample',
         'md5': '010082a2c802c5275bb00030743e75ad',
         'info_dict': {
             'id': '88abd86ea000cafe98f96321b23cc1206cbcbcc9',
             'ext': 'm4a',
-            'title': 'ytdl_Piano-sample',
-            'description': 'Royalty Free Sample Music'
+            'title': 'Piano sample',
+            'description': 'Royalty Free Sample Music',
+            'uploader': 'ytdl',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('title')
-        audio_title = mobj.group('user') + '_' + mobj.group('title')
+        display_id = mobj.group('display_id')
+
         webpage = self._download_webpage(url, display_id)
+
         audio_url = self._html_search_regex(
-            r'(?s)m4a\:\s"([^"]+)"', webpage, 'audio URL')
-        audio_id = re.split('\/|\.', audio_url)[-2]
+            r'(?s)m4a\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
+            'audio URL', group='url')
+
+        title = self._search_regex(
+            r'<div[^>]+\bclass=["\']jp-title[^>]+>([^<]+)',
+            webpage, 'title', default=display_id)
+
         description = self._html_search_regex(
-            r'(?s)<li>Description:\s(.*?)<\/li>', webpage, 'description',
-            fatal=False)
+            (r'(?s)<div[^>]+\bclass=["\']jp-description[^>]+>(.+?)</div>',
+             r'(?s)<li>Description:\s(.*?)<\/li>'),
+            webpage, 'description', fatal=False)
+
+        audio_id = self._search_regex(
+            r'/([^/]+)\.m4a', audio_url, 'audio id', default=display_id)
 
         return {
             'id': audio_id,
             'display_id': display_id,
             'url': audio_url,
-            'title': audio_title,
-            'description': description
+            'vcodec': 'none',
+            'title': title,
+            'description': description,
+            'uploader': mobj.group('user'),
         }
 
+
 class SoundgasmProfileIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?soundgasm\.net/u/(?P<id>[0-9a-zA-Z_\-]+)/?$'
+    IE_NAME = 'soundgasm:profile'
+    _VALID_URL = r'https?://(?:www\.)?soundgasm\.net/u/(?P<id>[^/]+)/?(?:\#.*)?$'
     _TEST = {
         'url': 'http://soundgasm.net/u/ytdl',
-        'playlist_count': 1,
         'info_dict': {
-            '_type': 'playlist',
             'id': 'ytdl',
-            'title': 'ytdl'
-        }
+        },
+        'playlist_count': 1,
     }
 
     def _real_extract(self, url):
         profile_id = self._match_id(url)
+
         webpage = self._download_webpage(url, profile_id)
 
-        ids = re.findall(r'''<a\s+href=".+?/u/%s/([^/]+)">''' % re.escape(profile_id), webpage)
-        ids = [clean_html(id) for id in ids]
-
-        entries = []
-        for id in ids:
-            entries.append({
-                '_type': 'url',
-                'url': ('http://soundgasm.net/u/%s/%s' % (profile_id, id))
-            })
-
-        info_dict = {
-            '_type': 'playlist',
-            'id': profile_id,
-            'title': profile_id,
-            'entries': entries
-        }
+        entries = [
+            self.url_result(audio_url, 'Soundgasm')
+            for audio_url in re.findall(r'href="([^"]+/u/%s/[^"]+)' % profile_id, webpage)]
 
-        return info_dict;
+        return self.playlist_result(entries, profile_id)