youtube-dl

commit 30e6161799dfdf9f53d3c8eaa9e10afe615bc5dd
parent dc24a7d4a2210d6bf282e234e49193d204c41124
Author: Sergey M․ <dstftw@gmail.com>
Date:   Thu, 26 Oct 2017 23:16:16 +0700

[soundgasm] Improve extraction (closes #14588)

Diffstat:
M
youtube_dl/extractor/soundgasm.py
 | 
35
++++++++++++++++++++++++-----------

1 file changed, 24 insertions(+), 11 deletions(-)
diff --git a/youtube_dl/extractor/soundgasm.py b/youtube_dl/extractor/soundgasm.py
@@ -8,36 +8,49 @@ from .common import InfoExtractor
 
 class SoundgasmIE(InfoExtractor):
     IE_NAME = 'soundgasm'
-    _VALID_URL = r'https?://(?:www\.)?soundgasm\.net/u/(?P<user>[0-9a-zA-Z_\-]+)/(?P<title>[0-9a-zA-Z_\-]+)'
+    _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(r'\/|\.', 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'),
         }