Fix extraction from youtube.

[youtubedl] / youtube_dl / extractor / hotnewhiphop.py

diff --git a/youtube_dl/extractor/hotnewhiphop.py b/youtube_dl/extractor/hotnewhiphop.py
index 80b48b1b3605a18fa11547ab890897b6b47ccaa5..4703e1894f08f70057ace62fc1411980ea7728b4 100644 (file)
--- a/youtube_dl/extractor/hotnewhiphop.py
+++ b/youtube_dl/extractor/hotnewhiphop.py
@@ -1,47 +1,44 @@
 from __future__ import unicode_literals
 
-import re
-import base64
-
 from .common import InfoExtractor
+from ..compat import compat_b64decode
 from ..utils import (
-    compat_urllib_parse,
-    compat_urllib_request,
     ExtractorError,
     HEADRequest,
+    sanitized_Request,
+    urlencode_postdata,
 )
 
 
 class HotNewHipHopIE(InfoExtractor):
-    _VALID_URL = r'http://www\.hotnewhiphop\.com/.*\.(?P<id>.*)\.html'
+    _VALID_URL = r'https?://(?:www\.)?hotnewhiphop\.com/.*\.(?P<id>.*)\.html'
     _TEST = {
         'url': 'http://www.hotnewhiphop.com/freddie-gibbs-lay-it-down-song.1435540.html',
-        'file': '1435540.mp3',
         'md5': '2c2cd2f76ef11a9b3b581e8b232f3d96',
         'info_dict': {
+            'id': '1435540',
+            'ext': 'mp3',
             'title': 'Freddie Gibbs - Lay It Down'
         }
     }
 
     def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('id')
-
-        webpage_src = self._download_webpage(url, video_id)
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
 
         video_url_base64 = self._search_regex(
-            r'data-path="(.*?)"', webpage_src, u'video URL', fatal=False)
+            r'data-path="(.*?)"', webpage, 'video URL', default=None)
 
         if video_url_base64 is None:
             video_url = self._search_regex(
-                r'"contentUrl" content="(.*?)"', webpage_src, u'video URL')
+                r'"contentUrl" content="(.*?)"', webpage, 'content URL')
             return self.url_result(video_url, ie='Youtube')
 
-        reqdata = compat_urllib_parse.urlencode([
+        reqdata = urlencode_postdata([
             ('mediaType', 's'),
             ('mediaId', video_id),
         ])
-        r = compat_urllib_request.Request(
+        r = sanitized_Request(
             'http://www.hotnewhiphop.com/ajax/media/getActions/', data=reqdata)
         r.add_header('Content-Type', 'application/x-www-form-urlencoded')
         mkd = self._download_json(
@@ -50,7 +47,7 @@ class HotNewHipHopIE(InfoExtractor):
         if 'mediaKey' not in mkd:
             raise ExtractorError('Did not get a media key')
 
-        redirect_url = base64.b64decode(video_url_base64).decode('utf-8')
+        redirect_url = compat_b64decode(video_url_base64).decode('utf-8')
         redirect_req = HEADRequest(redirect_url)
         req = self._request_webpage(
             redirect_req, video_id,
@@ -59,11 +56,11 @@ class HotNewHipHopIE(InfoExtractor):
         if video_url.endswith('.html'):
             raise ExtractorError('Redirect failed')
 
-        video_title = self._og_search_title(webpage_src).strip()
+        video_title = self._og_search_title(webpage).strip()
 
         return {
             'id': video_id,
             'url': video_url,
             'title': video_title,
-            'thumbnail': self._og_search_thumbnail(webpage_src),
+            'thumbnail': self._og_search_thumbnail(webpage),
         }