Browse Source

[openload] Improve ext extraction

Enes 7 years ago
parent
commit
85750f8972
3 changed files with 7 additions and 3 deletions
  1. 1 0
      test/test_utils.py
  2. 5 2
      youtube_dl/extractor/openload.py
  3. 1 1
      youtube_dl/utils.py

+ 1 - 0
test/test_utils.py

@@ -361,6 +361,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(determine_ext('http://example.com/foo/bar.nonext/?download', None), None)
         self.assertEqual(determine_ext('http://example.com/foo/bar/mp4?download', None), None)
         self.assertEqual(determine_ext('http://example.com/foo/bar.m3u8//?download'), 'm3u8')
+        self.assertEqual(determine_ext('foobar', None), None)
 
     def test_find_xpath_attr(self):
         testxml = '''<root>

+ 5 - 2
youtube_dl/extractor/openload.py

@@ -307,6 +307,10 @@ class OpenloadIE(InfoExtractor):
     }, {
         'url': 'https://oload.download/f/kUEfGclsU9o',
         'only_matching': True,
+    }, {
+        # Its title has not got its extension but url has it
+        'url': 'https://oload.download/f/N4Otkw39VCw/Tomb.Raider.2018.HDRip.XviD.AC3-EVO.avi.mp4',
+        'only_matching': True,
     }]
 
     _USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'
@@ -368,8 +372,7 @@ class OpenloadIE(InfoExtractor):
             'title': title,
             'thumbnail': entry.get('thumbnail') or self._og_search_thumbnail(webpage, default=None),
             'url': video_url,
-            # Seems all videos have extensions in their titles
-            'ext': determine_ext(title, 'mp4'),
+            'ext': determine_ext(title, None) or determine_ext(url, 'mp4'),
             'subtitles': subtitles,
             'http_headers': headers,
         }

+ 1 - 1
youtube_dl/utils.py

@@ -1228,7 +1228,7 @@ def unified_timestamp(date_str, day_first=True):
 
 
 def determine_ext(url, default_ext='unknown_video'):
-    if url is None:
+    if url is None or '.' not in url:
         return default_ext
     guess = url.partition('?')[0].rpartition('.')[2]
     if re.match(r'^[A-Za-z0-9]+$', guess):