Update upstream source from tag 'upstream/2020.06.16.1'

[youtubedl] / youtube_dl / extractor / shared.py

diff --git a/youtube_dl/extractor/shared.py b/youtube_dl/extractor/shared.py
index ff575f592003beb91314b0101fb97b3a27a67432..02295d1a4cacf570ee903c06161b072f35a6da77 100644 (file)
--- a/youtube_dl/extractor/shared.py
+++ b/youtube_dl/extractor/shared.py
@@ -1,13 +1,18 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from __future__ import unicode_literals
 
 from .common import InfoExtractor

-from ..compat import compat_b64decode
+from ..compat import (
+    compat_b64decode,
+    compat_urllib_parse_unquote_plus,
+)

 from ..utils import (
     determine_ext,
     ExtractorError,
     int_or_none,
 from ..utils import (
     determine_ext,
     ExtractorError,
     int_or_none,

+    js_to_json,

     KNOWN_EXTENSIONS,
     parse_filesize,
     KNOWN_EXTENSIONS,
     parse_filesize,

+    rot47,

     url_or_none,
     urlencode_postdata,
 )
     url_or_none,
     urlencode_postdata,
 )


@@ -112,16 +117,22 @@ class VivoIE(SharedBaseIE):
             webpage, 'filesize', fatal=False))
 
     def _extract_video_url(self, webpage, video_id, url):
             webpage, 'filesize', fatal=False))
 
     def _extract_video_url(self, webpage, video_id, url):

-        def decode_url(encoded_url):
+        def decode_url_old(encoded_url):

             return compat_b64decode(encoded_url).decode('utf-8')
 
             return compat_b64decode(encoded_url).decode('utf-8')
 

-        stream_url = url_or_none(decode_url(self._search_regex(
+        stream_url = self._search_regex(

             r'data-stream\s*=\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
             r'data-stream\s*=\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,

-            'stream url', default=None, group='url')))
+            'stream url', default=None, group='url')
+        if stream_url:
+            stream_url = url_or_none(decode_url_old(stream_url))

         if stream_url:
             return stream_url
         if stream_url:
             return stream_url

-        return self._parse_json(
+
+        def decode_url(encoded_url):
+            return rot47(compat_urllib_parse_unquote_plus(encoded_url))
+
+        return decode_url(self._parse_json(

             self._search_regex(
             self._search_regex(

-                r'InitializeStream\s*\(\s*(["\'])(?P<url>(?:(?!\1).)+)\1',
-                webpage, 'stream', group='url'),
-            video_id, transform_source=decode_url)[0]
+                r'(?s)InitializeStream\s*\(\s*({.+?})\s*\)\s*;', webpage,
+                'stream'),
+            video_id, transform_source=js_to_json)['source'])