Update upstream source from tag 'upstream/2020.06.16'

[youtubedl] / youtube_dl / extractor / shared.py

diff --git a/youtube_dl/extractor/shared.py b/youtube_dl/extractor/shared.py
index 89e19e9277f42b69ac6f01b03c78ced00f3ec990..02295d1a4cacf570ee903c06161b072f35a6da77 100644 (file)
--- a/youtube_dl/extractor/shared.py
+++ b/youtube_dl/extractor/shared.py
@@ -1,11 +1,19 @@
 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

-import base64
-

 from .common import InfoExtractor
 from .common import InfoExtractor

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

 from ..utils import (
 from ..utils import (

+    determine_ext,

     ExtractorError,
     int_or_none,
     ExtractorError,
     int_or_none,

+    js_to_json,
+    KNOWN_EXTENSIONS,
+    parse_filesize,
+    rot47,
+    url_or_none,

     urlencode_postdata,
 )
 
     urlencode_postdata,
 )
 


@@ -22,10 +30,8 @@ class SharedBaseIE(InfoExtractor):
 
         video_url = self._extract_video_url(webpage, video_id, url)
 
 
         video_url = self._extract_video_url(webpage, video_id, url)
 

-        title = base64.b64decode(self._html_search_meta(
-            'full:title', webpage, 'title').encode('utf-8')).decode('utf-8')
-        filesize = int_or_none(self._html_search_meta(
-            'full:size', webpage, 'file size', fatal=False))
+        title = self._extract_title(webpage)
+        filesize = int_or_none(self._extract_filesize(webpage))

 
         return {
             'id': video_id,
 
         return {
             'id': video_id,


@@ -35,6 +41,14 @@ class SharedBaseIE(InfoExtractor):
             'title': title,
         }
 
             'title': title,
         }
 

+    def _extract_title(self, webpage):
+        return compat_b64decode(self._html_search_meta(
+            'full:title', webpage, 'title')).decode('utf-8')
+
+    def _extract_filesize(self, webpage):
+        return self._html_search_meta(
+            'full:size', webpage, 'file size', fatal=False)
+

 
 class SharedIE(SharedBaseIE):
     IE_DESC = 'shared.sx'
 
 class SharedIE(SharedBaseIE):
     IE_DESC = 'shared.sx'


@@ -82,15 +96,43 @@ class VivoIE(SharedBaseIE):
             'id': 'd7ddda0e78',
             'ext': 'mp4',
             'title': 'Chicken',
             'id': 'd7ddda0e78',
             'ext': 'mp4',
             'title': 'Chicken',

-            'filesize': 528031,
+            'filesize': 515659,

         },
     }
 
         },
     }
 

-    def _extract_video_url(self, webpage, video_id, *args):
-        return self._parse_json(
+    def _extract_title(self, webpage):
+        title = self._html_search_regex(
+            r'data-name\s*=\s*(["\'])(?P<title>(?:(?!\1).)+)\1', webpage,
+            'title', default=None, group='title')
+        if title:
+            ext = determine_ext(title)
+            if ext.lower() in KNOWN_EXTENSIONS:
+                title = title.rpartition('.' + ext)[0]
+            return title
+        return self._og_search_title(webpage)
+
+    def _extract_filesize(self, webpage):
+        return parse_filesize(self._search_regex(
+            r'data-type=["\']video["\'][^>]*>Watch.*?<strong>\s*\((.+?)\)',
+            webpage, 'filesize', fatal=False))
+
+    def _extract_video_url(self, webpage, video_id, url):
+        def decode_url_old(encoded_url):
+            return compat_b64decode(encoded_url).decode('utf-8')
+
+        stream_url = self._search_regex(
+            r'data-stream\s*=\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
+            '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
+
+        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=lambda x: base64.b64decode(
-                x.encode('ascii')).decode('utf-8'))[0]
+                r'(?s)InitializeStream\s*\(\s*({.+?})\s*\)\s*;', webpage,
+                'stream'),
+            video_id, transform_source=js_to_json)['source'])