debian/README.source: Change 'markup'.

[youtubedl] / youtube_dl / extractor / shared.py

diff --git a/youtube_dl/extractor/shared.py b/youtube_dl/extractor/shared.py
index 931a0f70e0e0924689dfb6b71c374e6fa2c14457..ff575f592003beb91314b0101fb97b3a27a67432 100644 (file)
--- a/youtube_dl/extractor/shared.py
+++ b/youtube_dl/extractor/shared.py
@@ -3,8 +3,11 @@ from __future__ import unicode_literals
 from .common import InfoExtractor
 from ..compat import compat_b64decode
 from ..utils import (
+    determine_ext,
     ExtractorError,
     int_or_none,
+    KNOWN_EXTENSIONS,
+    parse_filesize,
     url_or_none,
     urlencode_postdata,
 )
@@ -22,10 +25,8 @@ class SharedBaseIE(InfoExtractor):
 
         video_url = self._extract_video_url(webpage, video_id, url)
 
-        title = compat_b64decode(self._html_search_meta(
-            'full:title', webpage, 'title')).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,
@@ -35,6 +36,14 @@ class SharedBaseIE(InfoExtractor):
             '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'
@@ -82,11 +91,27 @@ class VivoIE(SharedBaseIE):
             'id': 'd7ddda0e78',
             'ext': 'mp4',
             'title': 'Chicken',
-            'filesize': 528031,
+            'filesize': 515659,
         },
     }
 
-    def _extract_video_url(self, webpage, video_id, *args):
+    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(encoded_url):
             return compat_b64decode(encoded_url).decode('utf-8')