[thisav] Improve title extraction (closes #10682)

I didn't add a test case as the one in #10682 looks like a copyrighted
product.

diff --git a/ChangeLog b/ChangeLog
index a71fadfa75eeb3a77ac9b060963dd7667ad59cf7..18f9fa86117b587161f93d87f7d50c922f9d73fe 100644 (file)
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,9 @@
+version <unreleased>
+
+Extractors
+* [thisav] Improve title extraction (#10682)
+
+
 version 2016.09.18
 
 Core

diff --git a/youtube_dl/extractor/thisav.py b/youtube_dl/extractor/thisav.py
index 027a8e90712a7b8f84f8bd5096ea41a125ef9384..4473a3c773c3d9c4c26361e907769e8bb1ac9fad 100644 (file)
--- a/youtube_dl/extractor/thisav.py
+++ b/youtube_dl/extractor/thisav.py
@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .jwplatform import JWPlatformBaseIE
+from ..utils import remove_end
 
 
 class ThisAVIE(JWPlatformBaseIE):
@@ -35,7 +36,9 @@ class ThisAVIE(JWPlatformBaseIE):
 
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
-        title = self._html_search_regex(r'<h1>([^<]*)</h1>', webpage, 'title')
+        title = remove_end(self._html_search_regex(
+            r'<title>([^<]+)</title>', webpage, 'title'),
+            ' - 視頻 - ThisAV.com-世界第一中文成人娛樂網站')
         video_url = self._html_search_regex(
             r"addVariable\('file','([^']+)'\);", webpage, 'video url', default=None)
         if video_url: