[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / thisav.py

diff --git a/youtube_dl/extractor/thisav.py b/youtube_dl/extractor/thisav.py
index 027a8e90712a7b8f84f8bd5096ea41a125ef9384..dc3dd03c823d7ec7878943b5f99bc14d266a6257 100644 (file)
--- a/youtube_dl/extractor/thisav.py
+++ b/youtube_dl/extractor/thisav.py
@@ -3,12 +3,14 @@ from __future__ import unicode_literals
 
 import re
 
-from .jwplatform import JWPlatformBaseIE
+from .common import InfoExtractor
+from ..utils import remove_end
 
 
-class ThisAVIE(JWPlatformBaseIE):
+class ThisAVIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?thisav\.com/video/(?P<id>[0-9]+)/.*'
     _TESTS = [{
+        # jwplayer
         'url': 'http://www.thisav.com/video/47734/%98%26sup1%3B%83%9E%83%82---just-fit.html',
         'md5': '0480f1ef3932d901f0e0e719f188f19b',
         'info_dict': {
@@ -19,6 +21,7 @@ class ThisAVIE(JWPlatformBaseIE):
             'uploader_id': 'dj7970'
         }
     }, {
+        # html5 media
         'url': 'http://www.thisav.com/video/242352/nerdy-18yo-big-ass-tattoos-and-glasses.html',
         'md5': 'ba90c076bd0f80203679e5b60bf523ee',
         'info_dict': {
@@ -35,7 +38,9 @@ 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:
@@ -45,13 +50,17 @@ class ThisAVIE(JWPlatformBaseIE):
                 }],
             }
         else:
-            info_dict = self._extract_jwplayer_data(
-                webpage, video_id, require_title=False)
+            entries = self._parse_html5_media_entries(url, webpage, video_id)
+            if entries:
+                info_dict = entries[0]
+            else:
+                info_dict = self._extract_jwplayer_data(
+                    webpage, video_id, require_title=False)
         uploader = self._html_search_regex(
-            r': <a href="http://www.thisav.com/user/[0-9]+/(?:[^"]+)">([^<]+)</a>',
+            r': <a href="http://www\.thisav\.com/user/[0-9]+/(?:[^"]+)">([^<]+)</a>',
             webpage, 'uploader name', fatal=False)
         uploader_id = self._html_search_regex(
-            r': <a href="http://www.thisav.com/user/[0-9]+/([^"]+)">(?:[^<]+)</a>',
+            r': <a href="http://www\.thisav\.com/user/[0-9]+/([^"]+)">(?:[^<]+)</a>',
             webpage, 'uploader id', fatal=False)
 
         info_dict.update({