youtube-dl

commit 45aab4d30b7c3fc03c9be9680550cba88bd85b5c
parent ed2bfe93aaa11f49f7b2b92b581abb6aa385dfbf
Author: Yen Chi Hsuan <yan12125@gmail.com>
Date:   Sat,  3 Sep 2016 18:37:36 +0800

[youjizz] Fix extraction. The site has moved to HTML5

Closes #10437

Diffstat:
M
ChangeLog
 | 
1
+
M
youtube_dl/extractor/youjizz.py
 | 
43
+++++++++----------------------------------

2 files changed, 10 insertions(+), 34 deletions(-)
diff --git a/ChangeLog b/ChangeLog
@@ -1,6 +1,7 @@
 version <unreleased>
 
 Extractors
+* [youjizz] Fix extraction (#10437)
 + [foxnews] Add support for FoxNews Insider (#10445)
 + [fc2] Recognize Flash player URLs (#10512)
 
diff --git a/youtube_dl/extractor/youjizz.py b/youtube_dl/extractor/youjizz.py
@@ -1,21 +1,16 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-)
 
 
 class YouJizzIE(InfoExtractor):
     _VALID_URL = r'https?://(?:\w+\.)?youjizz\.com/videos/(?:[^/#?]+)?-(?P<id>[0-9]+)\.html(?:$|[?#])'
     _TESTS = [{
         'url': 'http://www.youjizz.com/videos/zeichentrick-1-2189178.html',
-        'md5': '07e15fa469ba384c7693fd246905547c',
+        'md5': '78fc1901148284c69af12640e01c6310',
         'info_dict': {
             'id': '2189178',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Zeichentrick 1',
             'age_limit': 18,
         }
@@ -27,38 +22,18 @@ class YouJizzIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
+        # YouJizz's HTML5 player has invalid HTML
+        webpage = webpage.replace('"controls', '" controls')
         age_limit = self._rta_search(webpage)
         video_title = self._html_search_regex(
             r'<title>\s*(.*)\s*</title>', webpage, 'title')
 
-        embed_page_url = self._search_regex(
-            r'(https?://www.youjizz.com/videos/embed/[0-9]+)',
-            webpage, 'embed page')
-        webpage = self._download_webpage(
-            embed_page_url, video_id, note='downloading embed page')
-
-        # Get the video URL
-        m_playlist = re.search(r'so.addVariable\("playlist", ?"(?P<playlist>.+?)"\);', webpage)
-        if m_playlist is not None:
-            playlist_url = m_playlist.group('playlist')
-            playlist_page = self._download_webpage(playlist_url, video_id,
-                                                   'Downloading playlist page')
-            m_levels = list(re.finditer(r'<level bitrate="(\d+?)" file="(.*?)"', playlist_page))
-            if len(m_levels) == 0:
-                raise ExtractorError('Unable to extract video url')
-            videos = [(int(m.group(1)), m.group(2)) for m in m_levels]
-            (_, video_url) = sorted(videos)[0]
-            video_url = video_url.replace('%252F', '%2F')
-        else:
-            video_url = self._search_regex(r'so.addVariable\("file",encodeURIComponent\("(?P<source>[^"]+)"\)\);',
-                                           webpage, 'video URL')
+        info_dict = self._parse_html5_media_entries(url, webpage, video_id)[0]
 
-        return {
+        info_dict.update({
             'id': video_id,
-            'url': video_url,
             'title': video_title,
-            'ext': 'flv',
-            'format': 'flv',
-            'player_url': embed_page_url,
             'age_limit': age_limit,
-        }
+        })
+
+        return info_dict