From 418c5cc3fc3ea99b791ad1774a6b03504eab7086 Mon Sep 17 00:00:00 2001
From: Yen Chi Hsuan <yan12125@gmail.com>
Date: Wed, 8 Apr 2015 17:26:51 +0800
Subject: [udn] Add new extractor

---
 youtube_dl/extractor/udn.py | 66 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 66 insertions(+)
 create mode 100644 youtube_dl/extractor/udn.py

(limited to 'youtube_dl/extractor/udn.py')

diff --git a/youtube_dl/extractor/udn.py b/youtube_dl/extractor/udn.py
new file mode 100644
index 000000000..2b9a733e8
--- /dev/null
+++ b/youtube_dl/extractor/udn.py
@@ -0,0 +1,66 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+from .common import InfoExtractor
+from ..utils import (
+    url_infer_protocol,
+    js_to_json
+)
+
+
+class UDNEmbedIE(InfoExtractor):
+    _VALID_URL = r'(?:https?:)?//video\.udn\.com/embed/news/(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://video.udn.com/embed/news/300040',
+        'md5': 'de06b4c90b042c128395a88f0384817e',
+        'info_dict': {
+            'id': '300040',
+            'ext': 'mp4',
+            'title': '生物老師男變女 全校挺"做自己"',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }, {
+        'url': '//video.udn.com/embed/news/300040',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        page = self._download_webpage(url, video_id)
+
+        options = json.loads(js_to_json(self._html_search_regex(
+            r'var options\s*=\s*([^;]+);', page, 'video urls dictionary')))
+
+        video_urls = options['video']
+
+        if video_urls.get('youtube'):
+            return self.url_result(video_urls.get('youtube'), 'Youtube')
+
+        try:
+            del video_urls['youtube']
+        except KeyError:
+            pass
+
+        formats = [{
+            'url': self._download_webpage(
+                url_infer_protocol(url, api_url), video_id,
+                'retrieve url for %s video' % video_type),
+            'format_id': video_type,
+            'preference': 0 if video_type == 'mp4' else -1,
+        } for video_type, api_url in video_urls.items()]
+
+        self._sort_formats(formats)
+
+        thumbnail = None
+
+        if options.get('gallery') and len(options['gallery']):
+            thumbnail = options['gallery'][0].get('original')
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': options['title'],
+            'thumbnail': thumbnail
+        }
-- 
cgit v1.2.3