[udn] Add new extractor

test/test_utils.py

@@ -53,6 +53,7 @@ from youtube_dl.utils import (
     uppercase_escape,
     url_basename,
     urlencode_postdata,
+    url_infer_protocol,
     version_tuple,
     xpath_with_ns,
     xpath_text,
@@ -296,6 +297,10 @@ class TestUtil(unittest.TestCase):
             url_basename('http://media.w3.org/2010/05/sintel/trailer.mp4'),
             'trailer.mp4')
 
+    def test_url_infer_protocol(self):
+        self.assertEqual(url_infer_protocol('http://foo.com/', '//bar.com/'), 'http://bar.com/')
+        self.assertEqual(url_infer_protocol('http://foo.com/', 'https://bar.com/'), 'https://bar.com/')
+
     def test_parse_duration(self):
         self.assertEqual(parse_duration(None), None)
         self.assertEqual(parse_duration(False), None)

youtube_dl/extractor/__init__.py

@@ -557,6 +557,7 @@ from .udemy import (
     UdemyIE,
     UdemyCourseIE
 )
+from .udn import UDNEmbedIE
 from .ultimedia import UltimediaIE
 from .unistra import UnistraIE
 from .urort import UrortIE

youtube_dl/extractor/generic.py

@@ -26,6 +26,7 @@ from ..utils import (
     unsmuggle_url,
     UnsupportedError,
     url_basename,
+    url_infer_protocol,
     xpath_text,
 )
 from .brightcove import BrightcoveIE
@@ -34,6 +35,7 @@ from .ooyala import OoyalaIE
 from .rutv import RUTVIE
 from .smotri import SmotriIE
 from .condenast import CondeNastIE
+from .udn import UDNEmbedIE
 
 
 class GenericIE(InfoExtractor):
@@ -650,6 +652,17 @@ class GenericIE(InfoExtractor):
                 'title': "PFT Live: New leader in the 'new-look' defense",
                 'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
             },
+        },
+        # UDN embed
+        {
+            'url': 'http://www.udn.com/news/story/7314/822787',
+            '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$',
+            }
         }
     ]
 
@@ -1268,6 +1281,13 @@ class GenericIE(InfoExtractor):
         if nbc_sports_url:
             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
 
+        # Look for UDN embeds
+        mobj = re.search(
+            r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage)
+        if mobj is not None:
+            return self.url_result(
+                url_infer_protocol(url, mobj.group('url')), 'UDNEmbed')
+
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
                 return True

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
+        }

youtube_dl/utils.py

@@ -1711,6 +1711,17 @@ def determine_protocol(info_dict):
     return compat_urllib_parse_urlparse(url).scheme
 
 
+def url_infer_protocol(ref_url, target_url):
+    """ Infer protocol for protocol independent target urls """
+    parsed_target_url = list(compat_urllib_parse_urlparse(target_url))
+    if parsed_target_url[0]:
+        return target_url
+
+    parsed_target_url[0] = compat_urllib_parse_urlparse(ref_url).scheme
+
+    return compat_urlparse.urlunparse(parsed_target_url)
+
+
 def render_table(header_row, data):
     """ Render a list of rows, each as a list of values """
     table = [header_row] + data