X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fdemocracynow.py;h=5c9c0ecdc0e20c1dbf9871edb65c43a24c714396;hb=a6211d237b4e7051ca018cc09440502561fedaa7;hp=824b8e2c5c6251960f26fdb049f9d033c9f00a45;hpb=dde9fe9788f23f168e0bddaf8ab0470f469165fa;p=youtube-dl

diff --git a/youtube_dl/extractor/democracynow.py b/youtube_dl/extractor/democracynow.py
index 824b8e2c5..5c9c0ecdc 100644
--- a/youtube_dl/extractor/democracynow.py
+++ b/youtube_dl/extractor/democracynow.py
@@ -13,15 +13,16 @@ from ..utils import (
 
 
 class DemocracynowIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?democracynow.org/(?P<id>[^\?]*)'
+    _VALID_URL = r'https?://(?:www\.)?democracynow\.org/(?P<id>[^\?]*)'
     IE_NAME = 'democracynow'
     _TESTS = [{
         'url': 'http://www.democracynow.org/shows/2015/7/3',
+        'md5': '3757c182d3d84da68f5c8f506c18c196',
         'info_dict': {
             'id': '2015-0703-001',
             'ext': 'mp4',
-            'title': 'July 03, 2015 - Democracy Now!',
-            'description': 'A daily independent global news hour with Amy Goodman & Juan González "What to the Slave is 4th of July?": James Earl Jones Reads Frederick Douglass\u2019 Historic Speech : "This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag : "We Shall Overcome": Remembering Folk Icon, Activist Pete Seeger in His Own Words & Songs',
+            'title': 'Daily Show for July 03, 2015',
+            'description': 'md5:80eb927244d6749900de6072c7cc2c86',
         },
     }, {
         'url': 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree',
@@ -31,21 +32,39 @@ class DemocracynowIE(InfoExtractor):
             'title': '"This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag',
             'description': 'md5:4d2bc4f0d29f5553c2210a4bc7761a21',
         },
+        'params': {
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
+
         webpage = self._download_webpage(url, display_id)
-        description = self._og_search_description(webpage)
 
-        js = self._parse_json(self._search_regex(
+        json_data = self._parse_json(self._search_regex(
             r'<script[^>]+type="text/json"[^>]*>\s*({[^>]+})', webpage, 'json'),
             display_id)
-        video_id = None
+
+        title = json_data['title']
         formats = []
 
-        default_lang = 'en'
+        video_id = None
+
+        for key in ('file', 'audio', 'video', 'high_res_video'):
+            media_url = json_data.get(key, '')
+            if not media_url:
+                continue
+            media_url = re.sub(r'\?.*', '', compat_urlparse.urljoin(url, media_url))
+            video_id = video_id or remove_start(os.path.splitext(url_basename(media_url))[0], 'dn')
+            formats.append({
+                'url': media_url,
+                'vcodec': 'none' if key == 'audio' else None,
+            })
+
+        self._sort_formats(formats)
 
+        default_lang = 'en'
         subtitles = {}
 
         def add_subtitle_item(lang, info_dict):
@@ -54,33 +73,24 @@ class DemocracynowIE(InfoExtractor):
             subtitles[lang].append(info_dict)
 
         # chapter_file are not subtitles
-        if 'caption_file' in js:
+        if 'caption_file' in json_data:
             add_subtitle_item(default_lang, {
-                'url': compat_urlparse.urljoin(url, js['caption_file']),
+                'url': compat_urlparse.urljoin(url, json_data['caption_file']),
             })
 
-        for subtitle_item in js.get('captions', []):
+        for subtitle_item in json_data.get('captions', []):
             lang = subtitle_item.get('language', '').lower() or default_lang
             add_subtitle_item(lang, {
                 'url': compat_urlparse.urljoin(url, subtitle_item['url']),
             })
 
-        for key in ('file', 'audio', 'video'):
-            media_url = js.get(key, '')
-            if not media_url:
-                continue
-            media_url = re.sub(r'\?.*', '', compat_urlparse.urljoin(url, media_url))
-            video_id = video_id or remove_start(os.path.splitext(url_basename(media_url))[0], 'dn')
-            formats.append({
-                'url': media_url,
-            })
-
-        self._sort_formats(formats)
+        description = self._og_search_description(webpage, default=None)
 
         return {
-            'id': video_id,
-            'title': js.get('title'),
+            'id': video_id or display_id,
+            'title': title,
             'description': description,
+            'thumbnail': json_data.get('image'),
             'subtitles': subtitles,
             'formats': formats,
         }