Merge pull request #8092 from bpfoley/twitter-thumbnail

[youtube-dl] / youtube_dl / extractor / revision3.py

diff --git a/youtube_dl/extractor/revision3.py b/youtube_dl/extractor/revision3.py
index fb2196ad678807ea3e1ed25d002d8adac5b661b1..99979ebe1a9fe82099076b46b576ef38a58bca8c 100644 (file)
--- a/youtube_dl/extractor/revision3.py
+++ b/youtube_dl/extractor/revision3.py
@@ -19,7 +19,7 @@ class Revision3IE(InfoExtractor):
         'url': 'http://www.revision3.com/technobuffalo/5-google-predictions-for-2016',
         'md5': 'd94a72d85d0a829766de4deb8daaf7df',
         'info_dict': {
-            'id': '73034',
+            'id': '71089',
             'display_id': 'technobuffalo/5-google-predictions-for-2016',
             'ext': 'webm',
             'title': '5 Google Predictions for 2016',
@@ -31,6 +31,7 @@ class Revision3IE(InfoExtractor):
             'uploader_id': 'technobuffalo',
         }
     }, {
+        # Show
         'url': 'http://testtube.com/brainstuff',
         'info_dict': {
             'id': '251',
@@ -41,7 +42,7 @@ class Revision3IE(InfoExtractor):
     }, {
         'url': 'https://testtube.com/dnews/5-weird-ways-plants-can-eat-animals?utm_source=FB&utm_medium=DNews&utm_campaign=DNewsSocial',
         'info_dict': {
-            'id': '60163',
+            'id': '58227',
             'display_id': 'dnews/5-weird-ways-plants-can-eat-animals',
             'duration': 275,
             'ext': 'webm',
@@ -55,7 +56,7 @@ class Revision3IE(InfoExtractor):
     }, {
         'url': 'http://testtube.com/tt-editors-picks/the-israel-palestine-conflict-explained-in-ten-min',
         'info_dict': {
-            'id': '73573',
+            'id': '71618',
             'ext': 'mp4',
             'display_id': 'tt-editors-picks/the-israel-palestine-conflict-explained-in-ten-min',
             'title': 'The Israel-Palestine Conflict Explained in Ten Minutes',
@@ -66,19 +67,29 @@ class Revision3IE(InfoExtractor):
             'upload_date': '20160120',
         },
         'add_ie': ['Youtube'],
+    }, {
+        # Tag
+        'url': 'http://testtube.com/tech-news',
+        'info_dict': {
+            'id': '21018',
+            'title': 'tech news',
+        },
+        'playlist_mincount': 9,
     }]
     _PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s'
     _API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62'
 
     def _real_extract(self, url):
         domain, display_id = re.match(self._VALID_URL, url).groups()
+        site = domain.split('.')[0]
         page_info = self._download_json(
             self._PAGE_DATA_TEMPLATE % (domain, display_id, domain), display_id)
 
         page_data = page_info['data']
         page_type = page_data['type']
-        if page_type == 'episode' or page_type == 'embed':
+        if page_type in ('episode', 'embed'):
             show_data = page_data['show']['data']
+            page_id = compat_str(page_data['id'])
             video_id = compat_str(page_data['video']['data']['id'])
 
             preference = qualities(['mini', 'small', 'medium', 'large'])
@@ -89,7 +100,7 @@ class Revision3IE(InfoExtractor):
             } for image_id, image_url in page_data.get('images', {}).items()]
 
             info = {
-                'id': video_id,
+                'id': page_id,
                 'display_id': display_id,
                 'title': unescapeHTML(page_data['name']),
                 'description': unescapeHTML(page_data.get('summary')),
@@ -98,6 +109,7 @@ class Revision3IE(InfoExtractor):
                 'uploader': show_data.get('name'),
                 'uploader_id': show_data.get('slug'),
                 'thumbnails': thumbnails,
+                'extractor_key': site,
             }
 
             if page_type == 'embed':
@@ -137,15 +149,20 @@ class Revision3IE(InfoExtractor):
             })
             return info
         else:
-            show_data = page_info['show']['data']
+            list_data = page_info[page_type]['data']
             episodes_data = page_info['episodes']['data']
             num_episodes = page_info['meta']['totalEpisodes']
             processed_episodes = 0
             entries = []
             page_num = 1
             while True:
-                entries.extend([self.url_result(
-                    'http://%s/%s/%s' % (domain, display_id, episode['slug'])) for episode in episodes_data])
+                entries.extend([{
+                    '_type': 'url',
+                    'url': 'http://%s%s' % (domain, episode['path']),
+                    'id': compat_str(episode['id']),
+                    'ie_key': 'Revision3',
+                    'extractor_key': site,
+                } for episode in episodes_data])
                 processed_episodes += len(episodes_data)
                 if processed_episodes == num_episodes:
                     break
@@ -155,5 +172,5 @@ class Revision3IE(InfoExtractor):
                     display_id)['episodes']['data']
 
             return self.playlist_result(
-                entries, compat_str(show_data['id']),
-                show_data.get('name'), show_data.get('summary'))
+                entries, compat_str(list_data['id']),
+                list_data.get('name'), list_data.get('summary'))