Merge branch 'vgtv' of https://github.com/mrkolby/youtube-dl into mrkolby-vgtv

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

diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py
index e2cf1ae5603c6ff32480eb0db07ece09b7ff6baf..3ab6017cdb51a3eaef6a3a1686719fba714780dd 100644 (file)
--- a/youtube_dl/extractor/yahoo.py
+++ b/youtube_dl/extractor/yahoo.py
@@ -14,14 +14,14 @@ from ..utils import (
 
 
 class YahooIE(InfoExtractor):
-    IE_DESC = 'Yahoo screen'
-    _VALID_URL = r'https?://screen\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html'
+    IE_DESC = 'Yahoo screen and movies'
+    _VALID_URL = r'(?P<url>https?://(?:screen|movies)\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html)'
     _TESTS = [
         {
             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
             'md5': '4962b075c08be8690a922ee026d05e69',
             'info_dict': {
-                'id': '214727115',
+                'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
                 'ext': 'mp4',
                 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
                 'description': 'Julian and Travis watch Julian Smith',
@@ -31,29 +31,60 @@ class YahooIE(InfoExtractor):
             'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
             'md5': 'd6e6fc6e1313c608f316ddad7b82b306',
             'info_dict': {
-                'id': '103000935',
+                'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
                 'ext': 'mp4',
                 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
                 'description': 'Agent Topple\'s mustache does its dirty work, and Nicole brokers a deal for peace. But why is the NSA collecting millions of Instagram brunch photos? And if your waffles have nothing to hide, what are they so worried about?',
             },
         },
+        {
+            'url': 'https://movies.yahoo.com/video/world-loves-spider-man-190819223.html',
+            'md5': '410b7104aa9893b765bc22787a22f3d9',
+            'info_dict': {
+                'id': '516ed8e2-2c4f-339f-a211-7a8b49d30845',
+                'ext': 'mp4',
+                'title': 'The World Loves Spider-Man',
+                'description': '''People all over the world are celebrating the release of \"The Amazing Spider-Man 2.\" We're taking a look at the enthusiastic response Spider-Man has received from viewers all over the world.''',
+            }
+        },
+        {
+            'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
+            'md5': '60e8ac193d8fb71997caa8fce54c6460',
+            'info_dict': {
+                'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
+                'ext': 'mp4',
+                'title': "Yahoo Saves 'Community'",
+                'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
+            }
+        },
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
+        url = mobj.group('url')
         webpage = self._download_webpage(url, video_id)
 
-        items_json = self._search_regex(r'mediaItems: ({.*?})$',
-            webpage, 'items', flags=re.MULTILINE)
-        items = json.loads(items_json)
-        info = items['mediaItems']['query']['results']['mediaObj'][0]
-        # The 'meta' field is not always in the video webpage, we request it
-        # from another page
-        long_id = info['id']
-        return self._get_info(long_id, video_id)
-
-    def _get_info(self, long_id, video_id):
+        items_json = self._search_regex(
+            r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
+            default=None)
+        if items_json is None:
+            CONTENT_ID_REGEXES = [
+                r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
+                r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"',
+                r'"first_videoid"\s*:\s*"([^"]+)"',
+            ]
+            long_id = self._search_regex(CONTENT_ID_REGEXES, webpage, 'content ID')
+            video_id = long_id
+        else:
+            items = json.loads(items_json)
+            info = items['mediaItems']['query']['results']['mediaObj'][0]
+            # The 'meta' field is not always in the video webpage, we request it
+            # from another page
+            long_id = info['id']
+        return self._get_info(long_id, video_id, webpage)
+
+    def _get_info(self, long_id, video_id, webpage):
         query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"'
                  ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2" AND region="US"'
                  ' AND protocol="http"' % long_id)
@@ -96,7 +127,7 @@ class YahooIE(InfoExtractor):
             'title': meta['title'],
             'formats': formats,
             'description': clean_html(meta['description']),
-            'thumbnail': meta['thumbnail'],
+            'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
         }
 
 
@@ -104,7 +135,7 @@ class YahooNewsIE(YahooIE):
     IE_NAME = 'yahoo:news'
     _VALID_URL = r'http://news\.yahoo\.com/video/.*?-(?P<id>\d*?)\.html'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
         'md5': '67010fdf3a08d290e060a4dd96baa07b',
         'info_dict': {
@@ -113,17 +144,14 @@ class YahooNewsIE(YahooIE):
             'title': 'China Moses Is Crazy About the Blues',
             'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
         },
-    }
-
-    # Overwrite YahooIE properties we don't want
-    _TESTS = []
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
         long_id = self._search_regex(r'contentId: \'(.+?)\',', webpage, 'long id')
-        return self._get_info(long_id, video_id)
+        return self._get_info(long_id, video_id, webpage)
 
 
 class YahooSearchIE(SearchInfoExtractor):