]> git.bitcoin.ninja Git - youtube-dl/commitdiff
[facebook] Add support for plugin video embeds and multiple embeds (closes #13493)
authorSergey M․ <dstftw@gmail.com>
Tue, 27 Jun 2017 15:25:34 +0000 (22:25 +0700)
committerSergey M․ <dstftw@gmail.com>
Tue, 27 Jun 2017 15:38:54 +0000 (22:38 +0700)
youtube_dl/extractor/buzzfeed.py
youtube_dl/extractor/facebook.py
youtube_dl/extractor/generic.py

index 75fa92d7cfc0204f4539e10c762585b2537abbb6..ec411091efe7dc15b28c7f4a3939bff89395fa59 100644 (file)
@@ -84,9 +84,10 @@ class BuzzFeedIE(InfoExtractor):
                 continue
             entries.append(self.url_result(video['url']))
 
-        facebook_url = FacebookIE._extract_url(webpage)
-        if facebook_url:
-            entries.append(self.url_result(facebook_url))
+        facebook_urls = FacebookIE._extract_urls(webpage)
+        entries.extend([
+            self.url_result(facebook_url)
+            for facebook_url in facebook_urls])
 
         return {
             '_type': 'playlist',
index b69c1ede0046d73e31df2098f78cf6dc20c254d4..4b3f6cc86b57f283f08faedfa481bfbc9719a879 100644 (file)
@@ -203,19 +203,19 @@ class FacebookIE(InfoExtractor):
     }]
 
     @staticmethod
-    def _extract_url(webpage):
-        mobj = re.search(
-            r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
-        if mobj is not None:
-            return mobj.group('url')
-
+    def _extract_urls(webpage):
+        urls = []
+        for mobj in re.finditer(
+                r'<iframe[^>]+?src=(["\'])(?P<url>https?://www\.facebook\.com/(?:video/embed|plugins/video\.php).+?)\1',
+                webpage):
+            urls.append(mobj.group('url'))
         # Facebook API embed
         # see https://developers.facebook.com/docs/plugins/embedded-video-player
-        mobj = re.search(r'''(?x)<div[^>]+
+        for mobj in re.finditer(r'''(?x)<div[^>]+
                 class=(?P<q1>[\'"])[^\'"]*\bfb-(?:video|post)\b[^\'"]*(?P=q1)[^>]+
-                data-href=(?P<q2>[\'"])(?P<url>(?:https?:)?//(?:www\.)?facebook.com/.+?)(?P=q2)''', webpage)
-        if mobj is not None:
-            return mobj.group('url')
+                data-href=(?P<q2>[\'"])(?P<url>(?:https?:)?//(?:www\.)?facebook.com/.+?)(?P=q2)''', webpage):
+            urls.append(mobj.group('url'))
+        return urls
 
     def _login(self):
         (useremail, password) = self._get_login_info()
index 8ef1a29801972995150600a48ed09364c9a77c3e..760a7f9c24d0c7de8f1c7e6e1a895effec67bb17 100644 (file)
@@ -2222,9 +2222,9 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'))
 
         # Look for embedded Facebook player
-        facebook_url = FacebookIE._extract_url(webpage)
-        if facebook_url is not None:
-            return self.url_result(facebook_url, 'Facebook')
+        facebook_urls = FacebookIE._extract_urls(webpage)
+        if facebook_urls:
+            return self.playlist_from_matches(facebook_urls, video_id, video_title)
 
         # Look for embedded VK player
         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)