Merge pull request #8611 from remitamine/ffmpegfd

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

diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py
index b354c5521852c2f78a18f83a70b3b473839a84f2..f5bbd39d2d0e90996c118e3fae325034fc2bbb6d 100644 (file)
--- a/youtube_dl/extractor/facebook.py
+++ b/youtube_dl/extractor/facebook.py
@@ -38,7 +38,8 @@ class FacebookIE(InfoExtractor):
                                 story\.php
                             )\?(?:.*?)(?:v|video_id|story_fbid)=|
                             [^/]+/videos/(?:[^/]+/)?|
-                            [^/]+/posts/
+                            [^/]+/posts/|
+                            groups/[^/]+/permalink/
                         )|
                     facebook:
                 )
@@ -102,6 +103,15 @@ class FacebookIE(InfoExtractor):
             'title': 'What the Flock Is Going On In New Zealand  Credit: ViralHog',
             'uploader': 'S. Saint',
         },
+    }, {
+        'note': 'swf params escaped',
+        'url': 'https://www.facebook.com/barackobama/posts/10153664894881749',
+        'md5': '97ba073838964d12c70566e0085c2b91',
+        'info_dict': {
+            'id': '10153664894881749',
+            'ext': 'mp4',
+            'title': 'Facebook video #10153664894881749',
+        },
     }, {
         'url': 'https://www.facebook.com/video.php?v=10204634152394104',
         'only_matching': True,
@@ -114,6 +124,9 @@ class FacebookIE(InfoExtractor):
     }, {
         'url': 'facebook:544765982287235',
         'only_matching': True,
+    }, {
+        'url': 'https://www.facebook.com/groups/164828000315060/permalink/764967300301124/',
+        'only_matching': True,
     }]
 
     def _login(self):
@@ -189,11 +202,12 @@ class FacebookIE(InfoExtractor):
 
         video_data = None
 
-        BEFORE = '{swf.addParam(param[0], param[1]);});\n'
+        BEFORE = '{swf.addParam(param[0], param[1]);});'
         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
-        m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
+        m = re.search(re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER), webpage)
         if m:
-            data = dict(json.loads(m.group(1)))
+            swf_params = m.group(1).replace('\\\\', '\\').replace('\\"', '"')
+            data = dict(json.loads(swf_params))
             params_raw = compat_urllib_parse_unquote(data['params'])
             video_data = json.loads(params_raw)['video_data']
 
@@ -280,8 +294,8 @@ class FacebookIE(InfoExtractor):
 
         if '/posts/' in url:
             entries = [
-                self.url_result('facebook:%s' % video_id, FacebookIE.ie_key())
-                for video_id in self._parse_json(
+                self.url_result('facebook:%s' % vid, FacebookIE.ie_key())
+                for vid in self._parse_json(
                     self._search_regex(
                         r'(["\'])video_ids\1\s*:\s*(?P<ids>\[.+?\])',
                         webpage, 'video ids', group='ids'),