Merge pull request #344 from aviperes/bugfix_facebook_unavilable_video

Bugfix facebook unavilable video

test/ci/test_facebook.py

@@ -1,5 +1,6 @@
 import unittest
 import youtube_dl
+from youtube_dl.utils import DownloadError
 
 
 class facebookMetaData(unittest.TestCase):
@@ -25,8 +26,15 @@ class facebookMetaData(unittest.TestCase):
         info = ydl.extract_info(url, download=False)
         self.assertGreater(info.get('comment_count'), 0)
 
+    def test_meta_data(self):
+        params = {}
+        url = "https://www.facebook.com/parapsychological.centr/videos/177407933624543/"
+        ydl = youtube_dl.YoutubeDL(params)
+        info = ydl.extract_info(url, download=False)
+        self.assertGreater(info.get('comment_count'), 0)
+
     def test_metadata_fetch_with_log_in(self):
-        url = "https://www.facebook.com/SerieA/videos/282581803097269"
+        url = "https://www.facebook.com/oristandup/videos/675360549895283"
         params = {}
         with open("cookie_file") as file:
             proxy = "ec2-3-221-82-67.compute-1.amazonaws.com:3128"
@@ -34,6 +42,8 @@ class facebookMetaData(unittest.TestCase):
             params['proxy'] = proxy
             ydl = youtube_dl.YoutubeDL(params)
             info = ydl.extract_info(url, download=False)
+            print (info.get('title'))
+            print (info.get('timestamp'))
             self.assertTrue(info.get('timestamp'))
             self.assertTrue(info.get('view_count'))
             self.assertTrue(info.get('comment_count'))
@@ -41,6 +51,19 @@ class facebookMetaData(unittest.TestCase):
             self.assertTrue(info.get('uploader_id'))
             self.assertTrue(info.get('thumbnail'))
 
+    def test_unavailable_video(self):
+        url = "https://www.facebook.com/101457238278830/videos/287839102599521/"
+        params = {}
+        with open("cookie_file") as file:
+            try:
+                proxy = "ec2-3-221-82-67.compute-1.amazonaws.com:3128"
+                params['cookiefile'] = file.name
+                params['proxy'] = proxy
+                ydl = youtube_dl.YoutubeDL(params)
+                info = ydl.extract_info(url, download=False)
+            except DownloadError:
+                self.assertRaises(DownloadError)
+
 
 if __name__ == '__main__':
     unittest.main()

youtube_dl/extractor/facebook.py

@@ -374,13 +374,7 @@ class FacebookIE(InfoExtractor):
         if not video_data:
             if not fatal_if_no_video:
                 return webpage, False
-            m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
+            self.validate_webpage(webpage)
-            if m_msg is not None:
-                raise ExtractorError(
-                    'The video is not available, Facebook said: "%s"' % m_msg.group(1),
-                    expected=True)
-            elif '>You must log in to continue' in webpage:
-                self.raise_login_required()
 
         if not video_data:
             info_dict = self.get_from_new_ui(webpage, tahoe_data, video_id)
@@ -806,13 +800,29 @@ class FacebookIE(InfoExtractor):
 
     def _resolve_thumbnail(self, webpage, tahoe_data):
         thumbnail = self._html_search_meta(['og:image', 'twitter:image'], webpage)
+
         if not thumbnail:
-            thumbnail = self._search_regex(r'"thumbSrc":"(.+?)"', tahoe_data.secondary, 'thumbnail', fatal=False)
+            thumbnail = self._search_regex(r'"thumbnailUrl":"(.+?)"', webpage, 'thumbnail', fatal=False)
             thumbnail = str(thumbnail).replace('\\', "")
         return thumbnail
 
     def _valid_video_title(self, video_title):
-        return video_title and not u'Log In or Sign Up to View' in video_title
+        if video_title:
+            video_title = video_title.lower()
+        return video_title and not u'log in or sign up to view' in video_title
+
+    def validate_webpage(self, webpage):
+        m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
+        if m_msg is not None:
+            raise ExtractorError(
+                'The video is not available, Facebook said: "%s"' % m_msg.group(1),
+                expected=True)
+        if 'Your Request Couldn\'t be Processed' in webpage:
+            raise ExtractorError(
+                'The video is not available, Facebook said: this content is not available',
+                expected=True)
+        elif '>You must log in to continue' in webpage:
+            self.raise_login_required()
 
 
 class FacebookTahoeData: