YoutubeIE: use a negative index when accessing the last element of the format list
[youtube-dl] / youtube_dl / extractor / youtube.py
index 8470c40e25e90c893d7d9aa0609b2dd89ebf6bb2..6782bbff6addf710ab17bec5baf73a2e11c50201 100644 (file)
@@ -34,7 +34,7 @@ class YoutubeIE(InfoExtractor):
                          (?:                                                  # the various things that can precede the ID:
                              (?:(?:v|embed|e)/)                               # v/ or embed/ or e/
                              |(?:                                             # or the v= param in all its forms
-                                 (?:watch(?:_popup)?(?:\.php)?)?              # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
+                                 (?:watch|movie(?:_popup)?(?:\.php)?)?              # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
                                  (?:\?|\#!?)                                  # the params delimiter ? or # or #!
                                  (?:.*?&)?                                    # any other preceding param (like /?s=tuff&v=xxxx)
                                  v=
@@ -81,6 +81,44 @@ class YoutubeIE(InfoExtractor):
         '46': '1080x1920',
     }
     IE_NAME = u'youtube'
+    _TESTS = [
+        {
+            u"url":  u"http://www.youtube.com/watch?v=BaW_jenozKc",
+            u"file":  u"BaW_jenozKc.mp4",
+            u"info_dict": {
+                u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
+                u"uploader": u"Philipp Hagemeister",
+                u"uploader_id": u"phihag",
+                u"upload_date": u"20121002",
+                u"description": u"test chars:  \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
+            }
+        },
+        {
+            u"url":  u"http://www.youtube.com/watch?v=1ltcDfZMA3U",
+            u"file":  u"1ltcDfZMA3U.flv",
+            u"note": u"Test VEVO video (#897)",
+            u"info_dict": {
+                u"upload_date": u"20070518",
+                u"title": u"Maps - It Will Find You",
+                u"description": u"Music video by Maps performing It Will Find You.",
+                u"uploader": u"MuteUSA",
+                u"uploader_id": u"MuteUSA"
+            }
+        },
+        {
+            u"url":  u"http://www.youtube.com/watch?v=UxxajLWwzqY",
+            u"file":  u"UxxajLWwzqY.mp4",
+            u"note": u"Test generic use_cipher_signature video (#897)",
+            u"info_dict": {
+                u"upload_date": u"20120506",
+                u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
+                u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c",
+                u"uploader": u"IconaPop",
+                u"uploader_id": u"IconaPop"
+            }
+        }
+    ]
+
 
     @classmethod
     def suitable(cls, url):
@@ -130,16 +168,14 @@ class YoutubeIE(InfoExtractor):
         self.to_screen(u'RTMP download detected')
 
     def _decrypt_signature(self, s):
-        """Decrypt the key the two subkeys must have a length of 43"""
-        if self._downloader.params.get('verbose'):
-            self.to_screen('encrypted signature length %d' % (len(s)))
+        """Turn the encrypted s field into a working signature"""
 
         if len(s) == 88:
             return s[48] + s[81:67:-1] + s[82] + s[66:62:-1] + s[85] + s[61:48:-1] + s[67] + s[47:12:-1] + s[3] + s[11:3:-1] + s[2] + s[12]
         elif len(s) == 87:
             return s[62] + s[82:62:-1] + s[83] + s[61:52:-1] + s[0] + s[51:2:-1]
         elif len(s) == 86:
-            return s[2:17] + s[0] + s[18:41] + s[79] + s[42:79] + s[82] + s[80:82] + s[41]
+            return s[2:63] + s[82] + s[64:82] + s[63]
         elif len(s) == 85:
             return s[76] + s[82:76:-1] + s[83] + s[75:60:-1] + s[0] + s[59:50:-1] + s[1] + s[49:2:-1]
         elif len(s) == 84:
@@ -150,7 +186,7 @@ class YoutubeIE(InfoExtractor):
             return s[36] + s[79:67:-1] + s[81] + s[66:40:-1] + s[33] + s[39:36:-1] + s[40] + s[35] + s[0] + s[67] + s[32:0:-1] + s[34]
 
         else:
-            raise ExtractorError(u'Unable to decrypt signature, subkeys length %d not supported; retrying might work' % (len(s)))
+            raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
 
     def _get_available_subtitles(self, video_id):
         self.report_video_subtitles_download(video_id)
@@ -366,6 +402,9 @@ class YoutubeIE(InfoExtractor):
         return video_id
 
     def _real_extract(self, url):
+        if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
+            self._downloader.report_warning(u'Did you forget to quote the URL? Remember that & is a meta-character in most shells, so you want to put the URL in quotes, like  youtube-dl \'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\' (or simply  youtube-dl BaW_jenozKc  ).')
+
         # Extract original video URL from URL with redirection, like age verification, using next_url parameter
         mobj = re.search(self._NEXT_URL_RE, url)
         if mobj:
@@ -520,6 +559,12 @@ class YoutubeIE(InfoExtractor):
                     if 'sig' in url_data:
                         url += '&signature=' + url_data['sig'][0]
                     elif 's' in url_data:
+                        if self._downloader.params.get('verbose'):
+                            s = url_data['s'][0]
+                            player = self._search_regex(r'html5player-(.+?)\.js', video_webpage,
+                                'html5 player', fatal=False)
+                            self.to_screen('encrypted signature length %d (%d.%d), itag %s, html5 player %s' %
+                                (len(s), len(s.split('.')[0]), len(s.split('.')[1]), url_data['itag'][0], player))
                         signature = self._decrypt_signature(url_data['s'][0])
                         url += '&signature=' + signature
                     if 'ratebypass' not in url:
@@ -541,7 +586,7 @@ class YoutubeIE(InfoExtractor):
             if req_format is None or req_format == 'best':
                 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
             elif req_format == 'worst':
-                video_url_list = [(existing_formats[len(existing_formats)-1], url_map[existing_formats[len(existing_formats)-1]])] # worst quality
+                video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
             elif req_format in ('-1', 'all'):
                 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
             else: