Add comment about weird .video extension for format 38

[youtube-dl] / youtube-dl

diff --git a/youtube-dl b/youtube-dl
index ba8a25a06148a234e76c663770a60d8ba8bbccde..481f9f400422149f552a490847558fb8cf43e960 100755 (executable)
--- a/youtube-dl
+++ b/youtube-dl
@@ -189,6 +189,7 @@ class FileDownloader(object):
        forcetitle:     Force printing title.
        simulate:       Do not download the video files.
        format:         Video format code.
+       format_limit:   Highest quality format to try.
        outtmpl:        Template for output names.
        ignoreerrors:   Do not stop on download errors.
        ratelimit:      Download speed limit, in bytes/sec.
@@ -679,7 +680,7 @@ class InfoExtractor(object):
 class YoutubeIE(InfoExtractor):
        """Information extractor for youtube.com."""
 
-       _VALID_URL = r'^((?:http://)?(?:\w+\.)?youtube\.com/(?:(?:v/)|(?:(?:watch(?:\.php)?)?[\?#](?:.+&)?v=)))?([0-9A-Za-z_-]+)(?(1).+)?$'
+       _VALID_URL = r'^((?:http://)?(?:youtu\.be/|(?:\w+\.)?youtube\.com/(?:(?:v/)|(?:(?:watch(?:_popup)?(?:\.php)?)?[\?#](?:.+&)?v=))))?([0-9A-Za-z_-]+)(?(1).+)?$'
        _LANG_URL = r'http://uk.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
        _LOGIN_URL = 'http://www.youtube.com/signup?next=/&gl=US&hl=en'
        _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
@@ -692,7 +693,7 @@ class YoutubeIE(InfoExtractor):
                '18': 'mp4',
                '22': 'mp4',
                '37': 'mp4',
-               '38': 'video',
+               '38': 'video', # You actually don't know if this will be MOV, AVI or whatever
                '43': 'webm',
                '45': 'webm',
        }
@@ -819,6 +820,13 @@ class YoutubeIE(InfoExtractor):
                        params = self._downloader.params
                        format_param = params.get('format', None)
                        if format_param == '0':
+                               format_limit = params.get('format_limit', None)
+                               if format_limit is not None:
+                                       try:
+                                               # Start at a different format if the user has limited the maximum quality
+                                               quality_index = self._available_formats.index(format_limit)
+                                       except ValueError:
+                                               pass
                                format_param = self._available_formats[quality_index]
                                best_quality = True
                        elif format_param == '-1':
@@ -839,7 +847,7 @@ class YoutubeIE(InfoExtractor):
                                return
 
                        # Attempt to extract SWF player URL
-                       mobj = re.search(r'swfConfig.*"(http://.*?watch-.*?\.swf)"', video_webpage)
+                       mobj = re.search(r'swfConfig.*"(http://.*?watch.*?-.*?\.swf)"', video_webpage)
                        if mobj is not None:
                                player_url = mobj.group(1)
                        else:
@@ -2111,6 +2119,8 @@ if __name__ == '__main__':
                                action='store_const', dest='format', help='alias for -f 22', const='22')
                video_format.add_option('--all-formats',
                                action='store_const', dest='format', help='download all available video formats', const='-1')
+               video_format.add_option('--max-quality',
+                               action='store', dest='format_limit', metavar='FORMAT', help='highest quality format limit for -b')
                parser.add_option_group(video_format)
 
                verbosity = optparse.OptionGroup(parser, 'Verbosity / Simulation Options')
@@ -2210,6 +2220,7 @@ if __name__ == '__main__':
                        'forcedescription': opts.getdescription,
                        'simulate': (opts.simulate or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription),
                        'format': opts.format,
+                       'format_limit': opts.format_limit,
                        'outtmpl': ((opts.outtmpl is not None and opts.outtmpl.decode(preferredencoding()))
                                or (opts.format == '-1' and opts.usetitle and u'%(stitle)s-%(id)s-%(format)s.%(ext)s')
                                or (opts.format == '-1' and opts.useliteral and u'%(title)s-%(id)s-%(format)s.%(ext)s')