Take format 37 into account (fixes issue #65)
[youtube-dl] / youtube-dl
index becf1d25de31ccbdbaaef76b517bb23490a9c001..52bda1b2533e9b8b5c63274074f285a8cccced95 100755 (executable)
@@ -27,6 +27,22 @@ std_headers = {
 
 simple_title_chars = string.ascii_letters.decode('ascii') + string.digits.decode('ascii')
 
+def preferredencoding():
+       """Get preferred encoding.
+
+       Returns the best encoding scheme for the system, based on
+       locale.getpreferredencoding() and some further tweaks.
+       """
+       def yield_preferredencoding():
+               try:
+                       pref = locale.getpreferredencoding()
+                       u'TEST'.encode(pref)
+               except:
+                       pref = 'UTF-8'
+               while True:
+                       yield pref
+       return yield_preferredencoding().next()
+
 class DownloadError(Exception):
        """Download Error exception.
        
@@ -205,11 +221,13 @@ class FileDownloader(object):
 
        @staticmethod
        def verify_url(url):
-               """Verify a URL is valid and data could be downloaded."""
+               """Verify a URL is valid and data could be downloaded. Return real data URL."""
                request = urllib2.Request(url, None, std_headers)
                data = urllib2.urlopen(request)
                data.read(1)
+               url = data.geturl()
                data.close()
+               return url
 
        def add_info_extractor(self, ie):
                """Add an InfoExtractor object to the end of the list."""
@@ -224,12 +242,12 @@ class FileDownloader(object):
        def to_stdout(self, message, skip_eol=False):
                """Print message to stdout if not in quiet mode."""
                if not self.params.get('quiet', False):
-                       print (u'%s%s' % (message, [u'\n', u''][skip_eol])).encode(locale.getpreferredencoding()),
+                       print (u'%s%s' % (message, [u'\n', u''][skip_eol])).encode(preferredencoding()),
                        sys.stdout.flush()
        
        def to_stderr(self, message):
                """Print message to stderr."""
-               print >>sys.stderr, message.encode(locale.getpreferredencoding())
+               print >>sys.stderr, message.encode(preferredencoding())
        
        def fixed_template(self):
                """Checks if the output template is fixed."""
@@ -291,15 +309,15 @@ class FileDownloader(object):
                # Do nothing else if in simulate mode
                if self.params.get('simulate', False):
                        try:
-                               self.verify_url(info_dict['url'])
+                               info_dict['url'] = self.verify_url(info_dict['url'].encode('utf-8')).decode('utf-8')
                        except (OSError, IOError, urllib2.URLError, httplib.HTTPException, socket.error), err:
                                raise UnavailableFormatError
 
                        # Forced printings
                        if self.params.get('forcetitle', False):
-                               print info_dict['title'].encode(locale.getpreferredencoding())
+                               print info_dict['title'].encode(preferredencoding())
                        if self.params.get('forceurl', False):
-                               print info_dict['url'].encode(locale.getpreferredencoding())
+                               print info_dict['url'].encode(preferredencoding())
 
                        return
                        
@@ -309,7 +327,7 @@ class FileDownloader(object):
                        filename = self.params['outtmpl'] % template_dict
                except (ValueError, KeyError), err:
                        self.trouble('ERROR: invalid output template or system charset: %s' % str(err))
-               if self.params['nooverwrites'] and os.path.exists(filename):
+               if self.params.get('nooverwrites', False) and os.path.exists(filename):
                        self.to_stderr(u'WARNING: file exists: %s; skipping' % filename)
                        return
 
@@ -320,7 +338,7 @@ class FileDownloader(object):
                        return
 
                try:
-                       success = self._do_download(filename, info_dict['url'])
+                       success = self._do_download(filename, info_dict['url'].encode('utf-8'))
                except (OSError, IOError), err:
                        raise UnavailableFormatError
                except (urllib2.URLError, httplib.HTTPException, socket.error), err:
@@ -384,7 +402,7 @@ class FileDownloader(object):
                        resume_len = os.path.getsize(filename)
                else:
                        resume_len = 0
-               if self.params['continuedl'] and resume_len != 0:
+               if self.params.get('continuedl', False) and resume_len != 0:
                        self.report_resuming_byte(resume_len)
                        request.add_header('Range','bytes=%d-' % resume_len)
 
@@ -512,12 +530,13 @@ class YoutubeIE(InfoExtractor):
        _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'
        _NETRC_MACHINE = 'youtube'
-       _available_formats = ['22', '35', '18', '5', '17', '13', None] # listed in order of priority for -b flag
+       _available_formats = ['37', '22', '35', '18', '5', '17', '13', None] # listed in order of priority for -b flag
        _video_extensions = {
                '13': '3gp',
                '17': 'mp4',
                '18': 'mp4',
                '22': 'mp4',
+               '37': 'mp4',
        }
 
        @staticmethod
@@ -567,10 +586,6 @@ class YoutubeIE(InfoExtractor):
                """Report attempt to extract video information."""
                self._downloader.to_stdout(u'[youtube] %s: Extracting video information' % video_id)
        
-       def report_video_url(self, video_id, video_real_url):
-               """Report extracted video URL."""
-               self._downloader.to_stdout(u'[youtube] %s: URL: %s' % (video_id, video_real_url))
-       
        def report_unavailable_format(self, video_id, format):
                """Report extracted video URL."""
                self._downloader.to_stdout(u'[youtube] %s: Format %s not available' % (video_id, format))
@@ -696,7 +711,6 @@ class YoutubeIE(InfoExtractor):
                        video_real_url = 'http://www.youtube.com/get_video?video_id=%s&t=%s&eurl=&el=detailpage&ps=default&gl=US&hl=en' % (video_id, token)
                        if format_param is not None:
                                video_real_url = '%s&fmt=%s' % (video_real_url, format_param)
-                       self.report_video_url(video_id, video_real_url)
 
                        # uploader
                        mobj = re.search(r'(?m)&author=([^&]+)(?:&|$)', video_info_webpage)
@@ -706,11 +720,11 @@ class YoutubeIE(InfoExtractor):
                        video_uploader = urllib.unquote(mobj.group(1))
 
                        # title
-                       mobj = re.search(r'(?m)&title=([^&]+)(?:&|$)', video_info_webpage)
+                       mobj = re.search(r'(?m)&title=([^&]*)(?:&|$)', video_info_webpage)
                        if mobj is None:
                                self._downloader.trouble(u'ERROR: unable to extract video title')
                                return
-                       video_title = urllib.unquote(mobj.group(1))
+                       video_title = urllib.unquote_plus(mobj.group(1))
                        video_title = video_title.decode('utf-8')
                        video_title = re.sub(ur'(?u)&(.+?);', self.htmlentity_transform, video_title)
                        video_title = video_title.replace(os.sep, u'%')
@@ -1076,6 +1090,22 @@ if __name__ == '__main__':
                import getpass
                import optparse
 
+               # Function to update the program file with the latest version from bitbucket.org
+               def update_self(downloader, filename):
+                       # Note: downloader only used for options
+                       if not os.access (filename, os.W_OK):
+                               sys.exit('ERROR: no write permissions on %s' % filename)
+
+                       downloader.to_stdout('Updating to latest stable version...')
+                       latest_url = 'http://bitbucket.org/rg3/youtube-dl/raw/tip/LATEST_VERSION'
+                       latest_version = urllib.urlopen(latest_url).read().strip()
+                       prog_url = 'http://bitbucket.org/rg3/youtube-dl/raw/%s/youtube-dl' % latest_version
+                       newcontent = urllib.urlopen(prog_url).read()
+                       stream = open(filename, 'w')
+                       stream.write(newcontent)
+                       stream.close()
+                       downloader.to_stdout('Updated to version %s' % latest_version)
+
                # General configuration
                urllib2.install_opener(urllib2.build_opener(urllib2.ProxyHandler()))
                urllib2.install_opener(urllib2.build_opener(urllib2.HTTPCookieProcessor()))
@@ -1092,6 +1122,8 @@ if __name__ == '__main__':
                                action='help', help='print this help text and exit')
                parser.add_option('-v', '--version',
                                action='version', help='print program version and exit')
+               parser.add_option('-U', '--update',
+                               action='store_true', dest='update_self', help='update this program to latest stable version')
                parser.add_option('-i', '--ignore-errors',
                                action='store_true', dest='ignoreerrors', help='continue on download errors', default=False)
                parser.add_option('-r', '--rate-limit',
@@ -1144,7 +1176,7 @@ if __name__ == '__main__':
                parser.add_option_group(filesystem)
 
                (opts, args) = parser.parse_args()
-
+        
                # Batch file verification
                batchurls = []
                if opts.batchfile is not None:
@@ -1157,8 +1189,6 @@ if __name__ == '__main__':
                all_urls = batchurls + args
 
                # Conflicting, missing and erroneous options
-               if len(all_urls) < 1:
-                       parser.error(u'you must provide at least one URL')
                if opts.usenetrc and (opts.username is not None or opts.password is not None):
                        parser.error(u'using .netrc conflicts with giving username/password')
                if opts.password is not None and opts.username is None:
@@ -1191,7 +1221,7 @@ if __name__ == '__main__':
                        'forcetitle': opts.gettitle,
                        'simulate': (opts.simulate or opts.geturl or opts.gettitle),
                        'format': opts.format,
-                       'outtmpl': ((opts.outtmpl is not None and opts.outtmpl.decode(locale.getpreferredencoding()))
+                       'outtmpl': ((opts.outtmpl is not None and opts.outtmpl.decode(preferredencoding()))
                                or (opts.usetitle and u'%(stitle)s-%(id)s.%(ext)s')
                                or (opts.useliteral and u'%(title)s-%(id)s.%(ext)s')
                                or u'%(id)s.%(ext)s'),
@@ -1204,6 +1234,17 @@ if __name__ == '__main__':
                fd.add_info_extractor(youtube_pl_ie)
                fd.add_info_extractor(metacafe_ie)
                fd.add_info_extractor(youtube_ie)
+
+               # Update version
+               if opts.update_self:
+                       update_self(fd, sys.argv[0])
+
+               # Maybe do nothing
+               if len(all_urls) < 1:
+                       if not opts.update_self:
+                               parser.error(u'you must provide at least one URL')
+                       else:
+                               sys.exit()
                retcode = fd.download(all_urls)
                sys.exit(retcode)