Handle weird OSX locale settings gracefully (fixes issue #43)

diff --git a/youtube-dl b/youtube-dl
index becf1d25de31ccbdbaaef76b517bb23490a9c001..472a8e1ccc677318be39d328c945be37b04a8435 100755 (executable)
--- a/youtube-dl
+++ b/youtube-dl
@@ -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.
+       """
+       try:
+               pref = locale.getpreferredencoding()
+               # Mac OSX systems have this problem sometimes
+               if pref == '':
+                       return 'UTF-8'
+               return pref
+       except:
+               sys.stderr.write('WARNING: problem obtaining preferred encoding. Falling back to UTF-8.\n')
+               return 'UTF-8'
+
 class DownloadError(Exception):
        """Download Error exception.
        
@@ -224,12 +240,12 @@ 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."""
@@ -297,9 +313,9 @@ class FileDownloader(object):
 
                        # 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
                        
@@ -1191,7 +1207,7 @@ 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'),