1 files changed, 24 insertions, 6 deletions
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 839da17d0..1f60d34ae 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -26,6 +26,11 @@ std_headers = {
 	'Accept-Language': 'en-us,en;q=0.5',
 }
 
+try:
+    compat_str = unicode # Python 2
+except NameError:
+    compat_str = str
+
 def preferredencoding():
 	"""Get preferred encoding.
 
@@ -83,7 +88,6 @@ class IDParser(HTMLParser.HTMLParser):
 		HTMLParser.HTMLParser.__init__(self)
 
 	def error(self, message):
-		print >> sys.stderr, self.getpos()
 		if self.error_count > 10 or self.started:
 			raise HTMLParser.HTMLParseError(message, self.getpos())
 		self.rawdata = '\n'.join(self.html.split('\n')[self.getpos()[0]:]) # skip one line
@@ -190,14 +194,28 @@ def timeconvert(timestr):
 	if timetuple is not None:
 		timestamp = email.utils.mktime_tz(timetuple)
 	return timestamp
-	
-def sanitize_filename(s):
-	"""Sanitizes a string so it could be used as part of a filename."""
+
+def sanitize_filename(s, restricted=False):
+	"""Sanitizes a string so it could be used as part of a filename.
+	If restricted is set, use a stricter subset of allowed characters.
+	"""
 	def replace_insane(char):
-		if char in u' .\\/|?*<>:"' or ord(char) < 32:
+		if char == '?' or ord(char) < 32 or ord(char) == 127:
+			return ''
+		elif char == '"':
+			return '' if restricted else '\''
+		elif char == ':':
+			return '_-' if restricted else ' -'
+		elif char in '\\/|*<>':
+			return '-'
+		if restricted and (char in '&\'' or char.isspace()):
 			return '_'
 		return char
-	return u''.join(map(replace_insane, s)).strip('_')
+
+	result = u''.join(map(replace_insane, s))
+	while '--' in result:
+		result = result.replace('--', '-')
+	return result.strip('-')
 
 def orderedSet(iterable):
 	""" Remove all duplicates from the input iterable """