diff -Ncr ejSplitter-0.3/__init__.py ejSplitter-0.3.fix/__init__.py
*** ejSplitter-0.3/__init__.py	2004-01-04 15:54:45.000000000 +0900
--- ejSplitter-0.3.fix/__init__.py	2004-01-10 12:57:53.000000000 +0900
***************
*** 3,9 ****
  
  def parseTreeNodeInit(self, value):
      """ Retrieve unicode to 'utf-8'. value is alredy UnicodeType"""
!     self._value = value.encode('utf-8')
  
  from Products.ZCTextIndex.ParseTree import ParseTreeNode
  ParseTreeNode.__init__ = parseTreeNodeInit
--- 3,10 ----
  
  def parseTreeNodeInit(self, value):
      """ Retrieve unicode to 'utf-8'. value is alredy UnicodeType"""
!     self._value = value
!     #self._value = value.encode('utf-8')
  
  from Products.ZCTextIndex.ParseTree import ParseTreeNode
  ParseTreeNode.__init__ = parseTreeNodeInit
diff -Ncr ejSplitter-0.3/ejSplitter.py ejSplitter-0.3.fix/ejSplitter.py
*** ejSplitter-0.3/ejSplitter.py	2004-01-05 16:39:02.000000000 +0900
--- ejSplitter-0.3.fix/ejSplitter.py	2004-01-10 12:57:37.000000000 +0900
***************
*** 50,59 ****
                    r"&[A-Za-z]+;"]
          for pat in remove:
              text = re.sub(pat, " ", text)
- 	
          return splitJapanese(text, glob)
      def _convert(self, text):
!         return unicode(text, 'utf-8')
  
  class ejSplitterEUC_JP(ejSplitter):
      def _convert(self, text):
--- 50,61 ----
                    r"&[A-Za-z]+;"]
          for pat in remove:
              text = re.sub(pat, " ", text)
          return splitJapanese(text, glob)
+ 
      def _convert(self, text):
!         if type(text) != UnicodeType:
!             return unicode(text, 'utf-8', 'ignore')
!         return text
  
  class ejSplitterEUC_JP(ejSplitter):
      def _convert(self, text):