--- cdli/cdliSplitter.py	2007/12/11 17:33:07	1.7.2.8
+++ cdli/cdliSplitter.py	2007/12/13 19:20:45	1.7.2.9
@@ -24,11 +24,12 @@ def getSupportedEncoding(encodings):
 ignoreLines=['$','@','#','&','>']
 separators=['']
 # kommas relevant for graphemes will not be deleted
-komma_exception="([^sStThH])," 
+komma_exception="([^sStThH]),"
+komma_exceptionex=re.compile(komma_exception)
 # grapheme boundaries
 #graphemeBounds="\{|\}|<|>|\(|\)|-|_|\#|,|\||\]|\[|\!|\?"
 graphemeBounds="\{|\}|<|>|-|_|\#|,|\]|\[|\!|\?|\""
-graphemeIgnore=""
+graphemeIgnore="<|>|\#|\||\]|\[|\!|\?"
 # for words 
 #wordBounds="<|>|\(|\)|_|\#|,|\||\]|\[|\!|\?"
 wordBounds="_|,|\""
@@ -41,7 +42,9 @@ class cdliSplitter:
     
     default_encoding = "utf-8"
     bounds=graphemeBounds
+    boundsex=re.compile(graphemeBounds)
     ignore=graphemeIgnore
+    ignorex=re.compile(graphemeIgnore)
     indexName="cdliSplitter"
     
     
@@ -80,11 +83,11 @@ class cdliSplitter:
                             lineNum=lineparts[0] 
                             
                         # delete kommata except kommata relevant for graphemes
-                        txt = re.sub(komma_exception,r"\1",txt)
-                        # replace letters to be ignored
-                        txt = re.sub(self.ignore,'',txt)
+                        txt = komma_exceptionex.sub(r"\1",txt)
                         # replace word boundaries by spaces
-                        txt = re.sub(self.bounds,' ',txt)
+                        txt = self.boundsex.sub(' ',txt)
+                        # replace letters to be ignored
+                        txt = self.ignorex.sub('',txt)
                         # split words
                         words = txt.split(" ")
                         for w in words:
@@ -98,12 +101,16 @@ class cdliSplitter:
 
 class graphemeSplitter(cdliSplitter):
     bounds=graphemeBounds
+    boundsex=re.compile(graphemeBounds)
     ignore=graphemeIgnore
+    ignorex=re.compile(graphemeIgnore)
     indexName="graphemeSplitter"
     
 class wordSplitter(cdliSplitter):
     bounds=wordBounds
+    boundsex=re.compile(wordBounds)
     ignore=wordIgnore
+    ignorex=re.compile(wordIgnore)
     indexName="wordSplitter"
       
 try: