--- cdli/cdli_files.py	2010/03/19 14:01:41	1.115
+++ cdli/cdli_files.py	2011/04/27 16:19:27	1.117
@@ -306,10 +306,15 @@ def manage_addCDLIFile(self,id,title,loc
 
 def checkUTF8(data):
     """check utf 8"""
+    if not isinstance(data, str):
+        logging.error("checkUTF8 data is not string! (%s)"%repr(data))
+
     try:
-        data.encode('utf-8')
+        data.decode('utf-8')
+        logging.debug("checkUTF8: ok!")
         return True
     except:
+        logging.debug("checkUTF8: false!")
         return False
     
 
@@ -335,7 +340,7 @@ def splitatf(fh,dir=None,ext=None):
     i=0
 
     #ROC: why split \n first and then \r???
-    if (type(fh) is StringType) or (type(fh) is UnicodeType):
+    if isinstance(fh, basestring):
         iter=fh.split("\n")
     else:
         iter=fh.readlines()
@@ -369,7 +374,7 @@ def splitatf(fh,dir=None,ext=None):
                     if dir:
                         filename=os.path.join(dir,filename)
                     nf=file(filename,"w")
-                    logging.info("open %s"%filename)
+                    logging.debug("open %s"%filename)
                 if nf:    
                     nf.write(line.replace("\n","")+"\n")
 
@@ -378,8 +383,9 @@ def splitatf(fh,dir=None,ext=None):
     except:
         pass
     
-    if not((type(fh) is StringType) or (type(fh) is UnicodeType)):
+    if not isinstance(fh, basestring):
         fh.close()
+        
     return ret,len(os.listdir(dir))
 
 
@@ -888,8 +894,9 @@ class CDLIRoot(Folder):
 
 
 
-    def searchText(self, query, index='graphemes'):
-        """searches query in the fulltext index and returns a list of file ids/P-numbers"""
+    def searchText(self, query, index='graphemes', resultFilter=None):
+        """searches query in the fulltext index and returns a list of file IDs/P-numbers
+           resultFilter is matched against the beginning of the file ID"""
         # see also: http://www.plope.com/Books/2_7Edition/SearchingZCatalog.stx#2-13
         logging.debug("searchtext for '%s' in index %s"%(query,index))
         #import Products.ZCTextIndex.QueryParser
@@ -899,8 +906,11 @@ class CDLIRoot(Folder):
         idx = getattr(self, self.file_catalog)
         # do search
         resultset = idx.search(query_request=idxQuery,sort_index='textid')
-        # put only the P-Number in the result 
-        results = [res.getId[:7] for res in resultset]
+        # put only the P-Number in the result
+        if resultFilter is None:
+            results = [res.getId[:7] for res in resultset]
+        else:
+            results = [res.getId[:7] for res in resultset if res.getId.startswith(resultFilter)]
         logging.debug("searchtext: found %d texts"%len(results))
         return results