Changeset 2179

Timestamp:

19/02/07 13:34:46 (13 years ago)

Author:

domlowe

Message:

Improved performance of two Nappy functions - readItemsFromUnknownLines and _checkForBlankLines

Location:

nappy/trunk

Files:

• naFile.py (modified) (2 diffs)
• textParser.py (modified) (4 diffs)

nappy/trunk/naFile.py

@@ -113 +113 @@
         rtlines=[]
         for line in datalines:
-            if re.match(r"^[\s\n\t\r]*$", line):
+            #if re.match(r"^[\s\n\t\r]*$", line):
+            if not line.strip(): #looks for blank line, quicker than reg ex
                 empties=1
             else:
@@ -274 +275 @@
             datalines=self._readData1(datalines, m)
             datalines=self._readData2(datalines, m)
-            m=m+1    
-
+            m=m+1
+            

nappy/trunk/textParser.py

@@ -19 +19 @@
 def readItemFromLine(line, rttype=str):
     rtitem=pattnNoQuotes.sub("", line.strip())
-    if rttype==int:
-        rtitem=int(rtitem)
-    elif rttype==float:
-        rtitem=float(rtitem)        
+    if rttype is not str:
+        rtitem = rttype(rtitem)   
     return rtitem
 
@@ -29 +27 @@
     if nitems and len(rtitems)!=nitems:
         raise "Incorrect number of items (%s) found in line: \n'%s'" % (nitems, line)
-    if rttype==int:
-        rtitems=map(lambda x: int(x), rtitems)
-    elif rttype==float:
-        rtitems=map(lambda x: float(x), rtitems)        
+    if rttype is not str:
+        rtitems = [rttype(x) for x in rtitems]
     return rtitems
 
@@ -39 +35 @@
     for line in lines:
         rtitems=rtitems+[readItemFromLine(line, rttype)]
-    if rttype==int:
-        rtitems=map(lambda x: int(x), rtitems)
-    elif rttype==float:
-        rtitems=map(lambda x: float(x), rtitems)            
+    if rttype is not str:
+        rtitems = [rttype(x) for x in rtitems]
     return rtitems
 
@@ -53 +47 @@
     """
     rtitems=[]
-    lines=[]    
-    if type(object)==list:
-        while len(rtitems)<nitems:
-            nextitem=object[0]
-            object=object[1:]
-            items=re.split(r"\s+", nextitem.strip())
-            lines.append(items)        
-            (rtitems, extras)=(rtitems+items[:nitems], items[nitems:])
+    lines=[]        
+    if type(object) is list: 
+        nextitem=object[0]
+        del object[0]
+        while len(rtitems)<nitems:   
+            #items=re.split(r"\s+", nextitem.strip())
+            items=nextitem.strip().split()
+            lines.append(items)
+            (rtitems,extras)=(rtitems+items[:nitems],items[nitems:])
         if len(extras)>0:
             raise "Could not split %s lines exactly into required number (%s) of items: \n%s" % (len(lines), nitems, lines)
-        if rttype==int:
-            rtitems=map(lambda x: int(x), rtitems)
-        elif rttype==float:
-            rtitems=map(lambda x: float(x), rtitems)
+        if rttype is not str:
+            rtitems = [rttype(x) for x in rtitems]
         return (rtitems, object)
     else:
+        lines =[]
         while len(rtitems)<nitems:
-            items=re.split(r"\s+", object.readline().strip())
-            lines.append(items)        
-            (rtitems, extras)=(rtitems+items[:nitems], items[nitems:])
+            #items=re.split(r"\s+", object.readline().strip())
+            items=object.readline().strip().split()
+            lines.append(items)
+            (rtitems,extras)=(rtitems+items[:nitems], items[nitems:])
         if len(extras)>0:
             raise "Could not split %s lines exactly into required number (%s) of items: \n%s" % (len(lines), nitems, lines)
-        if rttype==int:
-            rtitems=map(lambda x: int(x), rtitems)
-        elif rttype==float:
-            rtitems=map(lambda x: float(x), rtitems)     
+        if rttype is not str:
+            rtitems = [rttype(x) for x in rtitems]
         return rtitems