Bug 26826

Oops, forgot to remove my last printlns. Below a fresh patch:

Index: Util.java
===================================================================
RCS file:
/home/cvspublic/jakarta-taglibs/standard/src/org/apache/taglibs/standard/tag/common/core/Util.java,v
retrieving revision 1.12
diff -u -r1.12 Util.java
--- Util.java	13 Dec 2003 05:48:07 -0000	1.12
+++ Util.java	10 Feb 2004 12:16:08 -0000
@@ -152,23 +152,45 @@
      * See also OutSupport.out().
      */
     public static String escapeXml(String input) {
-        StringBuffer sb = new StringBuffer();
-        for (int i = 0; i < input.length(); i++) {
-            char c = input.charAt(i);
-            if (c == '&')
-                sb.append("&amp;");
-            else if (c == '<')
-                sb.append("&lt;");
-            else if (c == '>')
-                sb.append("&gt;");
-            else if (c == '"')
-                sb.append("&#034;");
-            else if (c == '\'')
-                sb.append("&#039;");
-            else
-                sb.append(c);
-        }
-        return sb.toString();
+		StringBuffer sb = null;
+		int length = input.length();
+		int currentPos;
+
+		//Check if there are any special characters to escape
+		for (currentPos = 0; currentPos < length; currentPos++) {
+			char c = input.charAt(currentPos);
+			if (c == '&' || c == '<' || c == '>' || c == '"' || c == '\'') {
+				sb = new StringBuffer(length+5);
+				sb.append(input.substring(0,currentPos)); // copy over the string up until
we found this char
+				break;
+			}
+		}
+
+		// If we didn't create a new buffer, then the input string didn't need any
escaping so we win big time
+		// and we can just return the same string they gave us (no object creation,
no copying).
+		if ( sb == null )
+		{
+			return input;
+		}
+		
+
+		// Oh well, we did have some escaping to do, so let's check the rest to see
if there are any more to do.
+		for (; currentPos < length;++currentPos) {
+			char c = input.charAt(currentPos);
+			if (c == '&')
+				sb.append("&amp;");
+			else if (c == '<')
+				sb.append("&lt;");
+			else if (c == '>')
+				sb.append("&gt;");
+			else if (c == '"')
+				sb.append("&#034;");
+			else if (c == '\'')
+				sb.append("&#039;");
+			else
+				sb.append(c);
+		}
+		return sb.toString();
     }  
     
     /**

My feedback from taglibs-dev (http://marc.theaimsgroup.com/?t=107639893000003):

It looks like most of the speedup of the new version can be gained by just
making the old version take the input length into account (based on my own
limited benchmarking, YMMV):

public static String escapeXml(String input) {
  String output = input;
  if (input != null) {
    int length = input.length();
    StringBuffer sb = new StringBuffer(length);
    for (int i = 0; i < length; i++) {
      char c = input.charAt(i);
      if (c == '&') sb.append("&amp;");
      else if (c == '<') sb.append("&lt;");
      else if (c == '>') sb.append("&gt;");
      else if (c == '"') sb.append("&#034;");
      else if (c == '\'') sb.append("&#039;");
      else sb.append(c);
    }
    output = sb.toString();
  }
  return output;
}

I suppose you could also add a test for input.length() == 0. If the time savings
aren't really that significant, I'd opt for simpler, easier to read code. BTW,
any modification should account for input == null. It doesn't make sense to
throw an NPE.

CC'ing the taglibs-dev address to all Standard bugs. 

*** This bug has been marked as a duplicate of 25382 ***