Float Conversion

Question

I am passing 2 strings of data Social Security Number and a string to be parsed
1) SSN 123456789
2)String PH80.00 170915.76 AAE
The string is parsed PH is extracted, 80.00 is extracted and 170915.76 is converted, however, it is converted to 170916.77. .01 is added to the value. Why does this occur? I think it has to do with the float. The code is as follows
* Created on Jun 13, 2007
* To change the template for this generated file go to
* Window>Preferences>Java>Code Generation>Code and Comments
package com.abbott.HRT.US018;
* @author Administrator
* To change the template for this generated type comment go to
* Window>Preferences>Java>Code Generation>Code and Comments
public class TransformData {
     * The below Function is for handling Compensation Update Records
     */ 2 examples of string passed MI80.00 242050.00 AAE Output produced is 242050.00
PH80.00 170915.76 AAE Output produced is 170915.77
     private String compensationUpdate(String sRecord) {
          //1.Map directly the source data to target data and make changes to below lengths.
          StringBuffer sBuf = new StringBuffer(sRecord);
          //          Correction : Make the length of sBuf equal to 21 in order to reduce the risk of runtime exception
          if (sBuf.length() < 21) {
               while (sBuf.length() != 21) {
                    sBuf.append(' ');
          //2.     Schedule Hours : Format the data into "ZZ9.99" at position 3-8
          String substring = sBuf.substring(2, 8);
          substring = formatNumber(substring, 4, 6, 2);
          sBuf.replace(2, 8, substring);
          //3.     Base Rate : Format the data into "ZZZZZZ9.99" at position 9-29. The target position will be 9-18
          substring = null;
          substring = sBuf.substring(8, 29);
          substring = formatNumber(substring, 8, 10, 2);
          sBuf.replace(8, 18, substring.substring(0, 10));
          sBuf.delete(18, 29);
          //4.     Performance Rating : For now no mapping is defined
          //At last append 434 spaces to the record
          sBuf = appendSpaces(sBuf, 434);
          return sBuf.toString();
     * The below Function is for formatting numbers Records
     private String formatNumber(
          String input,
          int position,
          int length,
          int decPlaces)
          throws NumberFormatException {
          //17.08.2007 : If empty string has come, send back the same
          int int1 = input.lastIndexOf(' ');
          int int2 = input.length();
          if (int1 == int2) {
               return input;
          // position = actual position of dec place (e.g.4 for ZZ9.99)
          //length = actual length (e.g. 6 for ZZ9.99)
          //decPlaces = (e.g. 2 for ZZ9.99)
          //First get the input converted into float and back so as to add a decimal sign if it does not exist
          float f;
          try {
               f = Float.parseFloat(input);
          } catch (NumberFormatException ex) {
               throw new NumberFormatException(
                    "Exception Occurred:" + ex.getMessage());
          StringBuffer sb = new StringBuffer(length);
          sb.insert(0, f);
          //Identify the position of decimal sign and then shift it gradually
          int index = sb.indexOf(".");
          if (index < position && index != -1) {
               while (index < (position - 1)) {
                    sb.insert(0, " ");
                    index = sb.indexOf(".");
               /*If there is no decimal sign, put one at the desired position                         
                                        if(index == -1){
                                             int int1 = Integer.parseInt(input);
                                             sb.ins;
                                             sb.insert(position,".");
          //Fill blank spaces after decimal sign with zero
          while (sb.length() < length) {
               sb.append("0");
          return sb.toString();
The code is as follows

Answer 1

Volume is not precision. You need to be precise and informative. This end is not served by simply dumping huge volumes of code or data into a help request. If you have a large, complicated test case that is breaking a program, try to trim it and make it as small as possible.
This is useful for at least three reasons. One: being seen to invest effort in simplifying the question makes it more likely that you'll get an answer, Two: simplifying the question makes it more likely you'll get a useful answer. Three: In the process of refining your bug report, you may develop a fix or workaround yourself.
-- http://www.catb.org/~esr/faqs/smart-questions.html
Example:
assert 170915.77F == Float.parseFloat("170915.76"); // why?And here's the answer:
http://en.wikipedia.org/wiki/Floating_point
http://java.sun.com/developer/JDCTechTips/2003/tt0204.html#2
http://docs.sun.com/source/806-3568/ncg_goldberg.html
Cheers!
~