Thread: Regex replacing entire text rather then match

Hello,

With the code below, I am trying to replace all regex matches for visa cards within a given text file.

My first test was with a text "new3.txt" exclusively containing the visa test card 4111111111111111. My objective was to replace the card with "xxxx-xxxx-xxxx-xxxx". This was successful.

However, when modifying the text file to include other characters and text before and after (ex: " qwerty 4111111111111111 adsf zxcv"), it gives mixed results. Although it successfully validates the match, it replaces the whole text in the file, rather than replacing solely the match.

When trying this search and replace with words (rather than a regex match), it does not have this behavior. What am I missing?

import java.io.*;
import java.util.regex.*;
public class BTest
    {
	//VISA Test
     private final static String PATTERN = "(?s).*\\b4[0-9]{12}(?:[0-9]{3})?\\b.*";	
     public static void main(String args[])
         {
         try
             {
             File file = new File("c:/eclipse/new3.txt");
             BufferedReader reader = new BufferedReader(new FileReader(file));
             String line = "", oldtext = "";
             while((line = reader.readLine()) != null)
                 {
                 oldtext += line + "\r\n";
                 }
             reader.close();
             String newtext = oldtext.replaceAll(PATTERN, "xxxx-xxxx-xxxx-xxxx");
 
             FileWriter writer = new FileWriter("c:/eclipse/new4.txt");
             writer.write(newtext);writer.close();
         }
         catch (IOException ioe)
             {
             ioe.printStackTrace();
         }
     }
}

Welcome to the forum! Thank you for taking the time to learn to post code correctly.

Your regex seems a bit complicated for it's required task, I suggest looking it over.

Originally Posted by KucerakJM

Your regex seems a bit complicated for it's required task, I suggest looking it over.

I am following a regex which validates VISA cards, as specified by an article on regular-expression dot info called "Finding or Verifying Credit Card Numbers" (cant seem to post the link).

edit:..... and you were right... on each end, I only truly needed the "\\b"... Now it behaves correctly. Thanks!

It's worth mentioning that the article you mentioned assumed that all non-numeric, or at least spaces and dashes, were removed from the string. This meant that the regex they provided wasn't exactly what you were looking for, but very close.

Very good point.

That is precisely an issue I'm trying to fix... My current "lazy" approach is to remove all special characters between 4 numbers... It's causing slight issues with my data, because I'm modifying more than the desired matches... But the impact is minor, so it's o.k. for the moment...

And also I guess this topic is for another thread...

Thanks!