Tokenizing Error: java.util.regex.PatternSyntaxException, dangling metacharacter '*'

Question

I am using split() to tokenize a String separated with * following this format:

name*lastName*ID*school*age
%
name*lastName*ID*school*age
%
name*lastName*ID*school*age

I'm reading this from a file named "entrada.al" using this code:

static void leer() {

    try {
        String ruta="entrada.al";
        File myFile = new File (ruta);
        FileReader fileReader = new FileReader(myFile);

        BufferedReader reader = new BufferedReader(fileReader);

        String line = null;

        while ((line=reader.readLine())!=null){
            if (!(line.equals("%"))){
                String [] separado = line.split("*"); //SPLIT CALL
                names.add(separado[0]);
                lastNames.add(separado[1]);
                ids.add(separado[2]);
                ages.add(separado[3]);
            }
        }

        reader.close();
    }

And I'm getting this exception:

Exception in thread "main" java.util.regex.PatternSyntaxException: Dangling meta character '*' near index 0 *

My guess is that the lack of a * after age on the original text file is causing this. How do I get around it?

Answer 1

No, the problem is that * is a reserved character in regexes, so you need to escape it.

String [] separado = line.split("\\*");

* means "zero or more of the previous expression" (see the Pattern Javadocs), and you weren't giving it any previous expression, making your split expression illegal. This is why the error was a PatternSyntaxException.

Answer 2

I had similar problem with regex = "?". It happens for all special characters that have some meaning in a regex. So you need to have "\\" as a prefix to your regex.

String [] separado = line.split("\\*");

Answer 3

The first answer covers it.

Im guessing that somewhere down the line you may decide to store your info in a different class/structure. In that case you probably wouldn't want the results going in to an array from the split() method.

You didn't ask for it, but I'm bored, so here is an example, hope it's helpful.

This might be the class you write to represent a single person:

class Person {
            public String firstName;
            public String lastName;
            public int id;
            public int age;

      public Person(String firstName, String lastName, int id, int age) {
         this.firstName = firstName;
         this.lastName = lastName;
         this.id = id;
         this.age = age;
      }  
      // Add 'get' and 'set' method if you want to make the attributes private rather than public.
} 

Then, the version of the parsing code you originally posted would look something like this: (This stores them in a LinkedList, you could use something else like a Hashtable, etc..)

try 
{
    String ruta="entrada.al";
    BufferedReader reader = new BufferedReader(new FileReader(ruta));

    LinkedList<Person> list = new LinkedList<Person>();

    String line = null;         
    while ((line=reader.readLine())!=null)
    {
        if (!(line.equals("%")))
        {
            StringTokenizer st = new StringTokenizer(line, "*");
            if (st.countTokens() == 4)          
                list.add(new Person(st.nextToken(), st.nextToken(), Integer.parseInt(st.nextToken()), Integer.parseInt(st.nextToken)));         
            else            
                // whatever you want to do to account for an invalid entry
                  // in your file. (not 4 '*' delimiters on a line). Or you
                  // could write the 'if' clause differently to account for it          
        }
    }
    reader.close();
}

Answer 4

It is because * is used as a metacharacter to signify one or more occurences of previous character. So if i write M* then it will look for files MMMMMM..... ! Here you are using * as the only character so the compiler is looking for the character to find multiple occurences of,so it throws the exception.:)