finding token data type (StringTokenizer)

[Pip Vo]

Does anyone know how to find the data type of
st.nextToken in the code below? I can't use
streamTokenizer's TT_WORD or TT_NUMBER b/c need
to parse words like G7790 that have both #s and
letters and streamTokenizer gets confused. Need something specific to stringTokenizer. Searched Sun's page w/ no luck. TIA

 try
  {
             FileReader f = new FileReader("tile1.txt");
             BufferedReader b = new BufferedReader(f);
             
             while ((line = b.readLine()) != null)
     {
                data = 0;
                StringTokenizer st = new StringTokenizer(line);
                while (st.hasMoreTokens())
        {
                   System.out.println("DATA: " +st.nextToken());
                   if (data <= 1) // need to find nextToken's data type to put in name1 or data1 array
      {name1[count1] = st.nextToken(); count1++;}
                   else
   {data1[count2] = st.nextToken(); count2++;}
                   data++;
            }
                } 
            }
          catch(IOException e) {System.out.println("ERROR!");
        }

[ January 29, 2002: Message edited by: Cindy Glass ]

[Cindy Glass]

What do you WANT G7790 to parse to?
Have you looked into using a StringTokenizer, perhaps inside the StreamTokenizer calls?

[Pip Vo]

Cindy,
Thanks for the reply.
I need to find out what tokens are doubles
and put them in the data1 array. Anything else
(like G7889) goes into the name1 array.
I've used StreamTokenizer in an earlier version,
and something like G7889 gets split up into G and 7889 which I don't want. So am not sure about
using again and trying to put StringTokenizer
inside it.
Pip

[Cindy Glass]

Oh, I think that I get it.
Can't you just use StringTokenizer, and then check each token to see if you can Long.parseLong(st.nextToken()) and if it pulls a NumberFormatException - then put it in as a String?