Assuming the text file is small enough to fit in memory and that you're
using only characters in the basic multilingual plane (e.g. no Ancient
Sumerian characters), the way I'd do this is to read in every line as a
String and keep them in an array of Strings. Then, for each line, I'd use
the charat() method to extract the character at a specific column within
that line.

   - Oliver

Your column values are really made of strings that represent device id and
address, right? Something like:

DEV-1  63412
DEV-23 45453

Given the device id you want the address.  If that's the case, I suggest a
simple reader to read the file, separate the columns by the space and put
the values into a Map indexed by the id.  Are there more than 2 columns?  Is
the separator always a space?

If it's just what's laid out above, it's really as simple as this (not
compiled or tested, sorely lacking in error handling.)

Map lookupTable = new HashMap ();
BufferedReader br = new BufferedReader (new FileReader (...));
String line = null;
while ((line = br.readLine()) != null) {
 int p = line.indexOf (" ");
 if (p > 0) {
   String deviceName = line.substring (0, p);
   String address = line.substring (p + 1);
   lookupTable.put (deviceName, address);
 }
}
br.close ();

If you have more columns there are more sophisticated ways to pick off the
values, such as using regular expressions (which can be good for handling
less perfect data).  It really depends on what you want to do with the
values.  If you really just want them as column values, that's possible too:

List rows = new ArrayList ();
BufferedReader br = new BufferedReader (new FileReader (...));
String line = null;
while ((line = br.readLine()) != null) {
 String [] row = line.split (" ");
 rows.add (row);
}
br.close ();

You can pick off any row from the list as an array of strings and pick off
the column value starting with 0 for the first column.

Matt Humphrey matth@ivizNOSPAM.com http://www.iviz.com/