Step 1, Get AWK. You could do this in Perl, but AWK is simpler. Myself, I
like cygwin (http://sources.redhat.com/cygwin) but that requires you to
install a "system". You'd probably be better off with one of the Win32 ports
of gAWK see http://www.faqs.org/faqs/computer-lang/awk/faq/


Step 2. Put this program in a file (say wordcount.awk):
BEGIN { RS="[ \t\n]" }
/^$/  { next; }
      {
#     word = $ 0    # use this for case-sensitive
      word = toupper($0)  # use this for not case-sensitive
      count[word]=count[word]+1;
      }
END   {
      for (x in count) {
        print x " " count[x];
        }
      }


Step 3. Execute: awk -f wordcount.awk test.txt

If test.txt has is in it:
This is a word counter
A word counter is this

Then the output will be:
A 2
IS 2
COUNTER 2
THIS 2
WORD 2

If you don't want to ignore case, just move the # side at the left margin
down to the next line. Then you'd get:

A 1
this 1
a 1
This 1
counter 2
word 2
is 2

Good luck!

Al Williams
AWC
* Floating point math for any microprocessor:
http://www.al-williams.com/awce/pak1.htm

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.