#! perl
#
# wfrq.pl Daniel Brockman 20030828 Word Frequency Counter
# Permission granted to copy with attribution to author.
#
# Usage: wfrq.pl < inputfile > outputfile
#
# wfrq reads the content of inputfile, counts the 
# occurrences of words and shows the most frequently
# used words and the number of times that each occurred.
# wfrq separates words on any character other than a 
# digit or a letter, ignoring all punctuation. wfrq 
# ignores very frequently occurring words including "a", 
# "and" and "the" and some others. Two words are considered
# the same word if their first $nl letters are the same.
# Only the first $nl characters are shown on the output 
# list.
#
# $nl is a parameter currently set to six.
#
# revised 070223 to use Wds.pm.

BEGIN {@INC=(".",@INC)} ; # test

use strict;
use Wds;

my (
  %tally,   # counts of occurrences by word
  %retally, # list of words by number of occurrences
  $i,$j,$k, # aux
  @list,    # aux
  $word,    # contains a word split out of %retally
  @numkeys, # number keys
  @lines,   # array of lines read from file
);

{ no strict;       # disable strict in this block so we can use the typeglob
&wfil(*STDIN,\@lines);             # append content of standard input to @lines
}

&wfee(\@lines,\%tally);            # count the occurrences

&wftl(\%retally,\%tally);          # group words by frequency

# reverse sort %retally
#@numkeys = sort { $a <=> $b } keys %retally ;
@numkeys = sort { $b <=> $a } keys %retally ;          # most frequent first

# write list of frequencies

for $i (@numkeys) {
  for $word (@list=split(/ /,$retally{$i})) {
    print "$word $i\n";
  } # for word
} # for i numkeys

#