# Wds.pm Daniel Brockman 070223 Wordcounting
# Creative Commons Attribution License.
# Permission granted to copy, use or modify, with attribution to author.
#
# Usage: use Wds.pm
#

use strict;

#__________________________________________________________
sub wfil {
  # 
  #
  # Usage: $rfreq = &wfil($rf,$rlines)
  #
  # $rlines is the reference to an array, to which wfil
  # appends, that contains the lines read from a file.
  #
  # Either
  # $rf is a reference to a file handle
  # or
  # $rf is a file name.
  # if $rf is a reference to a file handle, then the file 
  # must be open already on the file handle, and sub wfil won't close it.
  # if $rf is a file name, then sub wfil will open it, read it and close it.
  # wfil first assumes $rf is a file name and attempts to open it.
  # if wfil can't open the file, then wfil assumes $rf is a file handle
  # and attempts to read from it, and if the file isn't open, then
  # a fatal error will occur.
  #
  # wfil appends the content of the text file to array @$rlines.
  #
  # wfrq.pl Daniel Brockman 20030828 Word Frequency Counter
  # modified 070223
  
  my (
    $rf,                        # text input file name or handle
    $rlines,                    # ref to array of lines from file
    ) = @_;  
  
  my (
    @raw,     # lines read from file
    $i,$j,$k, # aux
    $fn,      # file name
    $rfh,     # ref to file handle
    $isfile,  # flag = 1 if $rf is file name
  );
  
  $isfile=1;                        # assume rf is a file name
  if (!open (FH,$rf)) {             # attempt to open file
    $isfile=0;                      # rf isn't a file
    $rfh=$rf;                       # we'll use $rf as a file handle
  }
  if ($isfile) {$rfh=\*FH;}         # rf is file name, rfh is handle
  
  @raw=<$rfh> ;             # suck in the whole file
  push(@$rlines,@raw) ;     # append to passed array

  if ($isfile) {close($rfh)}           # close file if opened
  
} # end sub wfil
#__________________________________________________________
#__________________________________________________________
sub wftl {
  # 
  #
  # Usage: $rfreq = &wftl($rfreq,$rword)
  #
  # $rword is the reference to a hash, 
  # for which the keys are numeric counts of word frequency.
  # One way to construct %$rfreq is by sub wfin.
  # Hash element order not guaranteed.
  # One way to obtain order is 
  #    @sortord = sort { $b <=> $a } keys %{$rword}
  #
  # $rfreq is the reference to a hash, to which wftl appends,
  # for which the keys are words.
  # Hash element order not guaranteed.
  # One way to obtain order is 
  #    @sortord = sort keys %{$rfreq}
  #
  # wfrq.pl Daniel Brockman 20030828 Word Frequency Counter
  # modified 070223
  
  my (
    $rfreq,                     # hash ref, { count => word, ... }
    $rword,                     # hash ref, { word => count, ... }
    ) = @_;  
  
  my (
    #@content, # file content
    #@token,   # line content
    $i,$j,$k, # aux
    #@list,    # aux
    $word, $line,   # items read from line, from file
    @keys,    # hash keys
    #  @numkeys, # number keys
  );
  
  
  @keys=keys %{$rword};        # get hash keys (words)
  for $word (@keys) {       # take each unique word
  
    $j=${$rword}{$word}; # count of occurrences of word
    if((!defined(${$rfreq}{$j})) or
       (${$rfreq}{$j} eq "")
     ){  # unranked?
      ${$rfreq}{$j}=$word;    # init list
    } else {
      ${$rfreq}{$j}=${$rfreq}{$j}." ".$word;  # append to list
    }  # if {$rfreq} j
  }  # for word keys
  
  ##################### test #############################
  ### write list of frequencies
  ##
  ##for $i (@numkeys) {
  ##  for $word (@list=split(/ /,${$rfreq}{$i})) {
  ##    print "$word $i\n";
  ##  } # for word
  ##} # for i numkeys
  ################### end test ###########################
  
} # end wftl
#
#__________________________________________________________
#__________________________________________________________
sub wfee {
  # 
  #
  # Usage: $rfreq = &wfee($rlines,$rword)
  #
  #
  # $rlines is the reference to an array
  # that contains the lines read from a file.
  # sub wfee doesn't change @$rlines.
  # One way to create the array @$rlines is to use
  # sub wfil.
  #
  # $rword is the reference to a hash, to which wfee appends,
  # for which the keys are numeric counts of word frequency.
  # Hash element order not guaranteed.
  # One way to obtain order is 
  #    @sortord = sort { $b <=> $a } keys %{$rword}
  #
  # wfee separates words on any character other than a 
  # digit or a letter, ignoring all punctuation. wfee 
  # ignores very frequently occurring words including "a", 
  # "and" and "the" and some others. 
  #
  # wfee considers two words as
  # the same word if their first $nl letters are the same.
  # Only the first $nl characters are shown in hash %$rfreq.
  #
  # $nl is a parameter currently set to 6.
  #
  # wfrq.pl Daniel Brockman 20030828 Word Frequency Counter
  # modified 070223
  
  my (
    $rlines,                    # ref to array of lines from file
    $rword,                     # hash ref, { word => count, ... }
    ) = @_;  
  
  my (
    @token,   # line content
    $i,$j,$k, # aux
    $word, $line,   # items read from line, from file
    $common,  # flag for common words
    @commoners,  # common words
    $nl,      # number of letters in a "word"
    $fn,      # file name
    $rfh,     # ref to file handle
    $isfile,  # flag = 1 if $rf is file name
  );
  
  # init
  
  $nl=6;
  
  # list of common words
  
  @commoners=(
    "a","and","the","of","to","in","with","for","or","is",
    "this","as","will","it","be","are","an","s"
  );
  
  foreach $line (@$rlines) {           # loop the lines
    if (!defined($line)){next}         # skip unassigned
    $line=~tr/A-Z/a-z/ ;               # lowercase
    @token=split(/[^0-9a-z]+/,$line);  # separate out words
    for $word (@token) {               # loop the words
      $word=substr($word,0,$nl);       # ch in one word
      $common=0;                       # assume not common
      for $j (@commoners) {            # compare with common words
        if($common=($word eq $j)){ last };
      };                               # for j commoners
      if($common) { next };            # skip if common
      ${$rword}{$word}++;              # count occurrences
      $i = ${$rword}{$word};           # test

    } # for $word
  } # foreach $line

} # end sub wfee
#__________________________________________________________
#__________________________________________________________
#__________________________________________________________
1