# call program as so:
# cat inputfile | perl transform.pl > outputfile

sub get_word_tag {
  $line = $_[0];
  @list = split(/\s+/,$line);
}

sub is_be {
  $line = $_[0];
  @list = split(/\s+/,$line);
  $word = $list[0];
  if ($word =~ /\b(be|is|am|are|was|were|being|been)\b/) {
    return 1
  }
  return 0
}

while (<>) {

  # 1. DT becomes ART
  s/\bDT\b/ART/g;
  # 2. POS stays as POS
  # 3. VBZ becomes 3PR
  s/\bVBZ\b/3PR/g;
  # 4. irregular past tense (IRPST) needs to be hand-checked
  # 5. NNS becomes PL
  s/\bNNS\b/PL/g;
  # 6. Auxiliary becomes AUXBE & 7. Copula becomes COP
  if (&is_be($prev_line)) {
    if (/(NN|NNS|NNP|NNPS|JJ|JJR|JJS)/) {
      $prev_line =~ s/VB[DGNPZ]?/COP/g;
    }	
    elsif (/(VBG|VBN)/) {
      $prev_line =~ s/\bVB[DGNPZ]?\b/AUXBE/g;
    }
  }
  # 8. VBG becomes PROG
  s/\bVBG\b/PROG/g;

  $prev_line = $_;
  print $prev_line;
}