#!/usr/bin/perl
# Copyright 13th. October 2008, Cotman University 
# http://www.cotmanuniversity.com
# Created by Dr. Peter Kjeldsen
# Department for Computational Linguistics
# Conversion of CVS format to WordFast TM format
# File name: csvwf.pl
#
# Instructions
# Change name of input file, source and target language
# as required.
# Run from command prompt perl csvwf.pl  
#

# Setting up formated variable for handling date and time
($sec, $min, $hour, $mday, $mon, $year) = localtime(time);

$mon++;
$year+= 1900;
$hour = sprintf("%02d", $hour);
$min = sprintf("%02d", $min);
$sec = sprintf("%02d", $sec);

$timestamp = "$year" . "$mon" . "$mday" . "~" . "$hour$min$sec";
$creationtool = "Lingualizer tmXl8r";
$userid = "PK";
$srclang = "EN-US";
$tgtlang ="DA-01";

# Opening input file
open (TXT, "<ordliste.csv") or die ("Cannot open file!");
@text = <TXT>;
close (TXT);

# Opening/creating output file
open (TM, ">>lingualizerWfTM.txt") or die ("Cannot open file!");

# Creation of header for output file
print TM '%'. "$timestamp\t". '%User ID,' ."$userid,\t" . '%' . "$srclang\t" . '%'. "$creationtool" . '/00' . "\t"  . '%' . "$tgtlang\t" . '%---89668861' . "\t\t\t\n.\n";


# Looping through input file content line by line
foreach $lines(@text)
{
# Splitting line content and creating translation units

    @line = split (/","/, $lines);
        
        # Chopping leading " from source string by reversing, chopping and reversing
        $words = reverse($line[0]);
        $char = chop($words);
        $rev2str = reverse($words);
        $CleanSource = $rev2str;
        
         
        # Chopping \n from the end of target string
        $char = chop($line[1]);
        $TargetChop = $line[1];
        # Chopping " from the end of the target string
        $char = chop($TargetChop);
        $CleanTarget = $TargetChop;   

    # Creating the translation unit
   print TM "$timestamp\t" . "$userid\t" . "0\t" ."$srclang\t" . "$CleanSource\t" . "$tgtlang\t" . "$CleanTarget\t\t\t\n";
}
# Closing output file
close (TM);