diff --git a/egs/gale_arabic/s5d/local/Eng2Ara.sh b/egs/gale_arabic/s5d/local/Eng2Ara.sh deleted file mode 100755 index f850ca5bc35..00000000000 --- a/egs/gale_arabic/s5d/local/Eng2Ara.sh +++ /dev/null @@ -1,109 +0,0 @@ -#!/usr/bin/env perl - -# Copyright 2014 QCRI (author: Ahmed Ali) -# Apache 2.0 - -use warnings; -use strict; -use Encode; -use utf8; - - - -if (@ARGV !=2 ) - {# - print "usage: $0 \n"; - exit (1); - } - -# <\check usage> -my $inFile = shift (@ARGV); -my $ouFile = shift(@ARGV); - - -open INFILE, "<$inFile" || die "unable to open the input file $inFile\n"; -binmode INFILE, ":encoding(utf8)"; - - -open OUTPUTFILE, ">$ouFile" or die "unable to open the output mlf file $ouFile\n"; -binmode OUTPUTFILE, ":encoding(utf8)"; - -while () { - my $BW = convertUTF8ToBuckwalter ($_); - print OUTPUTFILE "$BW"."\n"; -} -close INFILE; -close OUTPUTFILE; - - - - -# this function is copied from MADATools.pm: MADA Tools - sub convertUTF8ToBuckwalter { - - my ($line)= (@_); - $line =~ s/\x{0621}/\'/g; ## HAMZA - $line =~ s/\x{0622}/\|/g; ## ALEF WITH MADDA ABOVE - $line =~ s/\x{0623}/\>/g; ## ALEF WITH HAMZA ABOVE - $line =~ s/\x{0624}/\&/g; ## WAW WITH HAMZA ABOVE - $line =~ s/\x{0625}/\