01-README

echo -e 'Nuuk\nQaanaaq\nAasiaat' | ~/langtech/kal/tools/shellscripts/kal-tokenise | cg-conv -c -F | grep -F Prop+Abs | lg-multi.pl

02-lg-multi.pl

#!/usr/bin/env perl
# -*- mode: cperl; indent-tabs-mode: nil; tab-width: 3; cperl-indent-level: 3; -*-
use strict;
use warnings;
use utf8;

BEGIN {
	$| = 1;
	binmode(STDIN, ':encoding(UTF-8)');
	binmode(STDOUT, ':encoding(UTF-8)');
}
use open qw( :encoding(UTF-8) :std );
use feature 'unicode_strings';

sub trim {
   $_[0] =~ s/\s+$//sg;
   $_[0] =~ s/^\s+//sg;
   return $_[0];
}

while (my $l = <STDIN>) {
   chomp($l);
   $l =~ s@^([^\t]+)\t(.+)$@$2@;
   my $wf = $1;

   print "# $wf => $l\n";
   # List all replacements, including itself if that should be part of the output
   foreach my $v (('Abs', 'Abl', 'Trm', 'Lok')) {
      my $new = $l;
      # Replace the tag to vary with the variant
      $new =~ s@\+Abs\+@+$v+@;

      # Generate surface forms
      my $gen = trim(`echo '$new' | hfst-optimized-lookup -u -p ~/langtech/kal/src/generator-gt-norm.hfstol`);

      # For each generated form, inject Katersat semantics
      my @gs = split(/\n/, $gen);
      foreach my $g (@gs) {
         my ($ana,$surf) = ($g =~ m@^([^\t]+)\t(.+)$@);
         $ana = "DUMMY\t$ana";
         $ana = trim(`echo '$ana' | cg-conv -f -C | ~/langtech/nutserut/dev/kal2eng/bin/long2sem.pl | cg-conv -c -F`);
         $ana =~ s@DUMMY@$surf@;
         $ana = join("\t", reverse(split(/\t/, $ana)));
         print "$ana\n";
      }
   }
   print "\n";
}