From 7cc92b65a3185aa242088d830e166e495674efc9 Mon Sep 17 00:00:00 2001
From: redpony <redpony@ec762483-ff6d-05da-a07a-a48fb63a330f>
Date: Tue, 22 Jun 2010 05:12:27 +0000
Subject: initial checkin

git-svn-id: https://ws10smt.googlecode.com/svn/trunk@2 ec762483-ff6d-05da-a07a-a48fb63a330f
---
 word-aligner/support/classify.pl | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)
 create mode 100755 word-aligner/support/classify.pl

(limited to 'word-aligner/support/classify.pl')

diff --git a/word-aligner/support/classify.pl b/word-aligner/support/classify.pl
new file mode 100755
index 00000000..893c7b22
--- /dev/null
+++ b/word-aligner/support/classify.pl
@@ -0,0 +1,27 @@
+#!/usr/bin/perl -w
+use strict;
+
+die "Usage: $0 classes.txt corpus.txt" unless scalar @ARGV == 2;
+
+my ($class, $text) = @ARGV;
+open C, "<$class" or die "Can't read $class: $!";
+open T, "<$text" or die "Can't read $text: $!";
+
+my %dict = ();
+my $cc = 0;
+while(<C>) {
+  chomp;
+  my ($word, $cat) = split /\s+/;
+  die "'$word' '$cat'" unless (defined $word && defined $cat);
+  $dict{$word} = $cat;
+  $cc++;
+}
+close C;
+print STDERR "Loaded classes for $cc words\n";
+
+while(<T>) {
+  chomp;
+  my @cats = map { $dict{$_} or die "Undefined class for $_"; } split /\s+/;
+  print "@cats\n";
+}
+
-- 
cgit v1.2.3