From 3a2fc36378337147a956e439db31baf91bfb95c8 Mon Sep 17 00:00:00 2001
From: Chris Dyer <cdyer@cs.cmu.edu>
Date: Fri, 3 Feb 2012 18:03:49 -0500
Subject: escaping tool for grammar extractor

---
 sa-extract/escape-testset.pl | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)
 create mode 100755 sa-extract/escape-testset.pl

(limited to 'sa-extract/escape-testset.pl')

diff --git a/sa-extract/escape-testset.pl b/sa-extract/escape-testset.pl
new file mode 100755
index 00000000..02fd7445
--- /dev/null
+++ b/sa-extract/escape-testset.pl
@@ -0,0 +1,35 @@
+#!/usr/bin/perl -w
+
+use utf8;
+use strict;
+
+binmode(STDIN,":utf8");
+binmode(STDOUT,":utf8");
+
+my @fh = ();
+if (scalar @ARGV == 0) {
+  push @fh, \*STDIN;
+} else {
+  for my $file (@ARGV) {
+    my $f;
+    open $f, "<$file" or die "Can't read $file: $!\n";
+    binmode $f, ":utf8";
+    push @fh, $f;
+  }
+}
+
+my $id = -1;
+for my $f (@fh) {
+  while(<$f>) {
+    chomp;
+    die "Empty line in test set" if /^\s*$/;
+    die "Please remove <seg> tags from input:\n$_" if /^\s*<seg/i;
+    $id++;
+    s/&/\&amp;/g;
+    s/</\&lt;/g;
+    s/>/\&gt;/g;
+    print "<seg id=\"$id\"> $_ </seg>\n";
+  }
+}
+
+
-- 
cgit v1.2.3