From e419e52189ce8472bc5f5517796d00f0887affa7 Mon Sep 17 00:00:00 2001
From: Chris Dyer <cdyer@allegro.clab.cs.cmu.edu>
Date: Fri, 8 Mar 2013 22:44:49 -0500
Subject: few preproc fixes

---
 corpus/support/quote-norm.pl | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'corpus/support/quote-norm.pl')

diff --git a/corpus/support/quote-norm.pl b/corpus/support/quote-norm.pl
index d2980092..b104e73c 100755
--- a/corpus/support/quote-norm.pl
+++ b/corpus/support/quote-norm.pl
@@ -11,6 +11,8 @@ while(<STDIN>) {
   s/&\s*squot\s*;/'/gi;
   s/&\s*quot\s*;/"/gi;
   s/&\s*amp\s*;/&/gi;
+  s/&\s*nbsp\s*;/&/gi;
+  s/&\s*#\s*160\s*;/ /gi;
   s/ (\d\d): (\d\d)/ $1:$2/g;
   s/[\x{20a0}]\x{20ac}]/ EUR /g;
   s/[\x{00A3}]/ GBP /g;
-- 
cgit v1.2.3