summaryrefslogtreecommitdiff
path: root/corpus/support/tokenizer.pl
diff options
context:
space:
mode:
authorPaul Baltescu <pauldb89@gmail.com>2013-02-21 14:13:55 +0000
committerPaul Baltescu <pauldb89@gmail.com>2013-02-21 14:13:55 +0000
commitbca26d953a774b8efca12f30407390b3f5eef9d0 (patch)
treefe922de5c89b1844f677d550dcc24e87edd67a55 /corpus/support/tokenizer.pl
parent54a1c0e2bde259e3acc9c0a8ec8da3c7704e80ca (diff)
parent95c364f2cb002241c4a62bedb1c5ef6f1e9a7f22 (diff)
Merge branch 'master' of https://github.com/pauldb89/cdec
Diffstat (limited to 'corpus/support/tokenizer.pl')
-rwxr-xr-xcorpus/support/tokenizer.pl3
1 files changed, 2 insertions, 1 deletions
diff --git a/corpus/support/tokenizer.pl b/corpus/support/tokenizer.pl
index e9c3a37d..0350a894 100755
--- a/corpus/support/tokenizer.pl
+++ b/corpus/support/tokenizer.pl
@@ -73,6 +73,7 @@ my $dict_file = "$workdir/token_list";
my $word_patt_file = "$workdir/token_patterns";
open(my $dict_fp, "$dict_file") or die;
+binmode($dict_fp, ":utf8");
# read in the list of words that should not be segmented,
## e.g.,"I.B.M.", co-operation.
@@ -89,6 +90,7 @@ while(<$dict_fp>){
}
open(my $patt_fp, "$word_patt_file") or die;
+binmode($patt_fp, ":utf8");
my @word_patts = ();
my $word_patt_num = 0;
while(<$patt_fp>){
@@ -147,7 +149,6 @@ while(<STDIN>){
print STDOUT " $new_line\n";
}
-print STDERR "\n";
########################################################################