summaryrefslogtreecommitdiff
path: root/corpus/support/tokenizer.pl
diff options
context:
space:
mode:
Diffstat (limited to 'corpus/support/tokenizer.pl')
-rwxr-xr-xcorpus/support/tokenizer.pl2
1 files changed, 1 insertions, 1 deletions
diff --git a/corpus/support/tokenizer.pl b/corpus/support/tokenizer.pl
index f57bc87a..aa285be4 100755
--- a/corpus/support/tokenizer.pl
+++ b/corpus/support/tokenizer.pl
@@ -386,7 +386,7 @@ sub deep_proc_token {
}
##### step 1: separate by punct T2 on the boundary
- my $t2 = '\`|\!|\@|\+|\=|\[|\]|\<|\>|\||\(|\)|\{|\}|\?|\"|;';
+ my $t2 = '\`|\!|\@|\+|\=|\[|\]|\<|\>|\||\(|\)|\{|\}|\?|\"|;|●|○';
if($line =~ s/^(($t2)+)/$1 /){
return proc_line($line);
}