diff options
Diffstat (limited to 'corpus/tokenize-anything.sh')
| -rwxr-xr-x | corpus/tokenize-anything.sh | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/corpus/tokenize-anything.sh b/corpus/tokenize-anything.sh index 1a24193d..028992cf 100755 --- a/corpus/tokenize-anything.sh +++ b/corpus/tokenize-anything.sh @@ -9,5 +9,5 @@ $SUPPORT/utf8-normalize.sh | sed -e 's/ al - / al-/g' | $SUPPORT/fix-contract.pl | sed -e 's/^ //' | sed -e 's/ $//' | - perl -e 'while(<>){s/(\d+)(\.+)$/$1 ./;print;}' + perl -e 'while(<>){s/(\d+)(\.+)$/$1 ./; s/(\d+)(\.+) \|\|\|/$1 . |||/; print;}' |