#!/usr/bin/perl -w use strict; while() { chop; # avoid general madness s/[\000-\037]//g; s/\s+/ /g; s/^ //g; s/ $//g; # special characters in moses s/\&/\&/g; # escape escape s/\|/\&bar;/g; # factor separator s/\/\>/g; # xml s/\'/\'/g; # xml s/\"/\"/g; # xml s/\[/\[/g; # syntax non-terminal s/\]/\]/g; # syntax non-terminal # restore xml instructions s/\<(\S+) translation="([^\"]+)"> (.+?) <\/(\S+)>/\<$1 translation=\"$2\"> $3 <\/$4>/g; print $_."\n"; }