while (<BR>) {
# Only punctuation that is not part of a word is a token on its own.
# This only works on this text, because one flaw it has is decimal numbers
# and currency.
s/([,"$%\^\&*()+=\[\]!?])/ $1 /g;
s/[^A-Z.](\.)\s+/ $1 /g;
s/\s+/\n/g;
print $_;
}
Comments