t.multiple.tokens.head

一个C语言写的快速贝叶斯垃圾邮件过滤工具

#!/bin/sh

. ${srcdir=.}/t.frame

INP=${TMPDIR}/test.inp
REF=${TMPDIR}/test.ref
OUT=${TMPDIR}/test.out
CORRECT=${SYSTEST}/outputs/multiple.wordlists.ref

cat <<EOF > $INP
From [e-mail]
To:  [e-mail]

This is a test
EOF

cat <<EOF > $REF
::: 1 :::
head:From
head:somebody
head:bogofilter.org
to:somebody-else
to:bogofilter.org
This
is
test
::: 2 :::
head:From
head:somebody
head:From*somebody
head:bogofilter.org
head:somebody*bogofilter.org
to:somebody-else
to:bogofilter.org
to:somebody-else*bogofilter.org
This
is
This*is
test
is*test
::: 3 :::
head:From
head:somebody
head:From*somebody
head:bogofilter.org
head:somebody*bogofilter.org
head:From*somebody*bogofilter.org
to:somebody-else
to:bogofilter.org
to:somebody-else*bogofilter.org
This
is
This*is
test
is*test
This*is*test
::: 4 :::
head:From
head:somebody
head:From*somebody
head:bogofilter.org
head:somebody*bogofilter.org
head:From*somebody*bogofilter.org
to:somebody-else
to:bogofilter.org
to:somebody-else*bogofilter.org
This
is
This*is
test
is*test
This*is*test
::: 5 :::
head:From
head:somebody
head:From*somebody
head:bogofilter.org
head:somebody*bogofilter.org
head:From*somebody*bogofilter.org
to:somebody-else
to:bogofilter.org
to:somebody-else*bogofilter.org
This
is
This*is
test
is*test
This*is*test
EOF

for COUNT in 1 2 3 4 5 ; do
    echo "::: $COUNT :::" >> $OUT
    $BOGOLEXER -C -p --min-token-len 2 --multi-token-count $COUNT < $INP >> $OUT
done

if  [ $verbose -eq 0 ]; then
    diff $REF $OUT
    cmp  $REF $OUT
else
    diff $DIFF_BRIEF $REF $OUT
fi