nixCraft

#!/bin/bash
# bad words file
BFILE="/tmp/bfile.txt"

# ignore all words found in this file
IGGY="/tmp/bfile.iggy.txt"

# Admin Email ID
# Change this to your email id
MyEMAIL="root@localhost"

# Email subject
MySUBJECT="Bad words report dated $(date) @ $(hostname)"

# now start looking for all those bad words in directories 
# multiple dirs can be given at a time put single space between
# dirs 
# SDIRS="/home /www"
SDIRS="/home"

# ------------------------------- Do not change anything below
# this is new file do not write anything into it

# it will be created by script on fly
FBFILE="/tmp/bfile.final.txt"

# load iggy file words
iwords="$(cat $IGGY)"
# load bad file words 
bwords="$(cat $BFILE)"

# okay we need to iggy some words
for i in $iwords
do
   for j in $bwords
   do
      [ "$i" == "$j" ] && fwords="$fwords $i" || :
   done
done

# if final words file exists remot it
[ -f $FBFILE ] && rm -f $FBFILE || :
# copy our original bad words file to this
cp -f $BFILE $FBFILE
# create new words file ignoring all iggy words
for i in $fwords
do
    grep -v $i $FBFILE > /tmp/goodwords.txt
    mv -f /tmp/goodwords.txt $FBFILE
done


# Load all bad words 
WORDS="$(cat $FBFILE)"

# Out file
OUT="/tmp/out.$$"
FINAL="/tmp/out.$$.final"
>$OUT

# start to scan them
for d in $SDIRS # select first directory
do
    for b in $WORDS
    do
        # match whole bad word
        grep -E -w -r "$b" $d/* >> $OUT
    done
done
# create text file list in format
# file-name:bad-wordname
grep -E -w -v "^Binary file" $OUT > $FINAL

# send an email
mail -s "$MySUBJECT" $MyEMAIL < $FINAL