macOS script to collect and dedup MAC addresses from a network

autocollect.sh

#!/bin/bash
# SPDX-License-Identifier: MIT
# SPDX-FileCopyrightText: 2018 Ben Siraphob <bensiraphob@gmail.com>
command -v gshuf >/dev/null 2>&1 || { echo >&2 "gshuf is required but is not installed.  Run \"brew install coreutils\"."; control_c; }
command -v spoof-mac >/dev/null 2>&1 || { echo >&2 "spoof-mac is required but is not installed.  Run \"brew install spoof-mac\"."; control_c; }

# Colors
RED='\033[0;31m'
NC='\033[0m'

# Location of temporary data file during the run.
tmp="./.tmp"

# Location of where the MAC addresses (compressed) will be stored.
list="./list.txt"

# Name of the wireless network that you want the script to run.  This
# is to prevent accidentally running this script on other networks to
# ruin your sweet pure MAC address collection.
target_net="Wireless"

cleanup()
{
  rm -f $tmp
  return $?
}

control_c()
{
  (>&2 printf "\n${RED}Aborting...${NC}\n")
  cleanup
  exit 1
}

# Sometimes the script may take long, so trap C-c and clean up the temporary file.
trap control_c SIGINT

# Help argument
[[ $1 == "--help" ]] &&
echo "Options:
--help           Displays this help
--collect-only   Collects MAC addresses without spoofing
--spoof-only     Skip collection and spoof a random address"  && exit

# base64 + gzip reduces list size by a third. (e.g. 15 KB -> 5 KB)

# I used base64 encoding because A. it looks cooler but B. you can
# send it as plain text via any channel to help "bootstrap" other MAC
# address collecting computers.
(base64 --decode $list | gunzip) > $tmp

# How many MAC addresses do we have so far?
prev=$(cat $tmp | wc -l)

# If neither the --collect-only or --spoof-only options are supplied
# then the program collects AND spoofs.

# We're collecting.
if [[ $1 != "--spoof-only" ]]; then
    # Handy trick to determine current network name.
    currentNet="$(/System/Library/PrivateFrameworks/Apple80211.framework/Versions/Current/Resources/airport -I |
    awk '/ SSID/ {print substr($0, index($0, $2))}')"

    # Don't ruin our list.  Abort if we're connected to the wrong network.
    [[ $currentNet != $target_net ]] && (>&2 echo "You need to be connected to $target_net\!") && control_c
    echo "Collecting MAC addresses..."

    # Easy method to filter MAC addresses of the form XX:XX:XX:XX:XX
    # where XX is a hexadecimal number.  This is often the slowest
    # part of the script.
    arp -a | grep -o -E '([[:xdigit:]]{2}:){5}[[:xdigit:]]{2}' >> $tmp
    # Scrub bad addresses
    sed -i '' '/ff:ff:ff:ff:ff:ff/d' $tmp

    # We determine the number of online computers by checking how many
    # lines we've added.  This may include duplicates still.
    tmplen=$(cat $tmp | wc -l)
    online=$(($tmplen - $prev))

    # Sort and remove duplicates, then compress and base64 encode.
    (sort -u $tmp) | gzip | base64 > $list
    new=$(sort -u $tmp | wc -l)
    # Display status message.
    echo "From $(($online)) online, added $(($new - $prev)) addresses. Total Collected: $(($new))"
fi

# We're spoofing.
if [[ $1 != "--collect-only" ]]; then
    # This requires coreutils to be installed

    mac=$(gshuf -n 1 $tmp)
    echo "Attempting to spoof as $mac. A root password may be required."
    sudo spoof-mac set $mac en0 && echo "Spoofed as $mac."
fi

cleanup

exit