AWK concatenate output lines while extracting specific strings

Question

Given the output of nmap against a subnet

Nmap scan report for 192.168.1.20
Host is up (0.010s latency).
MAC Address: EC:13:B2:E2:20:48 (Netonix)
Nmap scan report for 192.168.1.21
Host is up (0.010s latency).
MAC Address: EC:13:B2:E2:13:68 (Netonix)
Nmap scan report for 192.168.1.23
Host is up (0.010s latency).
MAC Address: EC:13:B2:E1:AE:A8 (Netonix)
Nmap scan report for 192.168.1.99
Host is up (0.00076s latency).
MAC Address: 90:6C:AC:48:86:DA (Fortinet)

The goal is to get an output for csv or any other delimited format:

IP,MAC,RTT
192.168.1.20,EC:13:B2:E2:20:48,0.010s
192.168.1.21,EC:13:B2:E2:13:68,0.010s
...

The above has been accomplished in a two step process using AWK

$time_date=$(date "+%Y-%m-%d_%T")    # Get the immediate date and time
$nmap_command=$(nmap -sP 192.168.1.0/24 -n --max-rtt-timeout 50ms)

# Concatenate all lines including and between the patterns "Nmap scan report for"
c_list=$((awk '/Starting|done:/ {next} /Nmap scan report for/{ if (x) print x; x=""; }{ x=(!x)?$0:x" "$0; }END{ print x; }')<<<$nmap_command)

# Extract the values of interest from each line, add a header line, then add delimiter for every value extracted
list=$((awk -v OFS=',' -v date=$time_date 'BEGIN{ print "Date,IP,MAC,RTT" };{ gsub(/[()]/,""); for (I=1;I<=NF;I++) if ($I == "for") ip=$(I+1); else if ($I == "up") lat=$(I+1); else if ($I == "Address:") mac=$(I+1); } { print date, ip, mac, lat }')<<<$c_list)

printf '%s\n' "$list" > ~/Desktop/list.csv  # Send output to file

exit 0

The question is can the above two step awk process be consolidated into a single awk command. Thank you for the all the input beforehand

EDIT attempts to consolidate the two step with a some variations of

x_list=$((awk '/Starting|done:/ {next} /Nmap scan report for/{ if (x) print x; x=""; }{ x=(!x)?$0:x" "$0; }END I=x BEGIN{ print "Date,IP,MAC,RTT" };{ gsub(/[()]/,""); for (I=1;I<=NF;I++) if ($I == "for") ip=$(I+1); else if ($I == "up") lat=$(I+1); else if ($I == "Address:") mac=$(I+1); } { print date, ip, mac, lat }')<<<$list_ips)

was not promising. AWK would always give a good reason why to cease processing at the gap between the END and BEGIN. The exercise is actually an attempt to create a (bash) function that would be able to render down any data that is in a repeating pattern, the nmap output is just an example.

Answer 1

The normal thing to do is to store the needed info from all the lines into awk variables and then print out the data on the final line in the block. So for the example you would have something like this (I have not coded the host down latency case as the example data doesn't show it).

nmap -sP 192.168.1.0/24 -n --max-rtt-timeout 50ms |
  awk '/Nmap scan report for / {ip=$5}
       /Host is up/ { latency=$4 ; sub(/(/,"",latency) }
       /MAC Address: / { mac=$3; print(ip "," mac "," latency) ; ip=mac=latency="unknown" }
       BEGIN { print "IP,MAC,RTT"}

clearing the variables in the final line of the block may or may not be required, depending on if the data is always complete or not. If there are a lot of variables then consider using an array rather than scalar variables like ip, mac, latency as these can all be cleared with a single delete.