Getting Specific Value from Dmidecode Output with Bash Script

Below is a section from the dmidecode output. There is a part of this section that I want to break down. I want to get the 4 bytes after pattern 01 85 30. And I want to change the byte order. Then I will compare with the hex strings I have.

....... 
Handle 0x0027, DMI type 219, 106 bytes
OEM-specific Type
        Header and Data:
                DB 6A 27 00 01 04 01 45 02 00 90 06 *01 85 30* 20
                00 00 00 00 40 00 00 03 1F 24 02 C9 02 60 44 02
                FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
                FF FF FF FF FF FF FF FF 03 00 00 00 80 00 00 00
                00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
                00 04 FF FF FF FF FF FF FF FF FF FF FF FF FF FF
                FF FF FF FF FF FF FF FF FF FF
.......

Expected output : 00000020

I prepared a bash script and it works, is there a better way?

#!/bin/bash

id_array=("00000020", "00000040")
unique_id=$(dmidecode -t 219 | grep "01 85 30" -A 1 | xargs | cut -c 46-56 | tr -d '[:space:]' | grep -o .. | tac | paste -sd ''  -)
if [[ " ${id_array[*]} " == *"$unique_id"* ]];
then
  echo "1"
else
  echo "0"
fi

Awk solution allowing for line breaks that might disrupt the search pattern and the byte values being sought.

I've used two short awk procedures pipped together. The first formats the data such that the search pattern and following bytes are always separated by a single space, without line breaks, or any tabs or runs of spaces.

The second procedure divides the data at the now clean pattern set and extracts the first four bytes of the second record (the part immediately following the pattern), and prints them in reverse order.

whole procedure (file contains the original data)

awk 'BEGIN{ORS=" "; OFS=" "} {gsub(FS," ",$0); $1=$1; print$0}' file | awk -v pattern="01 85 30" 'BEGIN{RS=pattern} NR==2{print $4" "$3" "$2" "$1}'

output

00 00 00 20

Explanation of parts

part 1

takes the data file file and re-formats to produce an output that contains each white-space separated 'word' of the input file but now separated by a single space.

awk 'BEGIN{ORS=" "; OFS=" "} {gsub(FS," ",$0); $1=$1; print$0}' file |

This is achieved by setting both the output record separator ORS and output field separator OFS to a single space in the BEGIN block. The main block then substitutes the default input field separatorFS (white space of any length) for a single space character in each record, using gsub(FS," ",$0);. The record (line) is then printed (print $0) after forcing awk to reevaluate it (using $1=$1) to prevent the line being passed as it was received.

The output of the first part is:

Handle 0x0027, DMI type 219, 106 bytes OEM-specific Type Header and Data: DB 6A 27 00 01 04 01 45 02 00 90 06 01 85 30 20 00 00 00 00 40 00 00 03 1F 24 02 C9 02 60 44 02 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF 03 00 00 00 80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 04 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF

and is piped directly to the second awk procedure

part 2

| awk -v pattern="01 85 30" 'BEGIN{RS=pattern} NR==2{print $4" "$3" "$2" "$1}'

uses the variable flag v to pass the search pattern into the awk procedure -v pattern="01 85 30" allowing the combined procedure to be called repeatedly from a shell command if many patterns are to be processed (by setting the pattern to a shell variable and using the variable name in place of the string "01 85 30").

A BEGIN block sets the record separator to be the passed pattern (RS=pattern) so that subsequent blocks will examine chunks of the file either side of the pattern as individual records.

Since the required data immediately follows the pattern we are only interested in the second record so the main awk block filters for record 2 (NR==2), printing the first four fields in reverse order, separated by single spaces, with print $4" "$3" "$2" "$1