Script too slow ? Any particular reason?

Question

I have a script which does hexdump along with pipeline commands and finally I am redirecting it to a file. Now I am logging data from the serial port of my Raspberry Pi. I also call a python script(display time in milliseconds) from the shell script. I takes like 5 minutes for me to see the data on screen after I run the command.

            timeout 2s ./ExpHmax
            while :
            do
            sudo hexdump -v -e '1/1 "%02x\n"' /dev/ttyUSB0 | awk '$1=="ed" {
            getline byte1
            getline byte2
            number = strtonum("0x" substr(byte2,2,1))+2
            "./Time.py" |& getline timestamp;
            data = ":: "timestamp " ::ED." byte1"."byte2
            while(number-->0)
            {getline byte; data = data"."byte}
            print data"."
            }' |  tr '[a-z]' '[A-Z]'| grep -v '^.....................00' >      Thur.log
            done

Can anyone suggest me a way to do this script without doing the pipe. I guess that's causing the delay. Here is my Python script

   import time
   millis = int(round(time.time() * 1000))
   print millis

Answer 1

You could start by not instantiating Python on every matching byte (!):

#!/usr/bin/awk -f
function de() {
  getline < "/proc/uptime"
  close("/proc/uptime")
  return $0
}
BEGIN {
  ec = de()
}
$0 == "ed" {
  getline byte1
  getline byte2
  number = strtonum("0x" substr(byte2, 2, 1)) + 2
  data = ":: " de() - ec " ::ED." byte1 "." byte2
  while (number--) {
    getline byte
    data = data "." byte
  }
  print data "."
}