CODEWITHSUNDEEP
linuxbashshellunix
Abhishake Yadav
Abhishake Yadav

Reputation: 107

Loop through the files one by one in a directory as they appear and display result of command executed on them on the screen

I have a directory in which new files appear at a certain interval (10-15 seconds), all the files are in the same format (.ffid) and also numbered in a regular manner (1.ffid 2.ffid 3.ffid), I wand to execute some commands to extract information from n and (n-1) file and perform calculation on that information and display it on screen while the script runs.

This is the code that I have right now :

#!/bin/bash
#bash-hexdump
# Quick script to check delay of the shotpoints 

echo "please enter the complete line name as mentioned in the RAID2"

read  line

cd /argus/raid2/$line

dir=/argus/raid2/$line/

FILES="$dir"
while [ true ] 

do 


FFID=$(ls -lrt "$FILES" | grep -i ffid | tail -1)

echo "FFID Value is : "$FFID""


while [ $FFID = $(ls -lrt "$FILES" | grep -i ffid | tail -1) ]

do 

sleep 0.5

done 





ls -lrt "$dir" | awk '{print "     "$9}' | awk 'NR>1' | head -n -2 > /d/home/adira0151/Desktop/tmp/list_a


ls -lrt "$dir" | awk '{print "     "$9}' | awk 'NR>2' | head -n -2 > /d/home/adira0151/Desktop/tmp/list_b


paste /d/home/adira0151/Desktop/tmp/list_a  /d/home/adira0151/Desktop/tmp/list_b > /d/home/adira0151/Desktop/tmp/list_c




rm /d/home/adira0151/Desktop/tmp/list_a  /d/home/adira0151/Desktop/tmp/list_b



let ofst1=1840

let ofst2=1974

let ofst3=1798



 while    read  cffid nffid

 do 




        wd=$(hexdump -s $ofst1 -n 6 -e "64 \"%_p\" \"\\n\"" "$FILES""$cffid")  

        sp=$(hexdump -s $ofst3 -n 4 -e "64 \"%_p\" \"\\n\"" "$FILES""$cffid")  

        ct=$(hexdump -s $ofst2 -n 8 -e "64 \"%_p\" \"\\n\"" "$FILES""$cffid" | awk -F: '{print ($1 *3600) + ($2 * 60) + $3}')  

         nt=$(hexdump -s $ofst2 -n 8 -e "64 \"%_p\" \"\\n\"" "$FILES""$nffid" | awk -F: '{print ($1 *3600) + ($2 * 60) + $3}')  


          wbt=$(echo "$wd" | awk '{print (($1) *1.33)/1000}' | cut -c 1-3)

          spint=$(echo "$nt" "$ct" | awk '{print $1 - $2}') 

            tot=$(echo "$wbt" "$spint" | awk '{print $1 + $2}' |cut -c 1-2) 



          echo "  "  "    SP_NO  "  " "  "W_d"   "      "                                      "  Current_SP_Time "         "    "       "  Next_SP_Time  "      " W_B_T "          " SP_Int "           "   "            "  Total_time "  

          echo " "

          echo " " "  "  "  $sp "     "  $wd "         ""    "  "  "      "       " $ct "   "          "        "  $nt "         "   "    "   "  " $wbt   "          "  $spint "      "     "      "    "        "   $tot "     ""

          echo "                                                                                                                                                                           "



          echo " " 

          if [ $tot -lt 12 ] 

          then 

                paste /d/home/adira0151/Desktop/tmp/slow_down.txt 

                echo please slow down 



           fi

                done < /d/home/adira0151/Desktop/tmp/list_c 




   done

But it outputs repeated values from list_c , is there any way to just display the output line by line when new files appear in the directory.

Upvotes: 0

Views: 114

Answers (3)

Armali
Armali

Reputation: 19375

But it outputs repeated values …

You get repeated output for the same file because you have the setting

FFID=$(ls -lrt "$FILES" | grep -i ffid | tail -1)

of the old FFID value (to be compared with present ones to check if a new one has appeared) inside your outer while [ true ] loop, thus getting the same "old" value repeatedly. Rather move that line before the outer loop and set the variable FFID in the loop to the new FFID, using that for the comparison in the next loop cycle.

Upvotes: 0

Walter A
Walter A

Reputation: 19982

Decide how you want to see which files have been processed: You can write the filenames of the processed files to a logfile or move them to another dir (mkdir -p processed; mv ${file} processed). Process the new files and sleep a short time:

function process_file {
   your_logic_with $1
}

while [ x ]; do
   for file in *.ffid; do
      # grep "^${file}$" logfile | continue
      process_file "${file}"
      # mv "${file}" processed
      # Or
      # echo "${file}" >> processed
   done
   sleep 1
done

Upvotes: 1

SLePort
SLePort

Reputation: 15461

I suggest you use inotify to monitor newly created files.

In your case, to output (or do whatever you want) with newly created .ffid files : 

DIR="/tmp" # set DIR with your dir to monitor

inotifywait -m -e create $DIR | while read file
do
  if [[  ${file##*.} == "ffid" ]]; then
    if [[ -n "$prev_file" ]]; then
      printf "Command on previous file: %s\n" "$prev_file";
    fi
    if [[ -n "$file" ]]; then
      printf "Command on last file: %s\n" "$file";
    fi
    prev_file=$file  # stores the penultimate created file
  fi
done

Upvotes: 0

Related Questions