Shell Script which Works Similar to the Unix Command HEAD TAIL

Prerequisites: Bash Scripting, Shell Function Library

HEAD and TAIL command print a specified number of lines of a file from the beginning and from the end respectively. If we don't specify a number, the commands print the entire file. We will make a bash script display.sh which takes two to three arguments.

• -h or -t option to specify whether to print from head or tail
• NUM to specify the number of lines to be printed
  • If NUM is not specified, the script shall print the whole file.
  • If NUM exceeds the total number of lines in a file, the script shall print the whole file as usual.
• FILE address

Creating Shell Script

The name of the script will be display.sh. Make it an executable file.

$ touch display.sh
$ chmod +x display.sh

Step 1: Handling Arguments and Errors

$ ./display.sh #invalid syntax (no arguments provided)

$ ./display.sh FILE #invalid syntax (options not specified)

$ ./display.sh -option NUM FILE #correct syntax

$ ./display.sh -option FILE #correct syntax

Open the file and add the following script :

#!/bin/bash

# if the number of args is less than 2, 
# then show the correct syntax to the user
if [ $# -lt 2 ] ; then
  echo "Invalid Syntax!"
  echo "The valid syntax is ./$(basename $0) [-h|t] NUM [FILE]"
  exit 1
fi

# storing the option (-h or -t)
option=$1

# if second argument is the file to be displayed
# then by default we will print all the lines
# otherwise we will store the number of lines
# to be printed
if [[ -f $2 ]]; then
  NUM=$(wc -l < "$2") #wc -l returns number of lines in a file
  FILE="$2" #store the file address
else
  NUM=$2
  FILE="$3"
fi

• We will store the arguments in variable
• It works even if the file name contains spaces. Make sure to add double quotes. For example :   ./display.sh -h 10 "file name"
• wc -l command returns the total number of lines in a file.

Step 2: Create Main Function

case $option in
  "-h") print_head $NUM $FILE
        exit;;
  "-t") print_tail $NUM $FILE
        exit;;
     *) echo "Invalid Option"
        exit 1
        exit;;
esac
exit 0

• If the option is -h, the print_head function will be called.
• If the option is -t, the print_tail function will be called.
• If the option is invalid, the script will exit 1

Step 3: Create print_head function

function print_head ()
{
  #print first NUM lines
  NUM=$1 ; FILE=$2 ; COUNTER=0 ; 
  while IFS= read -r line #will keep reading the file till the end
  do
    echo "$line"
    (( COUNTER++ ))
    if [ $COUNTER = $NUM ] ; then #if we have printed NUM lines; break
      break
    fi
  done < "$FILE"                        
}

• read command trims white spaces. To avoid that, include IFS= in your script.
• -r option enables us to print escape sequences in the file as well.
• COUNTER variable keeps the count of lines that have been printed. 

Step 4: Create print_tail function

function print_tail ()
{
  #print last part of file
  NUM=$1 ; FILE="$2" ;
  total_lines=$(wc -l < "$FILE")

  #number of lines to be ignored
  if [ $NUM -ge $total_lines ] ; then
    IGNORE=0
  else
    IGNORE=$(( $total_lines - $NUM ))
  fi

  while IFS= read -r line
  do
    if [ $IGNORE -gt 0 ] ; then
      (( IGNORE-- ))
      continue
    fi
    echo "$line"
  done < "$FILE"
}

• We start reading the file line by line from the beginning.
• IGNORE variable will keep a track of how many lines at the beginning we have to ignore, and thus not print them.
• The rest of the lines will be printed once IGNORE becomes equal to 0.

Step 5: Putting together everything

#!/bin/bash

function print_head ()
{
  #print first NUM lines
  NUM=$1 ; FILE=$2 ; COUNTER=0 ; 
  while IFS= read -r line #will keep reading the file till the end
  do
    echo "$line"
    (( COUNTER++ ))
    if [ $COUNTER = $NUM ] ; then #if we have printed NUM lines; break
      break
    fi
  done < "$FILE"                        
}

function print_tail ()
{
  #print last part of file
  NUM=$1 ; FILE="$2" ;
  total_lines=$(wc -l < "$FILE")

  #number of lines to be ignored
  if [ $NUM -ge $total_lines ] ; then
    IGNORE=0
  else
    IGNORE=$(( $total_lines - $NUM ))
  fi

  while IFS= read -r line
  do
    if [ $IGNORE -gt 0 ] ; then
      (( IGNORE-- ))
      continue
    fi
    echo "$line"
  done < "$FILE"
}

# if the number of args is less than 2,
# then show the correct syntax to the user
if [ $# -lt 2 ] ; then
  echo "Invalid Syntax!"
  echo "The valid syntax is ./$(basename $0) [-h|t] NUM [FILE]"
  exit 1
fi

#storing the option (-h or -t)
option=$1

#if second argument is the file to be displayed
#then by default we will print all the lines
#otherwise we will store the number of lines to be printed
if [[ -f $2 ]]; then
  NUM=$(wc -l < "$2") #wc -l returns number of lines in a file
  FILE="$2" #store the file address
else
  NUM=$2
  FILE="$3"
fi

case $option in
  "-h") print_head $NUM $FILE
        exit;;
  "-t") print_tail $NUM $FILE
        exit;;
     *) echo "Invalid Option!"
        exit 1
        exit;;
esac
exit 0

Run the Script

$ ./display.sh -h FILE

$ ./display.sh -h 4 FILE

$ ./display.sh -t 6 file.txt
$ ./display.sh -t 1000 file.txt

Handles invalid cases as well