DevOps / Sys Admin Q & A #4 : Scripting

To run our script, enter:

1. $ ./script.sh
   The dot slash (./) denotes the current directory. This is necessary because the current directory is usually not in $PATH.
2. $ bash script.sh
3. $ source script.sh
   The file need not be executable but it must be a valid shell script. The file can be in current directory or in a directory in $PATH.
4. $ . script.sh
   Bash defined source as an alias to the dot command.

Note that sourcing will run the commands in the current shell process while bash executing will run the commands in a new shell process.

Output should look like this:

1 3 5 7 9

Shell 1

#!/bin/bash

for i in 1 3 5 7 9
do
   echo -n "$i "
done
echo

Shell 2

#!/bin/bash

for i in {1..9..2}
do
   echo -n "$i "
done
echo

Bash v4.0+ has inbuilt support for setting up a step value using {START..END..INCREMENT} syntax.

Shell 3

#!/bin/bash

for ((i=1; i<10; i+=2))
do
   echo -n "$i "
done
echo

Shell 4

#!/bin/bash

for i in {1..10}
do
  out=$(( $i % 2 ))
  if [ $out -eq 1 ]
  then
    echo -n "$i "
  fi
done

Shell 5

#!/bin/bash
c=1
while [ $c -le 10 ]
do
	echo -n "$c "
	(( c=$c+2 ))
done

Shell 6 - printing odd numbers 1 3 5 ...

#!/bin/bash
for i in {1..10}; do
   rem=$(( $i % 2 ))
   if [ $rem -ne 0 ]; then
	echo -n "$i "
   fi
done

# output 1 3 5 7 9

Ruby

Using "print" for no-new-line, the "puts" for a new line:

i = 1
loop do
   break if i > 10
   print "#{i} " unless i % 2 == 0
   i += 2
end
puts

We'll use 'vi' in the script:

#!/bin/bash

for file in sub/*
  do
    if [ -f $file ]
    then
      vi -c '$d' -c 'wq' "$file"
    fi
  done

The bash script will remove the last line from all files in a 'sub' directory.

When we need to replace a string across multiple files, we can use sed and the coding is similar to the one we used in the previous section:

#!/bin/bash

for file in ./*
  do
    if [ -f $file ]
    then
      sed -i 's/2018/2019/g' "$file"
    fi
  done

Note that we may tempted to do it with a single line with the "sed" command like this:

sed -i s/2018/2019/g *

But it won't work when we meet a directory. That's why we need to check if it is a file or not.

We want delete a line from a file and saved the modified file with the same name. The search string ss = ("justify"). We limit the file only with the *.php extension. Once we find the php files we loop through all the files. If a file has the search string, we then delete the line using "sed", redirect it to *.bak, and then move the *.bak to *.php.

#! /bin/bash

# search string
ss='("justify")'

# loop through *.php files
for f in $(find . -name '*.php' -print)
do
	# if a file has the string
	if [[ $(grep $ss $f) ]]  
	then
	    echo "$f"
	    
	    # delete the line and redirect the output 
	    sed "/$ss/d" $f > $f.bak

	    # move it back to original file
	    mv $f.bak $f
	fi
done

To be more familiarized with if statement, try the following script:

#!/bin/bash

TWO=2
for (( i=1;i<=3;i++ ))
do
  if [ ${i} -lt $TWO ]; then
    echo "${i} is less then $TWO"
  elif [ ${i} -eq $TWO ]; then
    echo "${i} is equal to $TWO"
  else
    echo "${i} is greater than  $TWO"
  fi
done  

Output:

1 is less then 2
2 is equal to 2
3 is greater than  2

To check if a string is not null nor a space:

#!/bin/bash
str1="Not Null"
str2=" "
str3=""
message="is not Null nor a space"

if [ ! -z "$str1" -a "$str1" != " " ]; then
  echo "str1 ${message}"
fi

if [ ! -z "$str2" -a "$str2" != " " ]; then
  echo "str2 ${message}"
fi

if [ ! -z "$str3" -a "$str3" != " " ]; then
  echo "str3 ${message}"
fi

Output:

str1 is not Null nor a space

We have a list of IP addresses from an input file.

1.2.3.4  5.6.7.8
11.12.13.14

The IP addresses are delimited by either space or '\n'. How we can construct a for loop to print out the data?

#!/bin/bash
for h in $(<hosts.txt)
do
  echo $h
done

Output:

1.2.3.4
5.6.7.8
11.12.13.14

The syntax looks like this:

command << delimiter document delimiter

The shell interprets the << operator as an instruction to read input until it finds a line containing the specified delimiter. All the input lines up to the line containing the delimiter are then fed into the standard input of the command.

Example 1:

$ wc -l << EOF
    This is a simple lookup program 
    for good (and bad) restaurants
    in Cape Town.
EOF
3
$

Example 2:

cat << EOF > /etc/apt/sources.list.d/kubernetes.list
deb http://apt.kubernetes.io/ kubernetes-xenial main
EOF

The command appends "deb http://apt.kubernetes.io/ kubernetes-xenial main" string to "/etc/apt/sources.list.d/kubernetes.list" file.

Example 3:

$ sed 's/_/-/g' << EOF
    2013_4_5
    2013_4_6
EOF
    2013-4-5
    2013-4-6

Example 4: writing a file from a shell script (a.sh):

#!/bin/bash
  
cat > myfile << EOF
000
111
222
EOF  

$ . a.sh

$ cat myfile
000
111
222

Here are other shell programming tutorials:

1. Linux shell programming - introduction
2. Linux shell programming - variables (readonly, unset)
3. Linux shell programming - special shell variables
4. Linux shell programming : arrays - three different ways of declaring arrays & looping with $*/$@
5. Linux shell programming : operations on array
6. Linux shell programming : variables & commands substitution
7. Linux shell programming : metacharacters & quotes
8. Linux shell programming : input/output redirection & here document
9. Linux shell programming : loop control - for, while, break, and break n
10. Linux shell programming : for-loop

DevOps

DevOps / Sys Admin Q & A