Bash script: echo output of variable and write to file

Question

I am trying to write a script which will check to see if the contents of a file match a certain value and if so, overwrite it with contents of another file.

Example:

file1 =

A, B, C, D  
1, 2, 3, 4

file2 =

E, F, G, H  
5, 6, 7, 8

1. Checks to see if grep of file1 = A, B, C, D
2. If yes, cat file2 and write output back to file 1
3. If no, echo "Setting already configured" and exit

Here is what I have:

#!/bin/bash

var01=$(cat /home/user/scripts/test01.txt | grep A)
var02=$(cat /home/user/scripts/test02.txt)


if [[ $var01 = "A, B, C, D" ]];
then
  [[ echo $var02 > /home/user/scripts/test01.txt ]]
else
  echo "Setting already configured"
fi

I keep getting the following errors for line 9:

line 9: conditional binary operator expected
line 9: syntax error near `$var02'
line 9: `  [[ echo $var02 > /home/user/scripts/test01.txt ]]'

Any pointers would be greatly appreciated. Thanks.

Answer 1

So, it looks like that you want to copy test02.txt to test01.txt if test01.txt has one and only one line that contains A and that line is A, B, C, D.

Then maybe that should be something like:

#! /bin/sh -

if
  matching_lines=$(grep A test01.txt) &&
    [ "$matching_lines" = 'A, B, C, D' ]
then
  cp test02.txt test01.txt
fi

Beware that except in zsh, shells can't store NULs into their variables. If some lines containing A also contain NULs, the behaviour will vary between shells. bash removes the NULs in command substitution, so you may get a false positive if test01.txt contains A\0\0\0\0\0, B, C, \0\0D for instance.

Note that the above always scans the whole file from beginning to end. Since you know the conditions don't match as soon as a line other than A, B, C, D is seen containing A or as soon as any line containing A is found after the right one has been found, you could optimise it with something like:

if
  awk '
    /A/ {
      if (found) {found = 0; exit}
      if ($0 == "A, B, C, D")
        found = 1
      else
        exit
    }
    END {exit !found}' test01.txt
then
  cp test02.txt test01.txt
fi

With several awk implementations including gawk and modern versions of mawk, that also works around the NUL character problem described above.

Answer 2

#!/bin/bash
#   -q makes grep not give any output AND quit after finding a first match
# 
if grep -q 'A, B, C, D' /home/user/scripts/test01.txt
then
    cp /home/user/scripts/test02.txt /home/user/scripts/test01.txt
else
    echo "Setting already configured"
fi

Answer 3

#!/bin/bash

if [ "$(grep A /home/user/scripts/test01.txt)" = "A, B, C, D" ]; then
  cp /home/user/scripts/test02.txt /home/user/scripts/test01.txt
  echo "Setting configured."
  exit 0
fi

echo "Setting already configured."

exit 1

You were on the right track.
I just changed the else to follow through to exit 1 and "success" will return 0. (i.e exit 0)