extracting a variable's value from text file using bash

Assuming the file is called vars.txt

sed -n 's/^VAR2=\(.*\)/\1/p' < vars.txt

You can use the value elsewhere like this using single back quotes

echo VAR2=`sed -n 's/^VAR2=\(.*\)/\1/p' < txt`

The simplest way might be to use source or simply . to read and execute the file. This would work with your example, because there are no spaces in the variable values. Otherwise you need to use grep + cut or awk, as stated in other answers.

. /path/to/your/file
echo $VAR2

[edit] As stated by dawg, this would make the other variables available in your script too, and possibly overwrite existing variables.

Given:

$ echo "$txt"
VAR1=100
VAR2=5
VAR3=0
VAR4=99

You can use awk:

$ echo "$txt" | awk -F= '/^VAR2/ { print $2 }'
5

Or grep and cut:

$ echo "$txt"  | egrep '^VAR2=\d+' | cut -d = -f 2
5

On Bash, you can insert the value of those assignments into the current shell using source and filter the lines you wish to use. In this case, only the line VAR2=5 will be used. You need to write that to a file and then source that file:

$ echo "$txt"  | grep '^VAR2' > tmp && source tmp && rm tmp
$ echo "$VAR2"
5

Or use process substitution to eliminate the tmp file. You can use either grep or awk here:

$ source <(awk '/^VAR2/{print}' <<< "$txt")
$ echo "$VAR2"
5

Assume this as your txt file, named test.txt

VAR2 = 5
VAR3 = 0
VAR4 = 99

you can cat test.txt | grep 'VAR2' | awk '{printf $3}'

and then your output will be: 5

Here, cat test.txt will display the content of test.txt in your terminal,grep 'VAR2' will list lines containing 'VAR2' and awk '{printf $3}' will print the value of the variable

For the files as described, you can just source the file as bash script which will run it's content and update you workspace environment with it. For example:

source file.txt
echo $VAR2