Extract variables values from command output Bash Shell - bash

Question

In bash shell I have a command that prints variable name and values such as:
Hello mate this is you variables:
MY_VAR1= this is the value of the first var
MY_VAR2= subvar1=27, subvar2=hello1
Have a good day!
The value of the variables in the output can contain characters as =,commas,;,:.., but I expect to find new line at the end of each variable value.
I need to create a short script which reads the values of MY_VAR1 and MY_VAR2.
So I need to end up with 2 variables as follows:
MY_VAR1 = this is the value of the first var
MY_VAR2 = subvar1=27, subvar2=hello1
I have a basic installation of CentOS 7, and I can't install additional stuff in that machine.
How can I achieve it?

Answer 1

I assume that the variable names are everything from the beginning of the lines, up to the = sign (excluded) with leading and trailing blanks removed (you cannot have blanks in your variable names).
If all you want is print the output you show, you can use something like:
while read line; do
if [[ $line =~ ^[[:blank:]]*([^[:blank:]]+)[[:blank:]]*=(.*)$ ]]; then
var="${BASH_REMATCH[1]}"
val="${BASH_REMATCH[2]}"
echo "$var = $val"
fi
done < <( my_command )
The =~ operator is a pattern matching with regular expressions. The regular expression ^[[:blank:]]*([^[:blank:]]+)[[:blank:]]*=(.*)$ models an output line of your command with variable assignment. It matches even if the variable name is surrounded with blanks. Two sub-expressions (enclosed in ()) are isolated: the variable name and the value. The BASH_REMATCH array contains the patterns that matched the sub-expressions in cells 1 and 2, respectively.
If you also want to assign the variables:
while read line; do
if [[ $line =~ ^[[:blank:]]*([^[:blank:]]+)[[:blank:]]*=(.*)$ ]]; then
var="${BASH_REMATCH[1]}"
val="${BASH_REMATCH[2]}"
echo "$var = $val"
declare $var="$val"
fi
done < <( my_command )