Replace/Update column value from text document in bash - linux

Question

I got a question on how can i update an specific column value from a row in a text document in bash.
So far I do this:
In this case I'm trying to update the 4th column in a line from an specific account number. take the current value, sum with a new value inserted by the user and then replace that current value with the result.
Open the script with two variable $1 holds the account I ll be looking for while $2 the file name.
The text document has information like this:
11101 : CAJA GENERAL : 111 : 0
11102 : CAJA CHICA : 111 : 0
112 : BANCOS : 11 : 0
11201 : CUENTAS CORRIENTES : 112 : 0
1120101 : Banco Agrícola S.A. : 11201 : 20
1120102 : Banco Hipotecario S.A. : 11201 : 0
11202 : CUENTAS DE AHORRO : 112 : 0
1120201 : Banco Agrícola S.A. : 11202 : 0
I use this code to find the correct row and assign the current value to the var "act" and aggregate it with another new value
read -p "Inserte monto" insert
act=$(grep -w "^$1" $2 | cut -d":" -f4)
vare=$(($act + $insert))
But then I need to place this new value to the exact column/row I took the original value.
How do i exactly do it? I'm quite exhausted from a long day of traveling and now I want to finish this and go to bed. Anyone could give me an idea or solution? I'd really appreciate any help right now.
EDIT:
Ok.. I went and try AWK.
Found this to "replace" text
awk -F':' -vOFS=' ' '{ $4 = "$vare"}1'
So far... once I enter that line the execution of the script stops.. or I dont know what but it doesn't continue anymore, nor shows any error.
Am I doing anything wrong?
EDIT 2:
Expected Input.
After a successful update I want the selected account to have the updated value in a existing document in my directory, the solutions so far allow me to see the updates within terminal, but is the original document the one I need to see the changes. Thanks Ed Morton for the tip

Answer 1

Maybe try awk:
$ cat file
11101 : CAJA GENERAL : 111 : 0
11102 : CAJA CHICA : 111 : 0
112 : BANCOS : 11 : 1
11201 : CUENTAS CORRIENTES : 112 : 0
1120101 : Banco Agrícola S.A. : 11201 : 20
1120102 : Banco Hipotecario S.A. : 11201 : 0
11202 : CUENTAS DE AHORRO : 112 : 0
1120201 : Banco Agrícola S.A. : 11202 : 0
$ insert=10
$ ident=112
$ awk -i inplace -F: "/^$ident /{gsub(/([0-9]+)$/, \$4+ $insert)};{print}" file
-or-
$ awk -i inplace -v ident="$ident" -v insert="$insert" '$1==ident{sub(/[0-9]+$/, $NF+insert)} 1' file
$ cat file
11101 : CAJA GENERAL : 111 : 0
11102 : CAJA CHICA : 111 : 0
112 : BANCOS : 11 : 11
11201 : CUENTAS CORRIENTES : 112 : 0
1120101 : Banco Agrícola S.A. : 11201 : 20
1120102 : Banco Hipotecario S.A. : 11201 : 0
11202 : CUENTAS DE AHORRO : 112 : 0
1120201 : Banco Agrícola S.A. : 11202 : 0

Answer 2

You can do it with sed:
id=$1
sed -i "/^$id *:/ s/[0-9]\+$/$vare/" $2