I know there's a lot of questions similar to mine, but I didn't found any about trapping signals sent when a subshell is taking place on the terminal. Let me explain:
#!/bin/sh
trap 'echo "exiting.."; exit 0;' INT
var1=$(echo "ab\nab" | fzf)
var2=$(echo "cd\ncd" | fzf)
fzf is a tool to display a selector of items of a list, so when the program reaches var1 a pretty selector covers my terminal screen prompting ab in one line and ab in the second, with the arrow keys I select the desired one and the result gets printed. Looks a lot like dmenu using this way.
The problem is that Ctrl-C doesn't get trapped by the script when fzf is prompting those lines. Instead, Ctrl-C kills fzf but the script keeps running.
So the question here is more about how to propagate this signal to the parent process.
There is no easy way to propagate the particular signal from child to parent.
However, exit code is 130 if the program is terminated by Ctrl-C in many cases. fzf follows it also.
Checking exit status and sending signal to own process is one of the workaround.
#!/bin/sh
trap 'echo "exiting.."; exit 0;' INT
var1=$(echo "ab\nab" | fzf)
[ $? -eq 130 ] && kill -INT $$
var2=$(echo "cd\ncd" | fzf)
[ $? -eq 130 ] && kill -INT $$
However, it is not common way.
If it needs to do something when terminated with any errors, using set -e and trapping EXIT or other signals is general way.
#!/bin/sh
set -e
trap 'echo "exiting.."; exit 0;' EXIT
var1=$(echo "ab\nab" | fzf)
var2=$(echo "cd\ncd" | fzf)