trap errors in bash scripts

Question

I have several questions I want to understand in regards to error trap.

1. why does the code below execute only 1 trap and not 2 and specifically only the code with "mydir2" and not "mydir1"?
2. why at all is it executing the trap?
3. why the command, if written as the marked code, will not generate a trap at all.
4. what is the best / proper way to write this code.

5. why and how do I get the correct line number for the error command and not the line of the function FuncA?

   #!/bin/bash
   
   set -o errtrace
   trap 'printerr' ERR
   function printerr(){
    local lc="$BASH_COMMAND" rc=$? ln=${BASH_LINENO[$i]}
    echo "$(date +%s) : Command [ $lc ] exited with code [ $rc ] in line [ $ln ]"
   }
   
   
   function FuncA(){
   
   [[ -d mydir1 ]] && echo mydir1OK
   
   [[ -d mydir2 ]] && echo mydir2OK
   
   #if [ -d /mtdir/ ] ;then
   #       echo OKMYDIR
   #fi
   
   }
   
   FuncA
   

Result:

1457453672 : Command [ [[ -d mydir2 ]] ] exited with code [ 1 ] in line [ 13 ]

Answer 1

It seems the ERR trap is executed because the function FuncA has a non-zero exit status. As documented in man bash

the exit status of a function is the exit status of the last command executed in the body.

Which, in this case, was the [[ -d mydir2 ]]. You can add a true as the last command in the function to prevent it.

The if construct is different, because (same manual):

The exit status is the exit status of the last command executed, or zero if no condition tested true.

I'd recommend using if rather than adding a true.