What is the difference when using ensure and else after rescue?

Question

1. What is the difference between using ensure:

   begin
     foo_process
   rescue
     bar_process
   ensure
     baz_process
   end
   

   and putting the process after the begin ... rescue ... end block?

   begin
     foo_process
   rescue
     bar_process
   end
   baz_process
   

2. What is the difference between using else:

   begin
     foo_process
   rescue
     bar_process
   else
     baz_process
   end
   

   and continuing that process before rescue (supposing that baz_process does not raise an error)?

   begin
     foo_process
     baz_process
   rescue
     bar_process
   end
   

Answer 1

1.

Using ensure, baz_process is alway executed even though foo_process throw exception, return from method.

For example,

def foo_process
  puts 'foo_process'
  return 111
end

def f1
  begin
    return foo_process
  rescue 
    puts 'bar_process'
  ensure 
    puts 'baz_process'
  end
end

def f2
  begin
    return foo_process
  rescue 
    puts 'bar_process'
  end
  puts 'baz_process'
end

>> f1
foo_process
baz_process
=> 111
>> f2
foo_process
=> 111

Answer 2

For question 1, using ensure, baz_process will always be run, even if an exception is not caught by the rescue block (i.e. in your example code, even if the exception is not a StandardError). It will also be run if bar_process raises an exception. Neither of these things are true for code placed after the begin...rescue...ensure block.

For question 2, the difference is that exceptions thrown by code in the else block are not rescued. This is more relevant when there is also an ensure block, as the code will run before it.