What is more efficient, a complex query or treat the results?

I can do this:

    $sql = 'SELECT * FROM '
                    . 'data_companies '
                    . 'INNER JOIN data_companies_extra_infos ON' 
                    . 'data_companies.cod = data_companies_extra_infos.relationship';
   $rs = \MySQL\MySQLSelector::select_by_query_string($sql);

And have the result that I want.

Or, I can do this:

  $sql = 'SELECT * FROM '
                . 'data_companies ';
  $data = \MySQL\MySQLSelector::select_by_query_string($sql);


  $sql_ex = 'SELECT * FROM '
                . 'data_companies_extra_infos ';
  $rs_ex = \MySQL\MySQLSelector::select_by_query_string($sql_ex);
  $data = [];
  foreach($rs as $row){
      foreach($rs_ex as $row_ex){
           if($row['cod'] == $row_ex['relationship']){
               $data[] = array_merge($row,$row_ex);
           }
      }
  }

And also have the result that I want. Well, this is a simple example, I believe the first option is better, but in cases where there will be a complex query involving calculations, 4 or 5 tables and multiple filters for WHERE .

  

Which of the two is better in terms of performance?

In other words:

  

Is it better for me to do a giant and complex query, or to make multiple simple querys and treat the results with PHP?

PS: The select_by_query_string () function executes the query, treats and encodes the results, and then returns an array.

PS2: I personally find the second option easier.