This is the code & its output I used to draw the inference below:

 class a {
    public $var1;
    public $var2;
    }

 $obj0 = new a;
 var_dump($obj0);

 class b {
    public $var1;
    public $var2;
    public $var3;
    }

 $obj1 = new b;
 var_dump($obj1);

 $obj2 = new stdClass;
 var_dump($obj2);

 $obj3 = new stdClass;
 var_dump($obj3); 

 $obj4 = new stdClass;
 var_dump($obj4);

 $obj5 = new stdClass;
 var_dump($obj5);

 var_dump(new stdClass); 

 $obj6 = new stdClass;
 var_dump($obj6);  

The output:

object(a)#1 (2) {
  ["var1"]=> NULL
  ["var2"]=> NULL
}
object(b)#2 (3) {
  ["var1"]=> NULL
  ["var2"]=> NULL
  ["var3"]=> NULL
}
object(stdClass)#3 (0) {

}
object(stdClass)#4 (0) {

}
object(stdClass)#5 (0) {

}
object(stdClass)#6 (0) {

}
object(stdClass)#7 (0) {

}
object(stdClass)#7 (0) {

}

The #<some-number> next to the line object(someClass) in var_dump of an object is actually #<count>. Where,

count is the number of objects / zval's for objects irrespective of which class it belongs to that has been created till now. Which keeps getting incrementing for every object created & gets decremented by 1 when a refcount of a zval reaches zero i.e. Garbage Collection.

Am I right?

Edit
Report