I'm transitioning to using OOP for all my projects, historically everything I've built has been pretty small and OOP didn't seem to be an efficient choice, but now with large projects it is. However more recently I've been coming across more and more "best practice" questions that I have and I can't find the answer for.
For example, imagine I have the following:
class numbers{
function __construct($number){
$this->number = (int)$number;
}
function add($add){
$this->added = $this->number + $add;
}
function multiply($multiply){
$this->multiplied = $this->number * $multiply;
}
function minus($minus){
$this->minused = $this->number - $minus;
}
function number(){
return $this->number();
}
}
Now let's say that I want to apply add
, then multiply
and then minus
. Each stage can possibly fail (I didn't include that in the example, but imagine it's there). This is where my problem lies, should I do:
$numbers = new numbers(8);
if($numbers->add(7)){
if($numbers->multiply(6)){
if($numbers->minus(7){
echo $numbers->number();
}else{
echo 'error minusing';
}
}else{
echo 'error multiplying number';
}
}else{
echo 'error adding number';
}
Or should I have that part in my constructor, like:
class numbers{
function __construct($number){
$this->add(6);
$this->multiply(9);
$this->minus(7);
if($this->error){
return false;
}else{
return true;
}
}
function add($add){
$this->added = $this->number + $add;
if(!this->added){
$this->error = "couldn't be added";
}
}
function multiply($multiply){
$this->multiplied = $this->number * $multiply;
if(!this->multiplied){
$this->error = "couldn't be multiplied";
}
}
function minus($minus){
$this->minused = $this->number - $minus;
if(!this->minused){
$this->error = "couldn't be minused";
}
}
function number(){
return $this->number();
}
function error(){
return $this->error();
}
}
and then simply do:
$numbers = new numbers(5);
if($numbers){
echo $numbers->number();
}else{
echo $numbers->error();
}
Sorry if the example is long winded (also ignore the errors, I wrote it here just to outline what I'm trying to do, this isn't code I'm using...) but I don't know how to phrase the question without an example. Basically, should I be doing checking for the errors inside of the class or should I be doing it outside when I call the class?
It almost always makes sense to enforce consistency on updates inside the class (i.e. any time a method can mutate the internal state of the class, check the input). Then use exceptions to make it the caller's problem if that input is not well-formed, or if that operation will violate some class invariant.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With