Do you know that objects in PHP5 are passed by reference, while arrays and other scalar variables are passed by value? Yes, you know it, but it’s not exactly true. Let’s see some example and let’s try to answer few questions.

// depending on the machine but both lines return
// expectedly equal values: 331208

// 331208
echo memory_get_usage();
echo memory_get_usage();

These two lines of code, expectedly return the same value (in my case 331208), which shows us that because nothing happened in between them the memory usage isn’t growing. Let’s now put some code in between them.

echo memory_get_usage(); // 331616
$a = 10;
echo memory_get_usage(); // 331696

Now because of the variable $a, we get a little more memory consuption! The same thing (with even more memory usage) happens if we have an array initialization.

echo memory_get_usage(); // 332128
$a = array(1, 2, 3, 'hello', 'world');
echo memory_get_usage(); // 332728

OK, now we see how much memory PHP is using for this very simple and small array. If we copy this array, we’d expect PHP to take twice as much memory, but that’s not the case!

echo memory_get_usage(); // 332336
$a = array(1, 2, 3, 'hello', 'world');
$b = $a;
echo memory_get_usage(); // 332984

Actually if we had $b = 10; this will consume more!!! memory than the code above.

echo memory_get_usage(); // 332336
$a = array(1, 2, 3, 'hello', 'world');
$b = 10;
echo memory_get_usage(); // 333016

This is simply because in the first case $b wasn’t a copy of $a, while in the second case we have a brand new variable on the ground, which, of course, requires memory.

Why Arrays aren’t Copied?

Actually now we see that copying by reference and by value is absolutely the same.

$a = array(1, 2, 3, 'hello', 'world');

// this line is exactly the same as ...
$b = $a;

// this line
$b = &$a;

That is because in both case the array is passed by reference. It is copied once we make changes to $b. Then in the first case $b becomes a copy of $a and it’s changed, while in the second case $b is exactly the same array as $a and every change to $b changes $a as well.

echo memory_get_usage();
$a = array(1, 2, 3, 'hello', 'world');
echo memory_get_usage();
$b = $a;
$b = array(1, 2, 3, 'goodby', 'world');
echo memory_get_usage();

Conclusion

If we take an example from Zend Framework, where often we work with arrays that are passed to the view as a “copy”:

$a = array(1, 2, 3, 'hello', 'world');
$this->view->a = $a; // this is NOT a copy

This will not consume more memory!!! The only way to make your application more memory inefficient is to change directly the $this->view->a array, so be careful!

If we talk about arrays and linked lists we know the pros and cons about both of them. No matter which programming language we use arrays benefit from direct access to its items, while linked lists are more memory efficient for particular tasks.

The items of a linked list keep a reference to their successor, so we can easily walk through the entire list. However we don’t have direct access to its elements. Thus we can’t go directly to its middle element! Even more – in particular implementations of a linked list we don’t know its length. But in some cases linked lists are far more effective than arrays. For instance reversing an array of non-numeric values require constant additional memory, but also requires n/2 exchanges. The same taks using linked lists is not only performed in linear time, but doesn’t require any additional memory. The only thing we need to do is to reverse the links – no movement of values and the items remain at the same place in the memory.

Merging of two arrays often require more space (proportional of the space of the two arrays) or many exchanges in case we try to do it in place. The same task on linked lists is far more effective with only changing pointers and without moving the values.

Arrays or Linked Lists are More Memory Efficient

Many developers consider linked lists as something used only in college, but actually they can be very useful in practice as well. However how practically useful they are? Let’s see the following PHP experiment.

Here we have one class called “Item”, which is designed to keep only one integer value as its key and to point to its successor. Practically this class is designed to be used by a singly linked list, but let say we put some of these objects into an array and the same amount of the “Item” objects into a linked lists so what are the results?

First let’s see the code!

class Item
{
    protected $_key = '';
    protected $_next = null;
    
    public function __construct($key)
    {
        $this->_key = $key;
    }
    
    public function setNext(&$next) { $this->_next = $next; }
    public function &getNext() { return $this->_next; }
    
    public function setKey($key) { $this->_key = $key; }
    public function getKey() { return $this->_key; }
    
    public function __toString()
    {
        return $this->_key . "\n";
    }
}

This is the “Item” class and here we have the Linked_List class. As you can see this is the very basic implementation of a linked list with only one “insert” method and the magic __toString() in order to print the entire list. The insert method pushes an item at the end of the list thus the insertion is O(1).

class Linked_List 
{
    protected $_head = null;
    protected $_tail = null;
    
    public function insert($item)
    {
        if ($this->_head == null) {
            $this->_head = $item;
            $this->_tail = $item;
            return;
        }
        
        $this->_tail->setNext($item);
        $this->_tail = $item;
    }
    
    public function __toString()
    {
        $current = $this->_head;
        $output = '';
        
        while ($current) {
            $output .= $current->getKey() . "\n";
            $current = $current->getNext();
        }
        
        return $output;
    }
}

Now let’s see the creation of an array with N objects of class “Item”.

$n = 10000;
$a = array();
for ($i = 0; $i < $n; $i++) {
    $a[$i] = new Item($i);
}

The same thing but using the Linked_List class follows on the lines below.

$n = 10000;
$a = new Linked_List();
for ($i = 0; $i < $n; $i++) {
    $a->insert(new Item($i));
}

And the Winner is ...

More memory efficient is ... the linked list! On the next chart we can see the results. It's clear that for 10K objects the array uses nearly 1MB more memory than the linked list!

So what do you think now? Will you use linked list in your code or not?

Final Words

Although the linked list seems to be more memory efficient we don't have direct acess to it's items. In the same time often we don't need direct access, we just need to walk through the array, which doesn't benefit from the direct access. In PHP this is usally done with some loop construction as "foreach". So why we have such results in the experiment above. First our linked list is really very basic. It doesn't have any functionality, which in fact shouldn't affect memory usage much more. The array in the other hand keeps indexes for each of its items so this results in additional space. This explains a bit the victory of the linked list in the memory efficiency test.

In the other hand PHP can't have the full benefit of using linked lists, trees and other data structures since it keeps them in memory only for the request. In this case C, C++, Java loads a data structure in memory till the software runs so unfortunately coding complex data structures in PHP doesn't look as a great option. Indeed here we have an entire "Item" class only to keep an integer. Instead we can use an array of integers!

PHP by Book

Here’s how PHP developer declare a string, which is something very common in any programming language.

$my_var = 'hello world';
// or
$my_var = "hello world";

PHP Tricks

What if you do the following:

echo hello;

That appears to be correct … Well, it’s not absolutely correct. You’ll be “noticed”.

// Notice: Use of undefined constant hello
echo hello;

However if you disable error reporting, the code will be completely fine.

error_reporting(0);

// no problem now
echo hello;

Variations

What follows from the thing above is that you can use strings without quotes:

// hello
echo hello;

// hello world (concatenated)
echo hello . ' world';

// helloworld
echo hello . world;

However you can’t have spaces and most of the “special” symbols.

// syntax error
echo hello world;

// syntax error
echo hello!;

Final Words

Although you can do this in PHP, that is completely wrong. The code becomes more difficult to read and understand. In the second place you can miss a $ sign in front of a variable declaration and thus the PHP interpreter will assume this is a string. So disable error reporting isn’t so great sometimes.

1. What will be the output of the following code?

echo date('Y-m-d', strtotime('-1 month ago'));

• Error
• One month ago from the current date and time
• One month in the future from the current date and time correct answer (ref)

2. Is it possible to extend an interface with an abstract class?

• Yes correct answer (ref)
• No

3. What is the output of the following code?

$arr = array(1 => 10, "1" => 20); 
echo count($arr);

• Error
• 0
• 1 correct answer (ref)
• 2

4. What is the output of the following code?

$str = "hello world"; echo $str{1};

• An error
• h
• e correct answer (ref)
• ello world

5. What is the output of the following code?

$str = "foo Bar"; 
echo ucwords($str);

• Foo bar
• Foo Bar correct answer (ref)
• FOO BAR
• foo bar

Passing Parameters by Reference

Clearly when we pass parameters to a function it’s not by reference. How to check this? Well, like this.

function f($param)
{
	$param++;
}

$a = 5;
f($a);

echo $a;

Now the value of $a equals 5. If it were passed by reference, it would be 6. With a little change of the code we can get it.

function f(&$param)
{
	$param++;
}

$a = 5;
f($a);

echo $a;

Now the variable’s value is 6.

So far, so good. Now what about copying objects?

Objects: A Copy or a Cloning?

We can check whether by assigning an object to a variable a reference or a copy of the object is passed.

class C
{
	public $myvar = 10;
}

$a = new C();
$b = $a;

$b->myvar = 20;

// 20, not 10
echo $a->myvar;

The last line outputs 20! This makes it clear. By assigning an object to a variable PHP pass its reference. To make a copy there’s another approach. We need to change $b = $a, to $b = clone $a;

class C
{
	public $myvar = 10;
}

$a = new C();
$b = clone $a;

$b->myvar = 20;

// 10
echo $a->myvar;

Arrays by Reference

What about arrays? What if I assign an array to a variable?

$a = array(20);

$b = $a;
$b[0] = 30;

var_dump($a);

What do you think is the value of $a[0]? Well, the answer is: 20! So $b is a copy of the array “a”. Instead you should assign explicitly its reference to make “b” point to “a”.

$a = array(20);

$b = &$a;
$b[0] = 30;

var_dump($a);

Now $a[0] equals 30!

I think this could be useful!

Interfaces

Object interfaces allow you to create code which specifies which methods a class must implement, without having to define how these methods are handled.
Interfaces are defined using the interface keyword, in the same way as a standard class, but without any of the methods having their contents defined.
All methods declared in an interface must be public, this is the nature of an interface.

To implement an interface, the implements operator is used. All methods in the interface must be implemented within a class; failure to do so will result in a fatal error. Classes may implement more than one interface if desired by separating each interface with a comma.

Abstract Classes

PHP 5 introduces abstract classes and methods. Classes defined as abstract may not be instantiated, and any class that contains at least one abstract method must also be abstract. Methods defined as abstract simply declare the method’s signature – they cannot define the implementation.

When inheriting from an abstract class, all methods marked abstract in the parent’s class declaration must be defined by the child; additionally, these methods must be defined with the same (or a less restricted) visibility. For example, if the abstract method is defined as protected, the function implementation must be defined as either protected or public, but not private. Furthermore the signatures of the methods must match, i.e. the type hints and the number of required arguments must be the same. This also applies to constructors as of PHP 5.4. Before 5.4 constructor signatures could differ.

Some Cases

Now let’s do a quick experiment. According to the definition of interfaces we can define an interface and than an abstract class can implement this interface.

interface A
{
	public function a();
}

abstract class B implements A
{
	public function a()
	{
		return 10;
	}
}

However the abstract class cannot be instantiated so we need to extend it.

interface A
{
	public function a();
}

abstract class B implements A
{
	public function a()
	{
		return 10;
	}
}

class C extends B
{
	...
}

In the first place is completely OK to define an abstract class that does not contain any abstract method. TUsually if a class has one or more abstract methods (Methods defined as abstract simply declare the method’s signature – they cannot define the implementation, according to the docs), it must be defined as abstract. But we can also define an abstract class which has no abstract methods.

abstract class B
{
	public function a()
	{
		return 5;
	}
}

class C extends B
{
	public function a()
	{
		return 10;
	}
}

$c = new C();
echo $c->a();

OK then what would happen if we had an interface that is implemented by an abstract class, which in turn is extended by a given class. First of all if a class implements an interface, even if it is an abstract class, it must implement all methods defined in the interface. So if the predefined method in the abstract class is marked as abstract, it would result into a fatal error.

interface A
{
	public function a();
}

abstract class B implements A
{
	abstract public function a();
}

class C extends B
{
	public function a()
	{
		return 5;
	}
}

$c = new C();
echo $c->a();

This is perfectly logical, because in fact each method of the interface must be implemented by his children.

“All methods in the interface must be implemented within a class; failure to do so will result in a fatal error.”

Fine, but we can have abstract classes with no abstract methods, let’s try this:

interface A
{
	public function a();
}

abstract class B implements A
{
	public function a()
	{
		return 10;
	}
}

class C extends B
{
	public function a()
	{
		return 5;
	}
}

$c = new C();
echo $c->a();

This code is working and the result is the returned value from C::a(). However class B is abstract so let’s go even further. Let’s define an abstract method, different from the method B::a().

interface A
{
	public function a();
}

abstract class B implements A
{
	public function a()
	{
		return 10;
	}
	
	abstract public function b();
}

class C extends B
{
	public function a()
	{
		return 5;
	}
	
	public function b()
	{
		return "hello world!";
	}
}

$c = new C();
echo $c->a();
echo $c->b();

Now there’s a fatal error. Why?

$str = 'hello world';
echo substr($str, 0, 1); // outputs "h"

But here’s something better and cleaner.

echo $str{0}; // outputs "h"

This code chunk return the first character of $str, but it can be used with the same success for any other character of the string. In my opinion this is more cleaner and its really syntactically self documented.

This approach can be useful when trying to check whether the first symbol for instance is “?” or “/”.

var a = 5;

function f() { return a; } // returns 5

Thus f will return the value of “a” which equals to 5. You can also change the value of the global variable in the function body.

var a = 5;
function f() { a = 10; return a; }
console.log(a); // equals to 10

Now after we call the function f the value of “a” will equal to 10. This is because we reference the global variable “a” into the function body without using the keyword “var”. This means that if you put the “var” keyword the variable “a” inside the function body is no longer the same variable as the variable defined outside the body. It becames “local” and it’s visible only inside the function.

Objects & JavaScript

Actually in JavaScript functions can be also objects. Any variable defined with the “var” keyword becomes private, because it’s only visible inside the function body, while by using the “this” keyword we can define global variables visible outside the function body. Let’s see na example.

var f = function() 
{
	var a = 10;
	this.b = 5;
}

var myfunc = new f();
myfunc.a; // is undefined because a is private
myfunc.b; // equals to 5 because b is public

Public and Private Methods in JavaScript

As you may know in JS you can define functions inside other functions. Actually this is how you can define classes in JavaScript.

var myClass = function() 
{
	var f = function() {
		return 10;
	}
}

But this in fact defines a local (private) function into myClass and we cannot access it from the outside world. Here’s a fully functional class with one public and one private method.

var myClass = function()
{
	// public method
	this.a = function() {
		return 10;
	}
	
	// private method
	var b = function() {
		return 5;
	}
}

var c = new myClass();
c.a(); // will return 10
c.b(); // is undefined, because is private

Accessing Private Methods from Public Methods

Easily we can access private methods from public ones.

var myClass = function()
{
	// public method
	this.a = function() {
		return b();
	}
	
	// private method
	var b = function() {
		return 5;
	}
}

However accessing public mtehods in private ones is more difficult. That’s because we cannot use “this” into the private methods, just because “this” refers to the private method
itself and not to the global method, which is “myClass” in our case.

var myClass = function()
{
	// public method
	this.a = function() {
		return 10;
	}
	
	// private method
	var b = function() {
		return this.a(); // this will result in an error
	}
}

Accessing Public Methods from Private Methods

To access public methods in private methods you need to define a variable that points to the global “this” object.

var myClass = function()
{
	var self = this; 
	
	// public method
	this.a = function() {
		return 10;
	}
	
	// private method
	var b = function() {
		return self.a(); // this will return 10
	}
}

This variable gives you the bridge between private methods and global “this” pointer.

Once I posted about JavaScript object coding style. Back than I made the analogy with PHP array coding style. In breve it’s useful to format the arrays in PHP simply like that:

$data = array(
	'key1' => 'value1',
	'key2' => 'value2',
	'key3' => 'value3',
	'key4' => 'value4',
	'key5' => 'value5',
);

Note that there is a trailing comma after the last key/value pair. This is not a syntax error and helps you add new elements to the array with no fear to forget the comma. This coding standard is quite well known in the PHP community, but in fact writing JavaScript objects can be “translated” to something very similar. The only problem is that the trailing comma in JavaScript will result to an error, especially in Internet Explorer, so it is important to remove it.

var obj = {
	key1 : 'value1',
	key2 : 'value2',
	key3 : 'value3',
	key4 : 'value4',
	key5 : 'value5'
};

The problem is that when you’ve to add one key/value pair, you’ve to add the comma after the last pair. This actually makes it useless.

Better Solution

There is another way, much better I think, that may help you more when adding new pairs to the object.

var obj = 
	{ key1 : 'value1'
	, key2 : 'value2'
	, key3 : 'value3'
	, key4 : 'value4'
	, key5 : 'value5'
	};

In this example you can simply copy/paste the last pair and change the key and value, or you can simply can continue writing the way the object is constructed.

Thus you don’t have the problem with the last comma and syntax errors.

First of all the use of some software is needed not because of itself, but because of the need of some specific functionality. In my case this was the need of real time news feed. Of course there is a way to make this without Node.js, as I’ll describe later in this post, but there are several disadvantages. However to begin from somewhere, let me explain what is Node.js.

Introducing Node.js

Perhaps the best way do describe what is Node.js is from its about page.

Node’s goal is to provide an easy way to build scalable network programs. In the “hello world” web server example above, many client connections can be handled concurrently. Node tells the operating system (through epoll, kqueue, /dev/poll, or select) that it should be notified when a new connection is made, and then it goes to sleep. If someone new connects, then it executes the callback. Each connection is only a small heap allocation.

…

In general Node is a program using the Google Chrome’s V8 JavaScript engine, which in turn is a program that can parse and execute code written in JavaScript. V8 is a very very interesting project itself. First of all from Google have developed this engine especially for one of his products – their browser Chrome. It pretends to be and by no means is one of the masterpieces of Google. It is fast and reliable engine, written in C++ and JavaScript, as Wikipedia’s page says. Actually this code is open source and can be embedded in whatever application written in C++. Thus you can have in your application a JavaScript engine.

Such kind of application/server is Node.js. With V8 embedded into, Node is intended to run on a machine and to serve some requests. Just like a normal web server, but with few differences. First of all Node is not only serving HTTP requests, but also TCP. Well I’m not so deep into this yet, but so far that works for me. What I need for now is its web serving functionality.

The way Node works, however is different from what we know from every other web server. Here it is a normal server in general:

Whenever a request comes to the server, it tries to return the response as fast as possible. This is one of the main goals of every web developer team – to make the application as fast as possible. However in Node the things work different:

Here the server doesn’t return immediately the response, but holds as long as possible until some “event” occurs. You can thing of Node as a classical chat server. Until one of the peers doesn’t write a message, the other doesn’t receive anything.

Some may be confused as how you can use both the typical web server, as Apache, with Node as another web server within the same page. The answer is – with AJAX. As you know thus you can have only one chunk of your page talking with the server – in this case the entire page is returned by Apache, but inside the page you can have a chat app using Node as a server, working perhaps on a different port on the same machine.

Typically this scenario can be simulated without Node, but as I said, with some disadvantages. Let’s say you have two browsers and one server – you can infinitely loop AJAX calls to the server and whenever there is new message on the server return it to the browser – this is quite inefficient. There are two approaches in this case.

1. You can make those AJAX calls too recently and you can have a call every N seconds. In that case if N is to short you’ll have too many calls and in case the other peer doesn’t make anything the server will be unusually loaded with no practical effect.
2. In the second case you can choose to make N too long – for instance 20 seconds. This is also bad, because during this period of time the peers may want to interact and chat.

In that case Node is a web server, but simply doesn’t return the response until a specific event doesn’t occur. How does this look like in a web browser. You may know how looking into the Firebug console for every AJAX call you see the request and how many seconds it has taken. If you make an AJAX call to a Node server app the request just doesn’t respond immediately.

Installing Node.js

The process of installation is described quite well, so I’m going to describe it in few steps.

1. Download the application from Git, note that V8 is embedded inside. The version while this post was written and of course was stable is v0.2.4
2. For those familiar with Unix/Linux systems, there are three simple steps following, however without GCC (Gnu Compiler Collection) you wont be able to proceed! I’m writing this especially for the Mac OS users, as they may know there is no built-in GCC there.
3. ./configure
4. make
5. make install
6. make test (optional)

This in general is enough. After all that you can write the node command in your terminal. However if there are some problems while compiling it you may refer to Git and the Node community for more help. Don’t forget that V8 can work only on x86, x64 and ARM architectures.

Summary

There are few things that can be summarized now:

1. First of all Node can be a web server, as Apache is, but with the option to respond on event firing.
2. You can request it either with a web page loaded in your browser or within a chunk of the page via AJAX.
3. To install Node you need GCC. It comes with V8 JavaScript engine inside, which can run only on x86, x64 and ARM.

So far so good! Now we have Node installed, practically useless, because there is no application running on it. In my next post I’ll describe how to run your first application.