Archive for the 'Design Patterns' Category

Page 2 of 14

Sandlight CMS III: PHP Abstract Factory

abfacNote: This is the third in a series for developing a CMS for Sandlight Productions. Drop by the Sandlight site to see the progress so far, and be sure to check your country’s flag count!

Now that the CMS has a filter for different devices, it now needs a pattern to take care of those devices and different types of content that they will display. Previous posts on this blog used the Bridge pattern and the Factory Method pattern. However, another pattern might be more useful for all of the different things that a Content Management System (CMS) might do. A review of the series of posts on this blog for how to select a design pattern, shows the different criteria to consider. The CMS has to create different elements of a Web page for different devices and that fact must be the focal point of the consideration. The section head for Creational Patterns in Learning PHP Design Patterns, lists the Abstract Factory pattern as a creational one, but that pattern was not discussed in the book nor on this blog. It would appear to be just what this CMS needs.

The Abstract Factory Design Pattern

The Abstract Factory pattern has features for families of factories and products instead of individual factories as does the Factory Method pattern. In comparing the relatively simple Factory Method pattern with the Abstract Factory, the Abstract Factory has multiple abstract product interfaces with multiple concrete factories for the families of the products.

Figure 1 shows the the Abstract Factory class diagram, and when you look at it, try to focus on the fact that the pattern has two types of interfaces: Factory and Product. So, if you understand the Factory Method pattern, you have a starting point for appreciating the Abstract Factory:

Figure 1: Abstract Factory class diagram

Figure 1: Abstract Factory class diagram

Unlike the Factory Method pattern, the Abstract Factory includes a Client class as an integral part of the pattern. Further, the Client holds an association between both the AbstractFactory (AbsFactory) and the two AbstractProduct classes. So while it shares some of the basic Factory Method characteristics, it is clearly a different pattern than the Factory Method.

Since the Abstract Factory may appear to be daunting, the color-coded the product instantiations (dashed lines) in the file diagram (appearing in the Play window) help show where each concrete factory method calls. Experiment with different combinations of factories (devices) and products (page parts) and look at the diagram so that you can see the path. Phone instantiations are in green, Tablet in red, and Desktop in blue. Experiment with the different single products first, and then click the bottom button to see what different “pages” each device factory displays.

Implementing the Abstract Factory in PHP

To see how this implementation of the Abstract Factory design pattern works, click the Play button. You will see both the interactive Abstract Pattern tester and the of the file diagram of this implementation of the Abstract Factory for the evolving Sandlight CMS. Click the Download button to see all of the files in the diagram. For this particular post, downloading all of the files is more important than usual because there are lots of them, and rather than having listings for all in this write-up, I’ve just selected representative ones.
PlayDownload

In addition to the Abstract Factory file diagram (viewed when you click the Play button), the following quick overview of the participants’ roles and how the CMS implements the Abstract Factory explain how this implementation works:

Client client: Only uses interfaces declared by IAbFactory (interface) and IProducts (abstract classes IHeaderProduct, IImageProduct and ITextProduct.) This means that the Client can only use the classes and methods implemented from those two interface types—factory or products. In other words, it should not directly implement a product (a page element) by directly using a product independent of the factory and product interfaces. Figure 2 illustrates this point:

Figure 2: Client works through Abstract Factory implementations

Figure 2: Client works through Abstract Factory implementations

In going over the other participants below, keep in mind that the Client can only implement concrete implementations of the IAbFactory to request products (page parts for different devices.)

IAbFactory interface: Establishes the methods for the concrete factories.

  • PhoneFactory implements operations to create phone products
  • TabletFactory implements operations to create tablet products
  • DesktopFactory implements operations to create desktop products

IHeaderProduct: abstract class Establishes the method for concrete header products and adds protected property for returning completed product.

  • PhoneHeader defines phone object to be created by the PhoneFactory and
    implements the IHeaderProduct interface
  • TabletHeader defines tablet object to be created by the Tabletactory and
    implements the IHeaderProduct interface
  • DesktopHeader defines desktop object to be created by the DesktopFactory and
    implements the IHeaderProduct interface

IImageProduct: abstract class Establishes the method for concrete image and/or video products and adds protected property for returning completed product.

  • PhoneImage defines phone object to be created by the PhoneFactory and
    implements the IImageProduct interface
  • TabletImage defines tablet object to be created by the Tabletactory and
    implements the IImageProduct interface
  • DesktopImage defines desktop object to be created by the DesktopFactory and
    implements the IImageProduct interface

ITextProduct: abstract class Establishes the method for concrete text products and adds protected property for returning completed product.

  • PhoneText defines phone object to be created by the PhoneFactory and
    implements the ITextProduct interface
  • TabletText defines tablet object to be created by the Tabletactory and
    implements the ITextProduct interface
  • DesktopText defines desktop object to be created by the DesktopFactory and
    implements the ITextProduct interface

Comparing the above outline with the Abstract Factory file diagram (seen when you click the Play button) shows that the Abstract Factory is bound to the idea of a factory implementing a product. The specific classes (products) requested are never directly referenced by the Client; rather it is through a factory. The CMS application requires factories for the different device categories; Phone, Tablet and Desktop. Each factory should be able to build the necessary parts (products) for each device. In this case (and for this example) the products are a Header, Graphic and Text. Each factory can build its own version of the products; so requesting a header, for example, is through a concrete factory, and depending on which concrete factory the clients requests, it builds the appropriate product.
Continue reading ‘Sandlight CMS III: PHP Abstract Factory’

Share

Sandlight CMS II : Mobile First!

mobileFirstI’m not a graphic designer, and so I depend on others for the graphic elements and arrangement of my Web pages. However, I strive to make a site that is clear, easy to understand and useful. My focus is on good user experience (UX) and information design—clear communication with the user. In order to have good UX, you need to know something about Responsive Web Design (RWD), and if you don’t, check out the RWD link. Further, if you are unfamiliar with the approaches to RWD, I’m sold on the Mobile First approach, but possibly for different reasons than designers. Let me explain.

In designing my own site, my focus is on content categories, ease of maintenance, which includes updates and changes, and device flexibility. So I have to keep all of those in mind. I want PHP to handle regular updates by using content from a MySql database (the Content Management of CMS), and I need it to work on different devices. By tackling mobile first, I have to create a diamond-tipped focus on exactly what I want the user to view because even with the new “Phablets,” I’m not dealing with a lot of screen real estate. Currently, my old working mobile phone has a CSS resolution of 320 x 480, and my Phablet is 414 x 736. That’s less that 100 units different. (By CSS resolution, I’m referring to what CSS reads as far as screen width is concerned. See this table.)

Choosing the Devices

In an another sniffer program using a Chain of Responsibility (CoR) design pattern and a PHP user agent ($_SERVER['HTTP_USER_AGENT']) posted on this blog, the sniffer detected the user agent and then found the handler responsible for that agent. Now that user agents have been replaced by CSS screen width (as determined by a JavaScript function) for determining the device, we can use the same CoR pattern making the necessary changes. However, instead of getting real pages, we can use stand-ins that only have the roughest page content. All of the content will be encapsulated into PHP classes using heredoc strings. Near-future posts cover the mechanics of working out the MySql to provide dynamic content for the pages, along with other details necessary for the CMS. For now, though, the dummy pages will only contain enough material to demonstrate that each is appropriate for the selected device. Use the buttons below to see the current state of the CMS and download the files for this portion:
PlayDownload

Note that all devices can now access the Flag Counter. Where is your country on the Flag Counter? (See the note about the Flag Counter at the end of this post.)

Back to the Chain of Responsibility Pattern (CoR)

The CoR pattern is handy because it’s easy to update and change. For example, suppose that having three device categories (e.g., phone, tablets and desktops) proves to be inadequate and you want to add two more; one for laptops and another for phablets. It’s a simple matter to add to the chain and add device classes to handle the new devices. Figure 1 shows the first design pattern to be used in the CMS:

Figure 1: Chain of Responsibility Implementation

Figure 1: Chain of Responsibility Implementation

In Part I of this series, you can see how the device width and height is determined using a JavaScript closure (object) to pass the information to HTML and on to PHP. Since we only need to find the width, the JavaScript code has been slightly altered and placed in a separate file (deviceCatcher.js) in case it needs to be reused.

?View Code JAVASCRIPT
//deviceCatcher.js
function getWide()
{
	var wide = screen.width;
	return function()
	{
		return wide;
	}
}
var w = getWide();
//Send data to PHP class, CoRClient.php	
var lambdaPass= function() {window.location.href = "CoRClient.php?hor=" + w();};

The HTML file simply calls the closure function which passes the values to PHP:

<html>
	<head>
		<title>Device Catcher</title>
		<script src="deviceCatcher.js" type="text/javascript"></script>
	</head>
	<body onload=lambdaPass()>
	</body>
</html>

The HTML file is a trigger to get the ball rolling with the client class (CoRClient).

Starting the Chain

The client pulls the viewing device’s width from the superglobal, and passes it to a PHP variable. Given the variability in the width of device screens, I made the decision to work with three sizes to get started: 1) phone, 2) tablet, and 3) desktop. So, depending on the width, the request would be handled by one of those three device groups. I used the following cutoff sizes:

  1. Phone: >= 480
  2. Tablet: >=481 and < 900
  3. Desktop: >= 900

I used this table as a guide, but cutoff points can be anything you want.

Getting the width from the superglobal is easy enough using a static variable:

self::$wide=$_GET['hor'];

The, using the cutoffs, the program needs to generate three strings, phone, tablet, and desktop to send to the Request class that stores the appropriate string. The most obvious way is to use conditional statements (if or switch) to generate the correct string for Request. For example an imperative algorithm such as the following would do the trick:

if(self::$wide < = 480)
{
	return "phone";
}
elseif (self::$wide >= 900)
{
	return "desktop";
}
else
{
	return "tablet";
}

However, a functional program would be more compact, and like the JavaScript closure used in Part I, it would be an “object.” Transformed into a functional closure, the operation would look like the following:

$beta = self::$wide >= 900 ? 'desktop' : 'tablet';
$lambda = function($x) use ($beta) {
	$alpha =  $x < = 480 ? 'phone' : $beta;
	return $alpha;};

Using ternary operations ?: , $alpha and $beta both have function-like qualities. for example, $beta could have been written as a function beta() as shown in Figure 2:

Figure 2: "Functional" variables

Figure 2: “Functional” variables

As you can see in Figure 2, $beta provides the same functionality as beta(), and $beta can be used as a reference in the $lambda function along with $alpha in a PHP closure. (For some reason, when $beta is assigned an anonymous function, I was unable to get it to be added as a closure in the $lambda anonymous function.)
Continue reading ‘Sandlight CMS II : Mobile First!’

Share

PHP Command Design Pattern: Part I

commandEncapsulate a Request

On occasion when developing an application, you may want to issue a request, but you may not know about the requested operation or its receiver. Imagine that you’re developing a game, and you’re working on a “Commander” (like Captain Kirk of Star Trek). The commander will issue commands, but during development or execution, you’re not sure who’s going to carry out the command or exactly how it’s going to be done. For instance, suppose that a Weapons Officer on the bridge simply pushes a button when the Captain issues a command to fire a photon torpedo. However, suppose that the bridge is damaged and the Weapons Officer wounded and cannot carry out the command. Any decent game designer would allow for someone else to launch the torpedoes; a crew member working in the torpedo bay, for example. Therefore, you would not want to tightly couple the action of firing a photon torpedo with the Weapons Officer. Something like;

weapons_officer->firePhotons()

would not be a good idea. Instead, you’d want the request to be handled by anyone who could get to the torpedoes triggering mechanism—no matter who it was or what triggering device was employed.

The Command design pattern encapsulates the request as an object, and allows you to add parameters to the clients with different requests. You can also support undoable operations—like firing off a photon torpedo! In this particular example, the participants and implementation is quite simple. I employed a helper class (Move) to add a little flair to the example, but otherwise, it’s a very simple implementation of the Command design pattern. Fire off the torpedo and download the files to get started:
torpedoDownload

The movement operations were taken from an earlier post on this blog of working with SVG files—once again illustrating the re-use functionality of OOP design. The helper class, Move is almost wholly a re-use of the earlier implementation. The sound was added to the class for this example.

The Wholly Involved Client

The requesting client is fully involved in the Command pattern. It is associated directly with both the Receiver and the ConcreteCommand classes. Figure 1 shows the pattern’s class diagram:

Figure 1: Command class diagram

Figure 1: Command class diagram

In this example, I employed a single Client, Receiver, Invoker and ConcreteCommand. Largely, I re-purposed an abstract example that Chandima Cumaranatunge had used in ActionScript 3.0 Design Patterns that we had co-authored in 2007—the main difference being that the example is written in PHP and I added an action that involved sound and animation. However, it is largely the same. In looking at the code in the Client, you can see that it creates instances of a ConcreteCommand, Receiver and Invoker.

<?php
error_reporting(E_ALL | E_STRICT);
ini_set("display_errors", 1);
// Autoload given function name.
function includeAll($className)
{
    include_once($className . '.php');
}
//Register
spl_autoload_register('includeAll');
 
/* Client */
class Client
{
    public static function request()
    {
        $rec = new Receiver();
        $concom = new ConcreteCommand($rec);
        $invoker = new Invoker();
        $invoker->setCommand($concom);
        $concom->execute();
    }
}
$worker=Client::request();
?>

As you can see, the Client, chooses the concrete command and the invoker. Figure 2 shows the file diagram for this Command implementation:

Figure 2: Command file diagram

Figure 2: Command file diagram

This pattern can have different commands and different invokers, but the key lies in the Command interface (ICommand) that includes an execute() method that invokes the command. It doesn’t concern itself with what other object does the invocation; it just provides the interface for some object to carry out the command.Using the bare minimum, the interface has a single abstract method:

<?php
interface ICommand
{
   function execute( );
}
?>

The design allows for a number of different ConcreteCommand classes, especially flexible because of the simplicity of the interface. However, in this example, only a single concrete command implements the interface.

<?php
class ConcreteCommand implements ICommand
{
    private $receiver;
    function __construct(Receiver $rec)
    {
        $this->receiver = $rec;
    }
    public function execute()
    {
        $this->receiver->action();
    }
}
?>

The receiver has been identified by the Client and passed to the parameterized ConcreteCommand by the Client. The Client acts very much like a conductor in this pattern. It pulls a the pieces together and determines how they will interact together. All of the work is to carry out the command realized in the Receiver class.

The Invoker Makes it Happen

Returning to the Star Trek example where the Captain commands the Weapons Officer to “fire a photon torpedo,” the Weapons Officer is the invoker. That is, by pressing the Fire button, she executes the command. However, as we noted, maybe the Weapons Officer is unable to carry out the order because she has been knocked out in a battle, and so someone else has to do it. Because the command and the invocation of the command are loosely coupled, any available invoker could carry out the command. First, take a look at the Invoker class:

<?php
class Invoker
{
    private $currentCommand;
    public function setCommand(ICommand $c)
    {
        $this->currentCommand = $c;
    }
    public function executeCommand()
    {
        $this->currentCommand->execute();
    }
}
?>

Note that both a method for setting the command and one for executing the command are part of the Invoker class. The Client sets the command, but how is it executed? Again, it is the Client, but the execute() method is through the ConcreteCommand instance; also called by the Client.

The Receiver Knows What to Do

In order for the command to be carried out, at least one object needs to do what the command requires. In this case, it’s the Receiver. Following the path so far:

  1. The Command tells what to do.
  2. The Invoker tells an object to carry out the command
  3. The Receiver carries out the requirements of the command.

Keep in mind that all of this is done with loosely coupled objects, and different participants can carry out the different roles.

<?php
class Receiver
{
   private $speed;
   private $photon;
 
   public function action()
   {
         $this->speed= 20;
         $this->photon = 16; 
         $launcher=new Move();
         echo $launcher->setVelocity($this->speed,$this->photon);
   }
}
?>

I suppose this is a bit elaborate for an abstract example, but it seemed a little more illustrative with something other than an echo statement that the Receiver object announced. Besides, it illustrates re-use of a class as a helper class for the Receiver to do a bit more. As you can see from Move helper class, a the property IDs from the original use have been retained—e.g., ‘ship’ and ‘torpedo’ used for the same purpose.

<?php
//Helper class
class Move
{
   private $velocity;
   private $capacity;
 
   public function setVelocity($speed,$ship)
   {
      $this->velocity=$speed;
      $this->capacity=$ship;
 
      $ship =<<<SHIP
      <!DOCTYPE HTML>
      <html>
         <head>
         <script type="text/ecmascript">
           // Single global variable
           var photon = 0; // Photon torpedo x position.
 
           function drawSVG()
           {
             //Interval Timer loop
             moveLoop = setInterval(fireTorpedo, $this->capacity);
           }
 
           function fireTorpedo()
           {
             // Change the photon torpedo's position
             photon += $this->velocity;
             photon = photon % 500;
             phpton.setAttribute("x", photon);
           }
         </script>
         </head>
           <body onload="drawSVG()">
             <!-- Create the SVG galaxy. -->
             <!--<?xml version="1.0" standalone="no"?> -->
             <!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd"> 
               <svg width="500" height="200" viewBox="0 0 500 200"
                  xmlns="http://www.w3.org/2000/svg" version="1.1">
                 <desc>Oopz</desc>
 
                 <!-- Show outline of 2D Galaxy using 'rect' element -->
                 <rect x="0" y="0" width="500" height="200"
                     fill="navy" stroke="blue" stroke-width="1"/>
                 <!-- Space craft Oopz -->
                 <rect x="100" y="100" width="30" height="20"
                     fill="yellow" stroke="#369" stroke-width=".4" id="phpton" />
               </svg>
               <audio autoplay>
                  <source src="photon.ogg" type="audio/ogg">
                  <source src="photon.mp3" type="audio/mpeg">
               </audio>
       </body>
       </html>
SHIP;
      return $ship;
   }
}

You can revise the actions of the Receiver to virtually anything you want. The important point with the Command class is that the original command (request) is encapsulated through the ICommand interface, and in the development process there can be any number of different requests fulfilled by different receivers and launched by different Invoker objects. Further with added commands, you can also add new invokers and ever receivers.

Beyond Abstract Command Structures

Even though this example has not been quite as abstract as originally planned, it is still pretty basic. What I’d like to do in subsequent examples is to build on this basic pattern where multiple commands, invokers and receivers work in concert through the Command design pattern. In the meantime, feel free to try out some of your own ideas and offer suggestions for further improvement and/or refinement.

Share

PHP Interpreter Design Pattern

interpreterInterpreting Postfix

Even though we may not realize it, one of the most common computer languages we use is Adobe’s PostScript. PostScript printers contain a language called PostScript, a postfix language based on a stack analogy. (Postfix also is called reverse Polish notation after the Polish mathematician Jan Łukasiewicz who developed the mathematical notation placing of the operator after the operands.)

Stack operations are compared to a cafeteria stack of trays. The last tray added to the stack is the first one to pop off. Imagine each tray with a value and each third tray with a mathematical operator. The bottom most tray is to the far left and the top tray is to the far right. With three trays, reading from left to right, we might find the first and second trays to be operands (numbers) and the third one to far right (top) an operator. So if the first operand is 5 and the second operand is 4 and the operator is add or (+), the three trays would be removed and replaced by a single tray with a value of 9—the sum of 5 and 4.

The following shows an example of this notation:

30 2 *
Result: 60

In order to correctly interpret a postfix expression as an infix one (the kind we’re used to using in PHP, like $total = 5 + 4;), this post uses an Interpreter design pattern that interprets a postfix expression so that PHP understands what to do with it as a math operation. The implementation of the Interpreter pattern is very simple and may lack some of the more subtle aspects of the pattern. However, it is one implementation that uses all of the pattern’s participants and should be easy to understand.

Using a simplified version of the Interpreter design pattern, you can practice math operations using a postfix calculator that interprets PostScript operations as PHP math operations. If the operation reminds you of certain of your calculator’s behavior, it’s probably because a lot of calculators use postfix to solve problems.

A further simplification is that this “calculator” only returns the results of a single two-operand calculation. A more common postfix expression may have several operators such as the following:

1 5 + 2 * 6 / result=2
1 and 5 add = 6
6 and 2 multiply = 12
12 and 6 divide = 2

This interpreter could be modified to take more complex expressions, but to get started, this one is simple by design. Click the play button to see how the program works and download the source code:
PlayDownload
Try some different entries using postfix notations with the caveat that all operators are spelled out—add instead of + and mod instead of %.

Interpreting Differences

The Interpreter design pattern as described by Gamma, et al can be used to express instances of a recurring problem as sentences. Interpreting the sentences then solves the problem. Right off the bat I was thinking, ¿Como está usted? translates to How’s it going?, but that’s not exactly what GoF had in mind. Any spoken language is a bit too big for this kind of interpreter. Instead, the pattern describes how to define a grammar for simple languages. Gamma and his associates use the example of regular expressions, so loved by Perl programmers.

So when I went looking at some Interpreter examples, besides regular expressions, I found converters from Roman numerals to regular numbers, a Boolean language, a musical note interpreter from do, rey, me, fa, so, la, ti to Hertz (Hz) values, calculations using postfix notations and some other fairly modest examples. Because of my experiences with postfix languages, I decided to do one that set up as a PostScript data entry that would resolve to an outcome (solution) to the results of a postfix statement. I decided on PostScript whose math operators are word-like and not symbols. (e.g., Instead of using / for division, it uses div.) Besides, most laser ink-jet printers are PostScript. This would be a little more than the usual minimalist example since the user can use it to practice and learn PostScript math entries.

Interpreter Design Pattern Formal Features

Figure 1 provides an overview of the Interpreter pattern. Note that the Client is part of the pattern (instead of implied or not at all). Also note that the Client holds references to both the Context class and AbstractExpression interface.

Figure 1: Interpreter Class Diagram

Figure 1: Interpreter Class Diagram


Of all of the design patterns I’ve seen, this one has the most loose ends. Any statement put into a string and then interpreted generally requires some kind of parsing. GoF note the need for parsing and point out it can be from a table driven source, a recursive descent (or some other hand-crafted parser) or in the Client. The Context class is used as a global entry point for the Interpreter. It’s probably heresy to do so, but I decided to put the parser in the Context. The Client sends the request to the Context and indirectly to the IAbstractExpression. The Context class still acts as a global entry point, but it also parses the data (statement in PostScript notation), but the Terminal Expression class acts more as a residual error-catcher than a terminal for multiple interpreted segments of a phrase. I don’t see it as adding tighter coupling. (If the parser were placed in the Client, I suppose it would be a purer version of the design pattern. The Client does, however, convert the string into an array using the preg_split() function; so, some partial parsing is done by the Client. The operators are treated as separate NonterminalExpression objects used to calculate the previous two elements on the stack (array). As such, they act as one-word interpreters.

The good news is that this implementation of the design pattern is wonderfully flexible and easy to update. If I wanted to change the language example from a postfix one to something like Scheme, the pattern would be able to handle it with ease. Of course, the purpose here is to examine the pattern with an illustration of its use, and while this implementation is basic, it’s a starting point.

Building a PostScript Learning Tool

The basic sequence for this little PostScript calculator is based on each the request passed from the HTML UI to the Client. It has the following sequence:

  1. User enters a string of two numbers and an operator in HTML UI.
  2. Client passes string to private variable and unsets superglobal
  3. Client converts string into a 3-element array and passes it to Context
  4. Context determines the correct expression class and returns class instance to Client
  5. Client removes the operator from the array and passes array through interpret() method of the object received from the Client.
  6. The expression class (IAbExpression implementation) carries out PHP (infix) math operation and returns result to Client
  7. Output displayed in iframe of HTML UI

Now having seen the sequence, Figure 2 shows the file diagram of the application:

Figure 2: File Diagram of Interpreter implementation

Figure 2: File Diagram of Interpreter implementation

I set the ValueTermExp.php file off to the side since it’s implemented more as an error message sender than its true role as a terminal expression. With a more sophisticated PostScript calculator (one that can handle a series of expressions), it can be upgraded to return the full calculation of the multiple NonTerminal classes.
Continue reading ‘PHP Interpreter Design Pattern’

Share

PHP Observer Design Pattern: The SplObserver

splObserverThe Standard PHP Library

For those of you in the Boston PHP 200 Days of Coding Advanced group and for those of you who develop with design patterns, you may have noticed that I have not taken advantage of PHP’s Standard Library (SPL). In part that’s because I don’t think about it, and the other is that often I have other features for the interface I’d like to add that are not in the SPL version. Further, there are only a couple, the Iterator that Larry talks about in his book, and the Observer pattern that I’m going to discuss in this post and on April 22 at Microsoft’s NERD Center next to the MIT campus.

In an earlier post on the Observer design pattern I noted that in a future post, I’d take a look at the built-in SplObserver and SplSubject interfaces. So as promised, here it is.

The class diagram is slightly different than the original, and to indicate the built-in interfaces, the backgrounds are filled with a light tint indicating, there’s nothing for you to add. If you compare Figure 1 with the class diagram in the other Observer pattern post, you can see certain fundamental differences:

Figure 1: Observer Class diagram with SPL Interfaces

Figure 1: Observer Class diagram with SPL Interfaces

With no abstract classes as part of the interface, everything passed to the concrete Subject and Observer is going to have to be implemented. With the Subject, that means implementing the notify() method and not having some protected properties, but otherwise, there’s not a lot of differences in the structure of the SplObserver and the Observer (from scratch.)

One Message: Lots of Ways to Configure the Message

Some developers treat the Observer design pattern like a magazine subscription—one magazine; lots of subscribers. I suppose that’s a legitimate use of the Observer, but it’s very expensive. Each concrete observer must be instantiated as a unique object. While the Subject (or SplSubject) supplies the data; the observers generally create something with that data; not just look at it. So, I decided to make a simple translation simulator. A club (like any of the many PHP groups in the world) has announcements, but rather than using the Observer for sending out notifications to subscribers, I created Observers who translated the event message from English to their own language. The concrete observers are named for their languages. But to get going, play the example and download the PHP code:
PlayDownload

When you test play the app, you’ll see that the all of the checkboxes on the left are checked and the text box is filled in with an event. Likewise, the data and day inputs are set to April and 22. You can just click the “Send Announcement” button to see the results. If you change the event to something other than “200 Days of Code” the only change you’ll see is in the English version. The other ones are only a simulation of a translation except for the month—they are actually translated. The idea would be to use a translation Web service to do the actual translations.

Beginning with the implementation of the SplSubject (ClubEvents), you can see the three abstract methods have been implemented using the signature from the SplMethod interface. (Their abstract form is shown at the top as comments.)

< ?php
class ClubEvents implements SplSubject
{
    /*Built-in abstract methods
     *abstract public void attach ( SplObserver $observer )
     *abstract public void detach ( SplObserver $observer )
     *abstract public void notify ( void )
    */
    private $member;
    private $observerSet = array();
    private $content= array();
 
    //add observer
    public function attach(SplObserver $observer)
    {
        array_push($this->observerSet, $observer);
    }
 
    //remove observer
    public function detach(SplObserver $observer)
    { 
      foreach($this->observerSet as $keyNow => $valNow)
      {
        if ($valNow == $observer)
        { 
          unset($this->observerSet[$keyNow]);
        }
      }   
    }
 
    //Set event name, month and day
    public function setState(array $content)
    {
        $this->content = $content;
        $this->notify();
    }
 
    public function getState()
    {
        return $this->content;
    }
 
    //Notify "subscribers"
    public function notify()
    {
        foreach ($this->observerSet as $value)
        {
            $value->update($this);
        }
    }
}
?>

If you’re familiar with the first example of the Observer on this blog, you’ll see similarities in the concrete Subject. About the only significant difference is the fact that the notify() method is wholly implemented in the concrete class (ClubEvents) instead of the Subject abstract class. The information (or data) generated by the SplSubject comes from the Client and triggers the notify() methods by calling setState() method. So instead of looking at the Observers next, we’ll look at the Client.

The Client’s Unlikely Role

Note: A primary principle of design patterns is,

Program to the interfaces; not the implementations.

In discussing the single SplSubject implementation (ClubEvents) and the many SplObserver implementations, I’ll be referring to them by the interface name except in certain specific cases. However, the references are to any implementation of either since we must assume (correctly) that any implementation contains a certain set of methods established in the abstract parent classes built into the Standard PHP Library (SPL).
Continue reading ‘PHP Observer Design Pattern: The SplObserver’

Share