Joseph Hallenbeck


October 13, 2014

DropFramwork

Filed under: Software Development — Tags: , , , — Joseph @ 6:00 pm

Flattr this!

This summer, I plunged into the depths of my back up drives and came up with some old projects that were growing some dust. Like most old projects, I find them, get excited. Decide to do a major revolutionary revamp, and ultimately just end up touching up some things and kicking them out the door. The DropFramework is one such thing. For a long time, I wanted to make my own micro-framework to compete with the likes of Slim or Silex. In the end though, I really feel that those two have the space of micro-frameworks very well covered. No one needs yet another PHP micro-framework with half-done ideas floating around. Still, I did want to show it off, so I polished it up a little bit and through it up on github. Below are my thoughts on “Yet Another PHP Microframework:”

Yet Another PHP Microframework

For all intensive purposes, you can consider this an abandoned project and I would not recommend anyone actually use this in production.

A few years ago when Code Igniter was still quite a hot thing and a lot of
servers were still running PHP 5.2, e.g. the “dark ages” before we got all the nice things that came along in PHP 5.3 it seemed to be quite the fashion for everyone to try their hand at writing their own framework.

This was my go at it.

You will find a lot of similarities with Code Igniter (since that is the framework I worked with at the time) and you might also find a lot of classes that look like they came straight out of PHP Objects, Patterns and Practice since that was my bible.

I wanted to do a few things in writing the DropFramework:

  1. I wanted to better understand the MVC pattern, the choices being made and how CI works.
  2. I wanted a framework that was small enough that I could read and understand every class in it.
  3. I wanted a framework with a very small footprint that worked by transforming HTTP requests into request objects / command objects. This allowed me to fire up multiple instances of the framework per HTTP request with the master application generating it’s own request objects that it would feed into it’s child instances and then construct a document out of the application responses from the children.
  4. I did not like at the time, and still do not like the major design patterns of a lot of ORM solutions which tend to treat the database as the authoritative model of the data. I rather turn this convention upside down: treat the database as just another form of user input. The model can then be constructed from any form of input — the database, an HTTP post, a file. The PHP object is then the authoritative source for how the data structure relates with other data. Any data coming into the model passes through a validation layer that translates it (or rejects it if it invalid).

Whether or not I succeeded at this items? I don’t think I would really know.

Version 0.4.0

The version of the framework that had been sitting on my hard disk for some time was 0.3.0. In deciding to release it I have done two major things:

  1. I created a simple example of the framework working. The code for this example is also up on github and a live version as well.
  2. I namespaced the entire framework and brought it into PSR-4 compliance allowing for installation via Composer and the use of the Composer autoloader. This defeats a lot of the purpose of the PHP 5.2 era frameworks which devoted a lot of their resources to locating and managing the loading of assets. This, of course, makes this no longer a PHP 5.2 compatible framework and probably even makes a lot of the framework look rather silly.

December 2, 2013

My Favorite PHP Helper Function — A Better Isset()

Filed under: Software Development — Tags: , , — Joseph @ 10:44 am

Flattr this!

My favorite helper function for CodeIgniter is a ridiculously simple function that has an amazing amount of utility. I stole, at least the idea of, this function from LemonStand and it has since made its way into nearly every CMS that I have worked on:

/**
*  Check if $var is set and if not return null or default
*  @param mixed $var The var to check if it is set.
*  @param mixed $default The value to return if var is not set.
*/
function h( &$var, $default = null) {

    return isset( $var ) ? $var: $default;

}

At first this doesn’t really seem to be doing much, after all at first glance it looks like it is nothing more than a wrapper for isset, but this improves heavily upon isset in two very important ways. First, let’s look at how this function works.

In the function definition we are taking a reference to a variable. Recall, a reference is pointing at the memory, not value, of a variable and so we can actually pass to our helper function a variable that has not yet been initialized. This saves us from receiving a notice that the variable does not exist. Our call to isset thus checks our memory to see if it is actually referencing a variable or nothing at all. If it is referencing an actual variable it returns that variable, otherwise it returns null (our default default) or whatever value has been assigned to $default.

The utility of this is best expressed in handful of examples. The biggest use of this is in a view. Let us look at a view constructed without the use of our helper function:

<h2><?= isset( $title ) ? $title : 'Blog Title' ?></h2>
<p><?= isset( $content ) ? $content : null ?></p>
etc.

In a sizable view the above can get quite long and quite cumbersome to maintain. Each call to isset is checking to see if the controller actually passed the value on to the view ( $title or $content ). If we did not do this we would get a notice from PHP. Sometimes this is resolved by programmers by using the error suppression symbal (@), however the notices will still end up in the logs of many frameworks that strictly check for errors. Contrast this with a view using our helper function:

<h1><?= h( $title, 'Blog Title' ) ?></h2>
<p><?= h( $content ) ?></p>
etc.

The above is a much, much more concise view that is easier to read and is still a strictly valid snippet of PHP that generates no warnings or notices. Once we start to use this helper function regularly all different kinds of uses come up for it, for example we can use it to see if a model returned a value:

# 1. Longer method without using the helper function
$page = $this->pages->getByURI( $url );
if( ! $page )
{
    $page = $this->pages->get404();
}  
$this->render( $page );

# 2. With helper function
$page = h( $this->pages->getByURI( $url ), $this->pages->get404() );
$this->render( $page );

The above snippets are fairly simple, but let’s walk through them. In both instances we need to pass some page object on to the render method. An error occurs if it does not get a valid page object so we must check after retrieving a page that it actually exists. In the first snippet we use four lines of code to first get a page by $url (the value of which is set somewhere else). Now if the pages model returns nothing then we enter a conditional statement that retrieves the 404 error page.

However, with the use of our helper function we can shorten the code in half and remove the conditional all together make it a much more readable snippet of code. The first line of the second snippet simply passes the return of the pages model and the get404 method into our helper function which returns the first if it returns something or the latter if it does not. The only downside is the additional load since the 404 page would also need to be loaded concurrent to the current page with each request, but in most cases this is going to be negligible.

Having looked at two different uses for our helper function, we can begin to see that we can get quite a bit out of some very very small functions. If you have your own favorite one-liner functions feel free to share in the comments below.


November 25, 2013

Improving Upon CodeIgniter’s Validation

Filed under: Software Development — Tags: , , , — Joseph @ 9:49 am

Flattr this!

In this article I plan on addressing CodeIgniter’s shortfalls as a framework for validating objects and introduce a method for improving the validator classes re-usability.

When To Validate?

The answer to this question is simple: whenever we are dealing with input. The (incorrect) assumption that CodeIgniter and many web-applications make is that user input comes in the form of GET and POST variables and a considerable amount of effort goes into validating inputs via these routes. However, GET and POST are not the only sources for user input. User input can come via external sources such as tying into a remote API, an RSS feed, or from the database itself. From each of these sources we could get an invalid state. In the case of the remote API or RSS feed this is easy to understand. The API could change, or the RSS feed could be malformed. In the case of the database the issue typically appears when data is stored into the database under one context but is then accessed with a different expectation.

Take for example a database with the following table:

CREATE TABLE IF NOT EXISTS `persons` (
  `id` int(10) NOT NULL AUTO_INCREMENT,  
  `name` varchar(256) NOT NULL,
  `birthdate` DATE NOT NULL,  
PRIMARY KEY (`id`) ) ;

Now say that we inserted a person with name “Bob” and birthdate “1975-01-01.” This passes the validator going into the database, but later on we pull this row from the database and use it to construct a plain PHP object with properties id, name, and birthdate which we pass onto the view and attempt to output the birthdate with the following line:

echo date('Y-m-d', $person->birthdate);

This is going to cause an error. Why? Because the date function is expecting the second parameter to be a UNIX timestamp, but birthdate is already a formatted date string. Now, we could solve this by changing the schema of the database or changing the schema of the person object, but it is important to note that even if we did fix the disparity between the two we would still not fix the issue that it is possible for the person object to exist in an invalid state.

So my answer is to when should validation occur is during object instantiation and setting. The properties of the object should not be able to be set to a value that the object cannot accept. This places validation clearly into the realm of the “M” in “MVC.”

Form Validation in CodeIgniter

CodeIgniter’s documentation offers a form validation class that makes the above mistake very clearly. It can only validate the POST super global and doesn’t really offer much of a solution towards validation of objects themselves. Furthermore, their example controller oddly mixes the issue of object validation, and thus business logic, inside the controller which tends to create in many CI application fairly bloated controllers:

public function index()
{
  $this->load->helper(array('form', 'url'));
  $this->load->library('form_validation');

  $this->form_validation->set_rules('username', 'Username', 'callback_username_check');
  $this->form_validation->set_rules('password', 'Password', 'required');
  $this->form_validation->set_rules('passconf', 'Password Confirmation', 'required');
  $this->form_validation->set_rules('email', 'Email', 'required|is_unique[users.email]');

  if ($this->form_validation->run() == FALSE) {
    $this->load->view('myform');
  } else {
    $this->load->view('formsuccess');
  }
}

I cannot offer a solution towards adapting the validation class to be fully object operating without a heavy rewrite of the class, but we can move this obtuse validation into a distinct model that encapsulates this behavior away from the controller.

Introducing the Abstract Validator Class

We can get the validation logic out of the controller by moving it into a Validator class. We begin with an abstract base class since each form will need their own validator classes:

abstract class Validator extends CI_Model 
{ 
  protected $rules = array(); 
  protected $fields = array();

  # Get keys of fields.
  public function getStructure()
  {
    return array_keys( $this->fields );
  }

  # Validate $_POST against the rules and fields.
  public function validate()
  {
    foreach( $this->rules as $key => $rule )
    {
      $this->form_validation->set_rules( $key, $this->fields[$key], $rule );
    }

    return $this->form_validation->run( $this );
  }   
}

We take advantage of the fact that the CI god to access the form_validation object inside the Validator instance to create the validate method which merely sets the validation rules and then runs them. The Validator has two properties $rules and $fields which we will use in sub-classes to provide the CI_Validator rules and fields strings. We can transform the above controller into the following subclass:

class LoginValidator extends Validator
{
  protected $rules = array(
                       'username' => 'callback_username_check',
                       'password' => 'required',
                       'passconf' => 'required',
                       'email' => 'required|is_unique[users.email]');
  protected $fields = array(
                        'username' => 'User Name',
                        'password' => 'Password',
                        'passconf' => 'Password Confirmation',
                        'email' => 'E-Mail');

  public function username_check( $name )
  {
    return $this->users->check_username( $name );
  }
}

Here we can see how the rules and fields are used as well as how we can extend the Validator class to add additional unique callback validations. This simplifies the controller significantly:

public function index() { 
  $this->load->library('form_validation');
  $this->load->model('loginvalidator');

  if ( $this->loginvalidator->validate() ) {
    $this->load->view('myform');
  } else {
    $this->load->view('formsuccess');
  }
}

The business logic is now gone and the controller is back to focusing on what it’s supposed to be doing — load resources and running paths.


September 23, 2013

CodeIgniter’s God Complex

Filed under: Software Development — Tags: , , , — Joseph @ 9:00 am

Flattr this!

I have worked with Code Igniter almost exclusively for the last nine months. In that time, I have found it to be a massive step ahead over working with some of the major CMS systems on the market (WordPress, I am looking at you). Nevertheless, there remains some major architectural and blind spots that exist in CodeIgniter as a framework. Some of these issues are resolvable (CodeIgniter’s presumption that you would only ever want to validate the POST superglobal), while others are inherent in it’s design. In this series I hope to look at some of these issues that I have found with CodeIgniter, showcase work-arounds where I can, or simply rant where no good solution exists. Today’s topic will be of the latter variety.

The God Object AntiPattern

Lets dip over to WikiPedia for the definition of a God Object:

In object-oriented programming, a god object is an object that knows too much or does too much… a program’s overall functionality is coded into a single “all-knowing” object, which maintains most of the information about the entire program and provides most of the methods for manipulating this data. Because this object holds so much data and requires so many methods, its role in the program becomes god-like (all-encompassing). Instead of program objects communicating amongst themselves directly, the other objects within the program rely on the god object for most of their information and interaction.

The God Object in CodeIgniter

CodeIgniter started as an early MVC framework that has maintained backwards compatibility with PHP5.2. It’s maintainers have insisted on maintaining this compatibility which has limited CI from taking advantage the advances that PHP5.3, 5.4, and 5.5 introduced to the language.

There remains nothing truly wrong with PHP5.2. While 5.3+ offers us many great advantages, a SOLID framework is still possible using the older version. CI’s architectural issues do not stem necessarily from it’s usage of the older version but rather the violation of SOLID principles in archetyping it’s interpretation of MVC.

In CI we have the CI super class (the idea of a super class alone should be a code smell) that is globally available via the get_instance() function. This returns an instance of CI_Controller, our main application controller handling the current request. This instance is our elusive beast. The God Object itself. We’ll call this object CI from here on out.

In any one request there can be only one instance of CI — it is essentially a singleton responsible for:

  1. Loading models
  2. Processing the request
  3. Returning the response

Overloaded Models

Here is where we get into the meat and potatoes.

The CI object begins its life by loading resources, that is it begins by loading various models and libraries and maintaining links to each of them like so:

public function __construct()
{
  $this->load->model('news');
  $this->load->model('events');
}

This code instantiates an instance of the news model and assigns a reference to news. It then instantiates an instance of events. In this manner every model that comes into existence during request process is held as a reference by the CI object and can be access latter on in the request, e.g.

public function index()
{
  $data = array();
  $data['articles'] = $this->news->get( 3 );
  $this->load->view( 'news', $data );
}

Once more, something very peculiar is done during this process. CI not only instantiates an instance of the given model but it also copies these references to every subsequently loaded model.

Thus every object that is loaded in this manner becomes aware of every object that had been loaded up-to that point regardless of whether that object really needed access to the behaviors of those objects. The model becomes unnecessarily bloated and the difficulty of debugging the behaviors of a given model increases. Unintended behaviors might be caused not by the model itself but by the combination of that particular model and the order or selection of previously loaded models.

Examine a Model’s State? No way.

Take for example the simple act of using var_dump to see the state of an object in memory. If we were to var_dump our instance of news we might as well call it a day as news contains a reference to everything that has been loaded into memory for our request. The server will proceed to dump the entirety of our application to the screen for us to wade through!

No Public Property is Safe

A larger issue is the assigning of the references themselves. Since the first act of initiating the model object is to copy CI’s massive registry of references to the model any properties or references set in the model’s constructor is at the mercy of the controller overwriting the model. Take for example, the events model. Let’s say the following was in the constructor:

public function __construct()
{
  $this->news = new News();
}

Following substantiation of the events object the Events object CI will immediately overwrite the news property with it’s own instance of the news property. Thus the events model would either need to make the news property private or protected which would generate an error when CI attempts to access it or we would always need to take care to keep our model properties from existing in the same namespace as CI.

I actually ran into a horrible bug where this very thing happened. I had a class named Validator that I loaded in with the controller. I also intended each of my models to load their own instances of the Validator class and to initialize their instances with own unique validation parameters. However, since the controller had already loaded an instance of Validator it immediately overwrote each of my model’s Validator‘s forcing them all to use the same instance of the class. The resolution to this problem was to have to name each instance of Validator something different, thus we had EventValidator, NewsValidator, etc.




Copyright 2011 - 2017 Joseph Hallenbeck Powered by WordPress