HomePHP Page 5 - Caching Result Sets in PHP: Porting the Code to PHP 5
The caching system’s core class: updating the “Cache” class - PHP
In this part of the series, you will see in detail an updated version of each class that composes the caching system, for a correct implementation in PHP 5. Also, the “Cache” class will be modified to work with an array processor class that handles array operations. It should help refresh your memory of techniques related to object-oriented programming.
Since the “Cache” class implements the necessary logic to manipulate data either from a query result or from the cache file, I’ll explain in detail the changes applied to its structure.
As you hopefully remember, the class utilizes an array structure for handling result sets, which are serialized for caching purposes and unserialized for post processing. Considering that the majority of the class operations are carried out on simple arrays, it’d be highly desirable to develop an array iterator class that handles the most common tasks associated with arrays, and then create an instance of it (again, composition plays an important role) within the “Cache” class.
Bearing this concept in mind, let’s have a look at the updated definition for the “Cache” class. It is as follows:
class Cache{ private $mysql; // instance of MySQL object private $result; // instance of Result object private $arrproc; // instance of arrayProcessor object private $expiry; // cache expire time in seconds private $cacheFile; // cache file private $data; // result data array // constructor public function __construct (&$mysql,$expiry=7200,$cacheFile='default_cache.txt'){ $this->mysql=&$mysql; $this->expiry=$expiry; $this->cacheFile=$cacheFile; $this->data=array(); } // if cache is valid, perform query against database. Otherwise, get results from cache file public function query($query){ if(!$this->isValid()){ $this->result=&$this->mysql- >query($query); // read data from MySQL $this->data=$this->write(); } else { // read data from cache file $this->data=$this->read(); } // create a new instance of arrayProcessor object $this->arrproc=new arrayProcessor($this- >data); } // write cache file private function write(){ if(!$fp=fopen($this->cacheFile,'w')){ throw new Exception('Error opening cache file'); } // lock cache file if(!flock($fp,LOCK_EX)){ throw new Exception('Unable to lock cache file'); } // get result array from query result while($row=$this->result->fetchRow()){ $contents[]=$row; } // write serialized data to cache file if(!fwrite($fp,serialize($contents))){ throw new Exception('Error writing to cache file'); } // unlock cache file flock($fp,LOCK_UN); fclose($fp); unset($fp,$row); // return query result return $contents; } // read cache file private function read(){ if(!$contents=unserialize (file_get_contents($this->cacheFile))){ throw new Exception('Error reading cache file'); } // return cached data return $contents; } // determine if cached data is valid or not (time expiry triggered cache) private function isValid(){ if(file_exists($this->cacheFile) &&filemtime($this->cacheFile)>(time()-$this->expiry)){ return true; } return false; } // fetch row public function fetchRow(){ if(!$row=$this->arrproc- >getCurrentElement()){ return false; } $this->arrproc->getNextElement(); return $row; } // fetch all rows public function fetchAll(){ $this->arrproc->countElements(); return $this->data; } // count rows public function countRows(){ return $this->arrproc->countElements(); } // fetch range of rows public function fetchRange($offset,$length){ return $this->arrproc->getRange ($offset,$length); } }
As you can see, now the class includes three objects for doing its work: a reference of a MySQL object, $this->mysql (this object is aggregated), an instance of a “Result” object, that is $this->result, and finally an instance of the array processor $this->arrproc, for performing array operations.
If this sounds rather confusing, let’s explain the logic of some methods. For instance, the “fetchRow()” method now uses the “getCurrentElement()” method that belongs to the array processor for fetching the current element within the result set array. Similarly, this technique is used to count rows, through the “countRows()” method, or even fetch a limited result set using “fetchRange()”. See how we’re using the functionality provided by the array processor? Now, the inclusion of this object should be clear for the correct implementation for the “Cache” class.
By assigning clearly delineated responsibilities to different classes, we’re writing better and more portable code that can be reused in any number of projects.
Of course, there are many things to be reviewed yet, in order to get the big picture about the caching system. Certainly, we need to see in detail the code for the array processor class, as well as how the classes fit into the application. All right, don’t get anxious, because these topics will be fully explained in the next part of the series.
To wrap up
Through the course of this part, we’ve seen in detail the updated version of each class that composes the caching system, for a correct implementation in PHP 5. Also, the “Cache” class has been modified to work with an array processor class that handles array operations. Hopefully, the article has been didactic and useful for refreshing techniques related to object-oriented programming.
With all of the code exposed here, I guess you’ll spend a while adapting it to suit your needs. However, don’t miss the next part, where the complete caching system is put to work. Until then, stay in touch!