Home >Backend Development >PHP Tutorial >The Forgotten Gem of PHP SPL_PHP Tutorial
The Forgotten Gem of PHP SPL_PHP Tutorial
- WBOYOriginal
- 2016-07-13 10:29:16962browse
The above article by Rafael Dohms amazed me so much that I couldn’t help but translate it and add some content.
SPL, PHP Standard Library (Standard PHP Library), is a built-in component and interface starting from PHP 5.0, and has gradually matured since PHP5.3. SPL is actually built into all PHP5 development environments and requires no setup.
It seems that many PHP developers basically don't use it, or even heard of it. The reason can be traced back to its snow-like documentation, which makes you ignore "its existence". SPL This gem is like Titanic's "Heart of the Ocean", which was sunk to the bottom of the sea. Now it should be picked up by us and worn in its rightful place, and this is the point of view expressed in this article.
So, what does SPL offer?
SPL extends the PHP engine with interfaces such as ArrayAccess, Countable, and SeekableIterator, which are used to manipulate objects as arrays. At the same time, you can also use other iterators such as RecursiveIterator and ArrayObjects to iterate data.
It also has several built-in objects such as Exceptions, SplObserver, Spltorage and splautoloadregister, splclasses, iteratorapply, etc. helper functions for overloading the corresponding Function.
These tools together are like a multi-functional Swiss Army knife. Making good use of them can qualitatively improve PHP code efficiency. So, how do we unleash its power?
Reload autoloader
If you are a "textbook programmer", then you must know how to use __autoload to replace the includes/requires operation to lazily load the corresponding class, right?
But over time, you will find that you have fallen into a dilemma. First, you have to ensure that your class files must be in the specified file path. For example, in the Zend framework, you must use "_" to separate class and method names (how do you solve this problem? this problem? ).
Another problem is that as the project becomes more and more complex, the logic within __autoload will also become correspondingly complex. In the end, you will even add exception judgment and write all the logic of loading classes into it.
Everyone knows that "eggs cannot be put in one basket", and SPL can be used to separate the loading logic of __autoload. Just write your own autoload function and overload it using the functions provided by SPL.
For example, in the Zend framework problem mentioned above, you can overload the corresponding method of Zend loader. If it does not find the corresponding class, then it will use the function you defined previously.
<?php
class MyLoader {
public static function doAutoload($class) {
// 本模块对应的 autoload 操作
}
}
spl_autoload_register( array('MyLoader', 'doAutoload') );
?>
As you can see, spl autoload register can also add multiple loading logic in the form of an array. At the same time, you can also use spl autoload unregister to remove loading logic that is no longer needed. This feature will always be used.
Iterator
Iteration is one of the common design patterns that is commonly used for unified traversal operations in a set of data. It is no exaggeration to say that SPL provides all the iterators you need for the corresponding data types.
A very good example is traversing directories. The conventional approach is to use scandir and then skip "." and "..", as well as other files that do not meet the conditions. For example, if you need to traverse a directory to extract image files, you need to determine whether they end in jpg or gif.
The following code is an example of using SPL's iterator to perform the above recursion to find image files in the specified directory:
<?php
class RecursiveFileFilterIterator extends FilterIterator {
// 满足条件的扩展名
protected $ext = array('jpg','gif');
/**
* 提供 $path 并生成对应的目录迭代器
*/
public function __construct($path) {
parent::__construct(new RecursiveIteratorIterator(new RecursiveDirectoryIterator($path)));
}
/**
* 检查文件扩展名是否满足条件
*/
public function accept() {
$item = $this->getInnerIterator();
if ($item->isFile() &&
in_array(pathinfo($item->getFilename(), PATHINFO_EXTENSION), $this->ext)) {
return TRUE;
}
}
}
// 实例化
foreach (new RecursiveFileFilterIterator('/path/to/something') as $item) {
echo $item . PHP_EOL;
}
?>You might say, doesn’t this take more code to do the same thing? So, look at the above code, don’t you have a highly reusable and testable code? :)
The following are other iterators provided by SPL:
- RecursiveIterator
- RecursiveIteratorIterator
- OuterIterator
- IteratorIterator
- FilterIterator
- RecursiveFilterIterator
- ParentIterator
- SeekableIterator
- LimitIterator
- GlobIterator
- CachingIterator
- RecursiveCachingIterator
- NoRewindIterator
- AppendIterator
- RecursiveIteratorIterator
- InfiniteIterator
- RegexIterator
- RecursiveRegexIterator
- EmptyIterator
- RecursiveTreeIterator
- ArrayIterator
Starting from PHP5.3, there will be more built-in iterators. I think you can try them. Maybe it will change your habit of writing traditional code.
SplFixedArray
SPL also has a series of built-in array manipulation tools. For example, you can use SplFixedArray to instantiate a fixed-length array. So why use it? Because it’s faster, and it even affects your salary :)
We know that PHP regular arrays contain keys of different types, such as numbers, strings, etc., and the length is variable. It is precisely because of these "advanced features" that PHP uses a hash to obtain the corresponding value through the key - in fact, this can cause performance problems in certain situations.
Since SplFixedArray uses fixed numeric keys, it does not use hash storage. Not exactly, you can even think of it as a C array. This is why SplFixedArray is faster than regular arrays (only in PHP5.3).
So how fast is it? The following set of data can give you a glimpse of it.
If you need a lot of array operations, then you can try it and believe that it is trustworthy.
Data structure
At the same time, SPL also provides the implementation of some basic types of data structures. Although we can use traditional variable types to describe data structures, such as using arrays to describe stacks (Strack) - and then use corresponding methods of pop and push (arraypop(), arraypush()), But you always have to be careful, because after all, they are not specifically designed to describe data structures - one wrong operation can destroy the stack.
SPL's SplStack object strictly describes data in the form of a stack and provides corresponding methods. At the same time, such code should also be able to understand that it is operating on a stack rather than an array, allowing your peers to better understand the corresponding code, and it will be faster.
Finally, maybe the above pale examples are not enough to "tempt you" to use SPL. Practice brings true knowledge, and you need to explore more and more powerful functions of SPL yourself. Only when it is carved slowly like a gem can it radiate its brilliance.