In-depth understanding of PHP optimization and precautions

This article is a detailed analysis and introduction of high-efficiency writing methods in PHP. The summary is very detailed and comprehensive. It is very helpful for everyone to improve their PHP level. Friends who need it can refer to it!

1. Make it as static as possible:

If a method can be static, declare it as static, and the speed can be improved. Improved by 1/4, even when I tested, this improved by almost three times.

Of course, this test method needs to be executed at level 100,000 or above for the effect to be obvious.

In fact, the main difference in the efficiency of static methods and non-static methods is memory: static methods generate memory when the program starts, and instance methods generate memory while the program is running, so static methods can be called directly, and instance methods must be created first. Generate instances and call methods through instances. Static speed is very fast, but too much will occupy memory.

Any language operates on memory and disk. As for whether it is object-oriented, it is just a question of the software layer. The bottom layer is the same, but the implementation method is different. Static memory is continuous because it is generated at the beginning of the program, while the instance applies for discrete space, so of course it is not as fast as the static method.

Static methods always call the same memory. The disadvantage is that they cannot be destroyed automatically, but can be destroyed by instantiation.

2. Echo is more efficient than print, because echo does not return a value, and print returns an integer;

Test:

Echo
0.000929 - 0.001255 s (平均 0.001092 seconds)
Print
0.000980 - 0.001396 seconds (平均 0.001188 seconds)

The difference is about 8%. Generally speaking, echo is faster.

Note that when echoing large strings, performance will be seriously affected if no adjustments are made. Use mod_deflate to open apached for compression or open ob_start to put the content into the buffer first.

3. Set the maximum number of loops before the loop, not during the loop;

Every fool understands this.

4. Destroy variables to release memory, especially large arrays;

Arrays and objects particularly occupy memory in PHP. This is due to the underlying zend engine of PHP. Caused by,

Generally speaking, the memory utilization of PHP arrays is only 1/10. In other words, an array with 100M of memory in C language requires 1G in PHP.

Especially in systems where PHP is used as a backend server, the problem of excessive memory consumption often occurs.

5. Avoid using magic methods such as __get, __set, __autoload, etc.;

For functions starting with __, name them magic functions. Such functions are in First visit under specific conditions. In general, there are the following magic functions

##__construct(), __destruct(), __get(), __set(), __unset(), __call(), __callStatic(), __sleep(), __wakeup(), __toString(), __set_state(), __clone(), __autoload()<strong></strong>

In fact, if __autoload cannot efficiently compare the class name with the actual disk file (note, this refers to the actual disk file, not just the file name), the system will have to make a large number of judgments whether the file exists (it needs to be searched in the path included in each include path), and Determining whether a file exists requires disk I/O operations. As we all know, the efficiency of disk I/O operations is very low, so this is the reason why the efficiency of the autoload mechanism is reduced.

Therefore, when we design the system, we need to define a clear mechanism for mapping class names to actual disk files. The simpler and clearer this rule is, the more efficient the autoload mechanism will be.

Conclusion: The autoload mechanism is not naturally inefficient. Only abuse of autoload and poorly designed autoload functions will lead to a reduction in efficiency.

So try to avoid using __autoload magic The method is open to question.

6.requiere_once() consumes more resources;

This is because requirere_once needs to determine whether the file has been referenced), so it should be used as little as possible. Commonly used require/include methods to avoid.

7. Use absolute paths in includes and requires.

#If a relative path is included, PHP will traverse the include_path to find the file.

Using absolute paths will avoid such problems, so it will take less time to resolve the operating system path.

8. If you need to get the time when the script is executed, $_SERVER['REQUSET_TIME'] is better than time();
As you can imagine. One is ready-made and can be used directly, and the other requires the result obtained by the function.

9. If you can use PHP's internal string manipulation functions, try to use them instead of regular expressions; because their efficiency is higher than regular expressions;
No need to say, Regularity consumes the most performance.

Are there any useful functions that you missed? For example: strpbrk()strncasecmp()strpos()/strrpos()/stripos()/strripos() speeds up strtr. If all that needs to be converted is a single character,

use a string instead of an array to do strtr:

<?php
$addr = strtr($addr, "abcd", "efgh"); // good
$addr = strtr($addr, array(&#39;a&#39; => &#39;e&#39;, )); // bad
?>

Efficiency improvement: 10 times.

10.str_replace character replacement is faster than regular replacement preg_replace, but strtr is 1/4 faster than str_replace;

另外不要做无谓的替换即使没有替换，str_replace 也会为其参数分配内存。很慢！解决办法：
用 strpos 先查找(非常快)，看是否需要替换，如果需要，再替换效率：- 如果需要替换：效率几乎相等，差别在 0.1% 左右。
如果不需要替换：用 strpos 快 200%。

11.参数为字符串

如果一个函数既能接受数组又能接受简单字符做为参数，例如字符替换函数，并且参数列表不是太长，可以考虑额外写一段替换代码，使得每次传递参数都是一 个字符，而不是接受数组做为查找和替换参数。大事化小，1+1>2;

12.最好不用@，用@掩盖错误会降低脚本运行速度;

用@实际上后台有很多操作。用@比起不用@，效率差距：3 倍。特别不要在循环中使用@，在 5 次循环的测试中，即使是先用 error_reporting(0) 关掉错误，在循环完成后再打开，都比用@快。

13.$row['id']比$row[id]速度快7倍

建议养成数组键加引号的习惯;

14.在循环里别用函数

例如For($x=0; $x < count($array); $x), count()函数在外面先计算;原因你懂的。

16.在类的方法里建立局部变量速度最快，几乎和在方法里调用局部变量一样快;

17.建立一个全局变量要比局部变量要慢2倍;

由于局部变量是存在栈中的，当一个函数占用的栈空间不是很大的时候，这部分内存很有可能全部命中cache，这时候CPU访问的效率是很高的。
相反，如果一个函数里既使用了全局变量又使用了局部变量，那么当这两段地址相差较大时，cpu cache需要来回切换，那么效率会下降。
(我理解啊)

18.建立一个对象属性(类里面的变量)例如($this->prop++)比局部变量要慢3倍;

19.建立一个未声明的局部变量要比一个已经定义过的局部变量慢9-10倍

20.声明一个未被任何一个函数使用过的全局变量也会使性能降低(和声明相同数量的局部变量一样)。
PHP可能去检查这个全局变量是否存在;

21.方法的性能和在一个类里面定义的方法的数目没有关系

因为我添加10个或多个方法到测试的类里面(这些方法在测试方法的前后)后性能没什么差异;

22.在子类里方法的性能优于在基类中;

23.只调用一个参数并且函数体为空的函数运行花费的时间等于7-8次$localvar++运算，而一个类似的方法(类里的函数)运行等于大约15次$localvar++运算;

24 用单引号代替双引号来包含字符串，这样做会更快一些。
因为PHP会在双引号包围的字符串中搜寻变量，单引号则不会。

PHP 引擎允许使用单引号和双引号来封装字符串变量，但是这个是有很大的差别的！使用双引号的字符串告诉 PHP 引擎首先去读取字符串内容，查找其中的变 量，并改为变量对应的值。一般来说字符串是没有变量的，所以使用双引号会导致性能不佳。最好是使用字

符串连接而不是双引号字符串。

BAD:
$output = "This is a plain string";
GOOD:
$output = 'This is a plain string';
BAD:
$type = "mixed";
$output = "This is a $type string";
GOOD:
$type = 'mixed';
$output = 'This is a ' . $type .' string';

25.当echo字符串时用逗号代替点连接符更快些。

echo一种可以把多个字符串当作参数的“函数”（译注：PHP手册中说echo是语言结构，不是真正的函数，故把函数加上了双引号）。
例如echo $str1,$str2。

26.Apache解析一个PHP脚本的时间要比解析一个静态HTML页面慢2至10倍。

尽量多用静态HTML页面，少用脚本。

28.尽量使用缓存，建议用memcached。

高性能的分布式内存对象缓存系统，提高动态网络应用程序性能，减轻数据库的负担;
也对运算码 (OP code)的缓存很有用，使得脚本不必为每个请求做重新编译。

29.使用ip2long()和long2ip()函数把IP地址转成整型存放进数据库而非字符型。

这几乎能降低1/4的存储空间。同时可以很容易对地址进行排序和快速查找;

30.使用checkdnsrr()通过域名存在性来确认部分email地址的有效性

这个内置函数能保证每一个的域名对应一个IP地址;

31.使用mysql_*的改良函数mysqli_*;

32.试着喜欢使用三元运算符(?：);

33.是否需要PEAR

在你想在彻底重做你的项目前，看看PEAR有没有你需要的。PEAR是个巨大的资源库，很多php开发者都知道;

35.使用error_reporting(0)函数来预防潜在的敏感信息显示给用户。

理想的错误报告应该被完全禁用在php.ini文件里。可是如果你在用一个共享的虚拟主机，php.ini你不能修改，那么你最好添加error_reporting(0)函数，放在每个脚本文件的第一行(或用
require_once()来加载)这能有效的保护敏感的SQL查询和路径在出错时不被显示;

36.使用 gzcompress() 和gzuncompress()对容量大的字符串进行压缩(解压)在存进(取出)数据库时。

这种内置的函数使用gzip算法能压缩到90%;

37.通过参数变量地址得引用来使一个函数有多个返回值。

你可以在变量前加个“&”来表示按地址传递而非按值传递;

38. 完全理解魔术引用和SQL注入的危险。

Fully understand “magic quotes” and the dangers of SQL injection. I'm hoping that most developers reading this are already familiar with SQL injection. However, I list it here because it's absolutely critical to understand. If you've never heard the term before, spend the entire rest of the day googling and reading.

39.某些地方使用isset代替strlen

当操作字符串并需要检验其长度是否满足某种要求时，你想当然地会使用strlen()函数。此函数执行起来相当快，因为它不做任何计算，只返回在zval 结构（C的内置数据结构，用于存储PHP变量）中存储的已知字符串长度。但是，由于strlen()是函数，多多少少会有些慢，因为函数调用会经过诸多步骤，如字母小写化（译注：指函数名小写化，PHP不区分函数名大小写）、哈希查找，会跟随被调用的函数一起执行。在某些情况下，你可以使用isset() 技巧加速执行你的代码。
（举例如下）

if (strlen($foo) < 5) { echo “Foo is too short”$$ }

（与下面的技巧做比较）

if (!isset($foo{5})) { echo “Foo is too short”$$ }

调用isset()恰巧比strlen()快，因为与后者不同的是，isset()作为一种语言结构，意味着它的执行不需要函数查找和字母小写化。也就是说，实际上在检验字符串长度的顶层代码中你没有花太多开销。

40.使用++$i递增

When incrementing or decrementing the value of the variable $i++ happens to be a tad slower then ++$i. This is something PHP specific and does not apply to other languages, so don't go modifying your C or Java code thinking it'll suddenly become faster, it won't. ++$i happens to be faster in PHP because instead of 4 opcodes used for $i++ you only need 3. Post incrementation actually causes in the creation of a temporary var that is then incremented. While preincrementation increases the original value directly. This is one of the optimization that opcode optimized like Zend's PHP optimizer. It is a still a good idea to keep in mind since not all opcode optimizers perform this optimization and there are plenty of ISPs and servers running without an opcode optimizer.

当执行变量$i的递增或递减时，$i++会比++$i慢一些。这种差异是PHP特有的，并不适用于其他语言，所以请不要修改你的C或Java代码并指望它们能立即变快，没用的。++$i更快是因为它只需要3条指令(opcodes)，$i++则需要4条指令。后置递增实际上会产生一个临时变量，这个临时变量随后被递增。而前置递增直接在原值上递增。这是最优化处理的一种，正如Zend的PHP优化器所作的那样。牢记这个优化处理不失为一个好主意，因为并不是所有的指令优化器都会做同样的优化处理，并且存在大量没有装配指令优化器的互联网服务
提供商（ISPs）和服务器。

40. 不要随便就复制变量

有时候为了使 PHP 代码更加整洁，一些 PHP 新手（包括我）会把预定义好的变量复制到一个名字更简短的变量中，其实这样做的结果是增加了一倍的内存消耗，只会使程序更加慢。试想一下，在下面的例子中，如果用户恶意插入 512KB 字节的文字到文本输入框中，这样就会导致 1MB 的内存被消耗！

BAD:
$description = $_POST['description'];

echo $description;
GOOD:
echo $_POST['description'];

41 使用选择分支语句

switch case好于使用多个if，else if语句,并且代码更加容易阅读和维护。

42.在可以用file_get_contents替代file、fopen、feof、fgets

在可以用file_get_contents替代file、fopen、feof、fgets等系列方法的情况下，尽量用file_get_contents，因为他的效率高得多！但是要注意file_get_contents在打开一个URL文件时候的PHP版本问题；

43.尽量的少进行文件操作，虽然PHP的文件操作效率也不低的；

44.优化Select SQL语句，在可能的情况下尽量少的进行Insert、Update操作(在update上，我被恶批过)；

45.尽可能的使用PHP内部函数

46.循环内部不要声明变量，尤其是大变量：对象

(这好像不只是PHP里面要注意的问题吧？)；

47.多维数组尽量不要循环嵌套赋值；

48.foreach效率更高，尽量用foreach代替while和for循环；

49.“用i+=1代替i=i+1。符合c/c++的习惯，效率还高”；

50.对global变量，应该用完就unset()掉；

51 并不是事必面向对象(OOP)，面向对象往往开销很大，每个方法和对象调用都会消耗很多内存。

52 不要把方法细分得过多，仔细想想你真正打算重用的是哪些代码？

53 如果在代码中存在大量耗时的函数，你可以考虑用C扩展的方式实现它们。

54、打开apache的mod_deflate模块，可以提高网页的浏览速度。
（提到过echo 大变量的问题）

55、数据库连接当使用完毕时应关掉，不要用长连接。

56、split比exploade快

split()
0.001813 - 0.002271 seconds (avg 0.002042 seconds)
explode()
0.001678 - 0.003626 seconds (avg 0.002652 seconds)
Split can take regular expressions as delimiters, and runs faster too. ~23% on average.

The above is the detailed content of In-depth understanding of PHP optimization and precautions. For more information, please follow other related articles on the PHP Chinese website!