使用正则表达式突出搜索结果

例：

<style>
	.word-0{ background-color: yellow; }
	.word-1{ border:1px solid red; }
</style>

<?php

header('Content-type:text/html;charset=utf-8');

/* 标记Web页面 */
$body = '
<p>I like pickles and hrrring.</p>
<a href="pickle.php"><img width="200" src="pickle.png">A pickle pic</a>
I have herringbone-patterned toaster cozy.
<herring>Herring is not a real HTML element!</herring>
';

$words = array('pickle', 'herring');
$replacements = array();
foreach($words as $i => $word) {
	$replacements[] = "<span class='word-$i'>$word</span>";
}

// 将页面分解为多个块
// 由看上去类似HTML元素的部分分隔
$parts = preg_split("{(<(?:"[^"]*"|'[^']*'|[^'">])*>)}", $body, -1, PREG_SPLIT_DELIM_CAPTURE);
//var_dump($parts);
/*
array (size=15)
  0 => string '
' (length=2)
  1 => string '<p>' (length=3)
  2 => string 'I like pickles and hrrring.' (length=27)
  3 => string '</p>' (length=4)
  4 => string '
' (length=2)
  5 => string '<a href="pickle.php">' (length=21)
  6 => string '' (length=0)
  7 => string '<img width="200" src="pickle.png">' (length=34)
  8 => string 'A pickle pic' (length=12)
  9 => string '</a>' (length=4)
  10 => string '
I have herringbone-patterned toaster cozy.
' (length=46)
  11 => string '<herring>' (length=9)
  12 => string 'Herring is not a real HTML element!' (length=35)
  13 => string '</herring>' (length=10)
  14 => string '
' (length=2)
*/

foreach($parts as $i => $part) {
	//如果这个部分是HTML元素则跳过
	if(isset($part[0]) && ($part[0] == '<')) { continue; }
	//将这些单词用<span/>包围起来
	$parts[$i] = str_replace($words, $replacements, $part);
}

$body = implode('', $parts);

echo $body;

说明：

preg_split() 函数中使用的正则表达式匹配 HTML 标签

<(?:"[^"]*"|'[^']*'|[^'">])*>

可以这样理解：

<                                //开始尖括号
    (?:                         //任意数量的 
        "[^"]*"                //双引号字符串
        |                        //或
        '[^']*'                  //单引号字符串
        |                        //或
        [^'">]                  //除去单引号、双引号和>的其他文本
    )*                          
>                                //结束尖括号

但是这种方法无法高亮最后一个 Herring，因为它的首字母是大写的。要完全不区分大小写的更改，需要把 str_replace() 方法 改为 preg_replace() 方法：

<style>
	.word-0{ background-color: yellow; }
	.word-1{ border:1px solid red; }
</style>

<?php

header('Content-type:text/html;charset=utf-8');

/* 标记Web页面 */
$body = '
<p>I like pickles and hrrring.</p>
<a href="pickle.php"><img width="200" src="pickle.png">A pickle pic</a>
I have herringbone-patterned toaster cozy.
<herring>Herring is not a real HTML element!</herring>
';

$words = array('pickle', 'herring');
$replacements = array();
foreach($words as $i => $word) {
	$patterns[] = '/'.preg_quote($word).'/i'; 
	//preg_quote()需要参数 str 并向其中 每个正则表达式语法中的字符前增加一个反斜线。正则表达式特殊字符有： .  + * ? [ ^ ] $ ( ) { } = ! < > | : -
	$replacements[] = "<span class='word-$i'>\0</span>";
}

// 将页面分解为多个块
// 由看上去类似HTML元素的部分分隔
$parts = preg_split("{(<(?:"[^"]*"|'[^']*'|[^'">])*>)}", $body, -1, PREG_SPLIT_DELIM_CAPTURE);
var_dump($parts);
/*
array (size=15)
  0 => string '
' (length=2)
  1 => string '<p>' (length=3)
  2 => string 'I like pickles and hrrring.' (length=27)
  3 => string '</p>' (length=4)
  4 => string '
' (length=2)
  5 => string '<a href="pickle.php">' (length=21)
  6 => string '' (length=0)
  7 => string '<img width="200" src="pickle.png">' (length=34)
  8 => string 'A pickle pic' (length=12)
  9 => string '</a>' (length=4)
  10 => string '
I have herringbone-patterned toaster cozy.
' (length=46)
  11 => string '<herring>' (length=9)
  12 => string 'Herring is not a real HTML element!' (length=35)
  13 => string '</herring>' (length=10)
  14 => string '
' (length=2)
*/

foreach($parts as $i => $part) {
	//如果这个部分是HTML元素则跳过
	if(isset($part[0]) && ($part[0] == '<')) { continue; }
	//将这些单词用<span/>包围起来
	$parts[$i] = preg_replace($patterns, $replacements, $part);
}

$body = implode('', $parts);

echo $body;

　　

参考：

<PHP Cookbook>3'rd

《精通正则表达式》第3版