如何防止用第二个正则表达式重新替换？ [英] How to prevent of re-replacing by second regex?

问题描述

我的输入方式有两个正则表达式，这些是：

  // //用一个URL替换一个URL链接，就像这样：[LinkName]（LinkAddress）
 $ str = preg_replace（/ \ [（[^] [] *）] \（（[^（）] *）\） /，< a href ='$ 2'target ='_ blank'> $ 1< / a>，$ str）; 
 
 //用链接取代常规URL 
 $ str = preg_replace（/（\ b（？:( ?: https？| ftp）：\ / \ / | www\） -  A-Z0-9 +功放; @＃\ /％=〜_ |：？！;] * [ -  A-Z0-9 +功放; @＃\ /％ = __]）/ i，< a href = \$ 1 \target = \_blank\> untitled< / a>，$ str）; 
  

现在有一个问题（某种程度上是碰撞）。对于普通的URL，一切都很好。但是对于基于模式的URL，存在一个问题：第一个正则表达式创建一个和第二个正则表达式的链接创建它的 href -attribute value。

我该如何解决它？

编辑：根据评论，我怎样才能创建一个正则表达式而不是这两个正则表达式？ （使用 preg_replace_callback ）。老实说，我尝试过，但它不适用于任何类型的网址。

是否可以组合它们？因为那些输出不一样。第一个有LinkName，第二个有一个常量字符串 untitled 作为它的LinkName。

解决方案

  $ str = preg_replace_callback（'/ \ [（[^] [] *）] \（（[^（）] *）\）| （\b（:( ?: HTTPS | FTP）：\ / \ / | www\？）[ -  A-Z0-9 +功放; @＃\ /％=〜_ |？！ ：，。;] * [ -  a-z0-9 +& @＃\ /％=〜_ |]）/ i'，
 function（$ matches）{
 if（isset $ matches [3]））{
 //用一个链接替换一个普通的URL 
 return< a href ='。$ matches [3]。''target ='_ blank'> untitled< / a>; 
} else {
 //用一个类似于这种模式的链接替换URL：[LinkName]（LinkAddress）
 return< a href = $ matches [2]。target ='_ blank'>。matches [1]。< / a>; 
} 
}，$ str）; 
 
 echo $ str; 
  

一种方法是这样做。您将两个表达式与替代字符 | 合并在一起。然后在你的回调函数中，你只需检查你的第三个捕获组是否被设置（ isset（$ matches [3]）），如果是，那么你的第二个正则表达式匹配字符串并且你替换一个正常的链接，否则你用link / linktext替换。

我希望你了解一切，我可以帮你。

I have two regex(s) on the way of my input, these:

// replace a URL with a link which is like this pattern: [LinkName](LinkAddress)
$str= preg_replace("/\[([^][]*)]\(([^()]*)\)/", "<a href='$2' target='_blank'>$1</a>", $str);

// replace a regular URL with a link
$str = preg_replace("/(\b(?:(?:https?|ftp):\/\/|www\.)[-a-z0-9+&@#\/%?=~_|!:,.;]*[-a-z0-9+&@#\/%=~_|])/i","<a href=\"$1\" target=\"_blank\">untitled</a>", $str);

Now there is a problem (somehow a collision). For regular URLs everything is fine. But for a pattern-based URLs, there is a problem: The first regex create a link of that and second regex again create a link of its href-attribute value.

How can I fix it?

Edit: According to the comments, how can I create a single regex instead of those two regex? (using preg_replace_callback). Honestly I tried it but it doesn't work for none kind of URLs ..

Is combining them possible? Because the output of those isn't identical. The first one has a LinkName and the second one has a constant string untitled as its LinkName.

解决方案

$str = preg_replace_callback('/\[([^][]*)]\(([^()]*)\)|(\b(?:(?:https?|ftp):\/\/|www\.)[-a-z0-9+&@#\/%?=~_|!:,.;]*[-a-z0-9+&@#\/%=~_|])/i', 
function($matches) {
    if(isset($matches[3])) {
        // replace a regular URL with a link
        return "<a href='".$matches[3]."' target='_blank'>untitled</a>";
    } else {
        // replace a URL with a link which is like this pattern: [LinkName](LinkAddress)
        return "<a href=".$matches[2]." target='_blank'>".$matches[1]."</a>";
    }
}, $str);

echo $str;

One way would be to do it like this. You merge your two expressions together with the alternative character |. Then in your callback function you just check if your third capture group is set (isset($matches[3])) and if yes, then your second regular expression matched the string and you replace a normal link, otherwise you replace with link/linktext.

I hope you understand everything and I could help you.

这篇关于如何防止用第二个正则表达式重新替换？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！