Question

What i tried(it’s a static for cpp only)=>

$str1 = "<pre                class="brush:cpp">";

$temp =  preg_replace('/&lt;pre\s+class=&quot;brush:cpp&quot;&gt;/','<pre class="brush:cpp">',$str1);

echo $temp . "\n";

That outputs=>

<pre class="brush:cpp">

But the $str1 can be

"<pre class="brush:cpp">"
"<pre class="brush:java">"
"<pre class="brush:php">"
"<pre class="brush:python">"

For those the output should be=>

<pre class="brush:cpp">
<pre class="brush:java">
<pre class="brush:php">
<pre class="brush:python">

Note: I can’t use html_entity_decode because the texts will contain other normal string and <br> for <br/>, i don’t want to do html_entity_decode for all texts.

I need a generic regular expression to catch cpp/java/php/python. How can i write a generic regular expression to save that part of pattern and keep it as it is in the replace string.

Answer 1

I believe something like this will work:

preg_replace('/&lt;pre\s+class=&quot;brush:(cpp|java|php|python)&quot;&gt;/','<pre class="brush:$1">',$str1);

It uses a capturing group to capture which ending is present, and it can be one of cpp/java/php/python. The replacement is made with the backreference #1, which will place whichever ending was captured.

Here is an example.