在php中的字符串之间替换文本

Question

I need to find all instances of a two strings and replace all the text enclosed between them. Eg:

[LINK]mySite.com[NAME]Click to visit my site[ENDLINK]

Needs to be replaced with:

<a href="mySite.com">Click to visit my site</a>

So far I have the following which is working:

    if(stristr("$text","[LINK]") && stristr("$text","[ENDLINK]")):
        $text = str_ireplace("[LINK]","\n<a href=\"" ,"$text");
        $text = str_ireplace("[NAME]", "\" target=\"_self\"><u>", "$text");
        $text = str_ireplace("[ENDLINK]","</u></a>","$text");
    endif;

However, it replaces [NAME] in all the other custom tags I am using, for example I also have:

   if(stristr("$text","[FILE]") && stristr("$text","[ENDFILE]")):
        $text = str_ireplace("[FILE]","\n<a href=\"" ,"$text");
        $text = str_ireplace("[NAME]", "\" target=\"_blank\"><u>", "$text");
        $text = str_ireplace("[ENDFILE]","</u></a>","$text");
    endif;

As you can see [FILE] uses target="_blank" and [LINK] uses target="_self", when I use the above script it replaces all instances of [NAME] to "_blank".

Renaming the tag [NAME] isn't a viable option as there are 1000's of entries already using it in the database and the users are accustomed to using it as is.

Is there a simple method to get this working?

Answer 1

$text = "[LINK]picture-large.jpg[NAME][IMAGE]picture-small.jpg[ENDIMAGE][ENDLINK] ... [LINK]mySite.com[NAME]Click to visit my site[endlink] lorem ipsum dolor sit amet ... [LINK]anothersite.com[name]Don't click here[ENDLINK] and here goes the file: [FILE]file.pdf[NAME]This is link to file[ENDFILE]";

function replaceFiles($match) {
    return '<a href="'.$match[1].'" target="_blank"><u>'.$match[2].'</u></a>';
}

function replaceImages($match) {
    return '<img src="'.$match[1].'" />';
}

function replaceLinks($match) {
    return '<a href="'.$match[1].'" target="_self"><u>'.$match[2].'</u></a>';
}


$text = preg_replace_callback('/\[FILE\](.*?)\[NAME\](.*?)\[ENDFILE\]/i', 'replaceFiles', $text);
$text = preg_replace_callback('/\[IMAGE\](.*?)\[ENDIMAGE\]/i', 'replaceImages', $text);
$text = preg_replace_callback('/\[LINK\](.*?)\[NAME\](.*?)\[ENDLINK\]/i', 'replaceLinks', $text);

Answer 2

This will only return if the entire pattern is matched:

public function replaceLinks($text){
    return preg_replace('/\[LINK](.*?)\[NAME](.*?)\[ENDLINK]/', "<a href='$1'>$2</a>", $text);
}

First of all it is working by using a pattern which matches the whole thing, not just parts of the link.

Here is a walkthrough of the regular expression (not a literal step by step of how a regular expression engine works before you regex nerds complain):

1. Forward slashes are used to contain the expression.
2. We put \\ before the [ to escape it, otherwise it is interpreted as the start of a character class.
3. We match the literal LINK] .
4. We create a capturing group for any character . , any number of times * , but with a lazy modifier ? . This is back referenced by the $1 in the replacement string.
5. I won't explain the rest because it is the same process for the rest of the string.

Here is a PHPUnit class which demonstrates the correct behaviour:

class Test extends PHPUnit_Framework_TestCase {

  public function replaceLinks($text){
    return preg_replace('/\[LINK](.*?)\[NAME](.*?)\[ENDLINK]/', "<a href='$1'>$2</a>", $text);
  }

  public function testCase1(){

    // Must replace links
    $input = "[LINK]mySite.com[NAME]Click to visit my site[ENDLINK]";
    $output = $this->replaceLinks($input);
    $expected = "<a href='mySite.com'>Click to visit my site</a>";
    $this->assertEquals($expected, $output);

  }

  public function testCase2(){

    // Must not replace files 
    $input = "[FILE]mySite.com[NAME]Click to visit my site[ENDFILE]";
    $output = $this->replaceLinks($input);
    $expected = "[FILE]mySite.com[NAME]Click to visit my site[ENDFILE]";
    $this->assertEquals($expected, $output);
  }

}