Question

ok I have two strings.

(I use this for a language library system to allow translators to provide translations with placeholders).

In the first string, there are two instances. note that it’s not always a single instance, some cases it will be none, one, two, or more.

This is a {[John Doe]} and this is {[Jane Doe]}

and then I have a string that is stored like this:

C’est {[1]} et c’est {[2]}

(translation)
This is a {[1]} and this is a {[2]}

so what I need to do is take the first string, replace everything between {[]} of the starting string and match each instance, i.e. first of first string with {[1]} of second string etc. keep in mind that the reason I am using {[1]} and {[2]} is because in some languages, terms may appear in a different order for gramatical accuracy, but are still terms that don’t need translation them selves (names).

so the question is. how do I do this? am thinking preg_split and then match index+1 of each with the second string. that part I can handle. the problem I am having is getting the right regex search going..

this is as close as I could get it..

preg_split('/[(\{\[).*(\]\})]/', $str, -1, PREG_SPLIT_NO_EMPTY | PREG_SPLIT_DELIM_CAPTURE);

that returns an array of everything before and after each instance of {[ and ]} when I am just trying to get the contents of inbetween the two..

EDIT: solution derived from NikiC’s answer.

function lang($str){
    $nwStr = $str;
    preg_match_all('(\{\[(.+?)\]\})', $str, $placeholders);
    foreach ($placeholders[0] as $mk => $match) {
        $pos = $mk+1;
        $nwStr = str_replace("$match","{[$pos]}",$nwStr);
    }
    $result = preg_replace_callback('(\{\[(\d+)\]\})', function ($matches) use ($placeholders) {
        $n = $matches[1]-1;
        return $placeholders[1][$n];
    }, $translation);
    return $result;

}

basically what i am doing here is first looping through to replace the matches with the placeholders so that I can match the proper placeholder text in my language files. (i.e. create the right label string out of the input string)

Answer 1

First grab the placeholders from the string:

preg_match_all('(\{\[(.+?)\]\})', $string, $matches);
$placeholders = $matches[1];

Now replace with a callback:

$result = preg_replace_callback('(\{\[(\d+)\]\})', function ($matches) use ($placeholders) {
    $n = $matches[1] + 1;
    return $placeholders[$n];
}, $translation);