Question

The middle of the HTML string viewed as text, after/before nearest sentence.

Example:

$str = 'First sentence. Second sentence! <a title="Hello. World">
        Third sentence. </a> Fourth sentence. Fifth sentence?';

$middle = strlen($str) / 2;

what I did is to append a keyword to all stop characters like ., ! or ?, but only to the ones outside HTML tags:

$str = 'First sentence. Second sentence! <a title="Hello. World">
        Third sentence. </a> Fourth sentence. Fifth sentence?';
$str = preg_replace_callback("/(\.|\?|\!)(?!([^<]+)?>)/i",  function($matches){
   return $matches[1].'<!-- stop -->';
}, $str);

This turns $str into:

First sentence.<!-- stop --> Second sentence!<!-- stop --> <a title="Hello. 
World"> Third sentence.<!-- stop --> </a> Fourth sentence.<!-- stop --> Fifth
sentence?<!-- stop -->

Now, how can I find the <!-- stop --> sub-string that’s nearest to the $middle position, and insert my text before it?

Is there a way I can find the position of $matches[1] inside my preg_replace callback relative to the whole string? That way I could just compare it with $middle

Answer 1

<?php
    $str    = 'First sentence.<!-- stop --> Second sentence!<!-- stop --> <a title="Hello. 
    World"> Third sentence.<!-- stop --> </a> Fourth sentence.<!-- stop --> Fifth
    sentence?<!-- stop -->';
    $str1   =   explode('<!-- stop -->',$str);
    $str1[round(sizeof($str1)/2)-2].="!Append text!";
    $glue=htmlentities('<!-- stop --/>');
    echo $str2  =   implode($glue,$str1);
?>

how is this?