Plain Text Cleaner

Code

<?php

/*
 * Copyright (c) 2026 Jeffery L. Paris <jparis@phpog.com>.
 * Free for personal and internal use. Paid project use requires visible credit
 * to Jeffery L. Paris. Corporate use requires a paid license fee unless a
 * separate written license states otherwise.
 */

/**
 * Cleans plain text while preserving normal punctuation, accents, and allowed line breaks.
 *
 * @param string $text Raw plain text.
 * @param int $max_length Maximum allowed length after cleaning. Use 0 for no cap.
 * @param bool $allow_line_breaks Whether to preserve line breaks.
 * @return array Cleaned text result.
 */
function ogCleanPlainText($text, $max_length = 5000, $allow_line_breaks = true) {
	$result = array(
		'success' => false,
		'message' => '',
		'data' => array()
	);

	$text = (string)$text;
	$max_length = (int)$max_length;
	$allow_line_breaks = (bool)$allow_line_breaks;

	$text = str_replace("\0", '', $text);
	$text = strip_tags($text);
	$text = str_replace(array("\r\n", "\r"), "\n", $text);

	if (!$allow_line_breaks) {
		$text = preg_replace('/\s+/', ' ', $text);
	} else {
		$text = preg_replace('/[\t ]+/', ' ', $text);
		$text = preg_replace('/\n{3,}/', "\n\n", $text);
	}

	$text = trim($text);
	$was_truncated = false;
	if ($max_length > 0 && strlen($text) > $max_length) {
		$text = substr($text, 0, $max_length);
		$text = rtrim($text);
		$was_truncated = true;
	}

	$result['success'] = true;
	$result['message'] = 'Plain text cleaned.';
	$result['data'] = array(
		'text' => $text,
		'length' => strlen($text),
		'was_truncated' => $was_truncated
	);

	return $result;
}