Benchmark for validating ascii characters

01_the_result_of_benchmark_for_validating_ascii_characters.txt

[ctype]
valid:true
time:4.9642469882965
[byte comparison]
valid:true
time:5.0199990272522
[preg_match]
valid:true
time:5.9788029193878
[mb_check_encoding]
valid:true
time:6.5968248844147

02_benchmark_for_validating_ascii_characters.php

<?php
// for avoiding Mac OSX's bug in C library
setlocale(LC_ALL, 'C');

$result = [
    'ctype' => timer('is_ascii'),
    'byte comparison' => timer('is_ascii2'),
    'preg_match' => timer('is_ascii3'),
    'mb_check_encoding' => timer('is_ascii4')
];

foreach ($result as $desc => $elem) {

    echo '[', $desc, ']', PHP_EOL,
         'valid:', $ret['valid'] ? 'true' : 'false', PHP_EOL,
         'time:', $ret['time'], PHP_EOL;

}

function timer(callable $callable)
{
    $ret = [];
    $expected = array_map('chr', range(0, 0x7F));
    $start = microtime(true);

    for ($i = 0; $i < 0x10FFFF; $i += 1) {

        $char = utf8_chr($i);

        if ($callable($char)) {
            $ret[] = $char;
        }

    }

    $stop = microtime(true);

    return ['valid' => $expected === $expected, 'time' => $stop - $start];
}


function utf8_chr($code_point)
{
    if ($code_point < 0 || (0xD800 <= $code_point && $code_point <= 0xDFFF) || 0x10FFFF < $code_point) {
        return false;
    }

    $ret = str_repeat('0', 8 - strlen(dechex($code_point))).dechex($code_point);
    $ret = hex2bin($ret);
    
    return mb_convert_encoding($ret, 'UTF-8', 'UTF-32BE');
}


function is_ascii($str)
{
    return ctype_print($str) || ctype_cntrl($str);
}


function is_ascii2($str)
{
    $len = strlen($str);

    for ($i = 0; $i < $len; $i += 1) {

        if ("\x7F" < $str[$i]) {

            return false;

        }

    }

    return true;
}

function is_ascii3($str)
{
    return preg_match('/[\x00-\x7F]/', $str);
}

function is_ascii4($str)
{
    return mb_check_encoding($str, 'ASCII');
}