rybakit/bench_utf8_regex_vs_u.php

## bench_utf8_regex_vs_u.php
<?php

const UTF8_REGEX = '/\A(?:
          [\x00-\x7F]++                      # ASCII
        | [\xC2-\xDF][\x80-\xBF]             # non-overlong 2-byte
        |  \xE0[\xA0-\xBF][\x80-\xBF]        # excluding overlongs
        | [\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}  # straight 3-byte
        |  \xED[\x80-\x9F][\x80-\xBF]        # excluding surrogates
        |  \xF0[\x90-\xBF][\x80-\xBF]{2}     # planes 1-3
        | [\xF1-\xF3][\x80-\xBF]{3}          # planes 4-15
        |  \xF4[\x80-\x8F][\x80-\xBF]{2}     # plane 16
        )*+\z/x';

const UTF8_U = '//u';

function utf8_regex($n, $str)
{
    for ($i = 0; $i < $n; ++$i) {
        \preg_match(UTF8_REGEX, $str);
    }
}

function utf8_u($n, $str)
{
    for ($i = 0; $i < $n; ++$i) {
        \preg_match(UTF8_U, $str);
    }
}

function empty_loop($n, $str)
{
    for ($i = 0; $i < $n; ++$i) {
    }
}

function getmicrotime()
{
    $t = \gettimeofday();

    return $t['sec'] + $t['usec'] / 1000000;
}

function start_test()
{
    \ob_start();

    return getmicrotime();
}

function end_test($start, $name, $overhead = null)
{
    global $total;
    global $last_time;
    $end = getmicrotime();
    \ob_end_clean();
    $last_time = $end - $start;
    $total += $last_time;
    $num = \number_format($last_time, 3);
    $pad = \str_repeat(' ', 24 - \strlen($name) - \strlen($num));
    if (null === $overhead) {
        echo $name.$pad.$num."\n";
    } else {
        $num2 = \number_format($last_time - $overhead, 3);
        echo $name.$pad.$num.'    '.$num2."\n";
    }
    \ob_start();

    return getmicrotime();
}

function total()
{
    global $total;
    $pad = \str_repeat('-', 24);
    echo $pad."\n";
    $num = \number_format($total, 3);
    $pad = \str_repeat(' ', 24 - \strlen('Total') - \strlen($num));
    echo 'Total'.$pad.$num."\n";
}

const N = 200000;

$str1_s = \str_repeat('c', 100);
$str1_m = \str_repeat('c', 1000);
$str1_l = \str_repeat('c', 10000);

$str2_s = \str_repeat('c', 49)."\x80".\str_repeat('c', 50);
$str2_m = \str_repeat('c', 499)."\x80".\str_repeat('c', 500);
$str2_l = \str_repeat('c', 4999)."\x80".\str_repeat('c', 5000);

$t0 = $t = start_test();
empty_loop(N, $str);
$t = end_test($t, 'empty_loop');
$overhead = $last_time;

\ob_end_clean();
echo "\nutf8:\n\n";
\ob_start();


utf8_regex(N, $str1_s);
$t = end_test($t, 'S regex', $overhead);
utf8_u(N, $str1_s);
$t = end_test($t, 'S //u', $overhead);

utf8_regex(N, $str1_m);
$t = end_test($t, 'M regex', $overhead);
utf8_u(N, $str1_m);
$t = end_test($t, 'M //u', $overhead);

utf8_regex(N, $str1_l);
$t = end_test($t, 'L regex', $overhead);
utf8_u(N, $str1_l);
$t = end_test($t, 'L //u', $overhead);


\ob_end_clean();
echo "\nnon-utf8:\n\n";
\ob_start();

utf8_regex(N, $str2_s);
$t = end_test($t, 'S regex', $overhead);
utf8_u(N, $str2_s);
$t = end_test($t, 'S //u', $overhead);

utf8_regex(N, $str2_m);
$t = end_test($t, 'M regex', $overhead);
utf8_u(N, $str2_m);
$t = end_test($t, 'M //u', $overhead);

utf8_regex(N, $str2_l);
$t = end_test($t, 'L regex', $overhead);
utf8_u(N, $str2_l);
$t = end_test($t, 'L //u', $overhead);


total();

## output.txt
$ php -n -d pcre.jit=1 bench_utf8_regex_vs_u.php                                                                                                                                                                [40/1527]
empty_loop         0.003

utf8:

S regex            0.036    0.033
S //u              0.055    0.052
M regex            0.120    0.117
M //u              0.215    0.212
L regex            0.928    0.925
L //u              1.843    1.840

non-utf8:

S regex            0.030    0.027
S //u              0.023    0.021
M regex            0.072    0.069
M //u              0.106    0.103
L regex            0.475    0.472
L //u              0.932    0.929
------------------------
Total              4.837


$ php -n -d pcre.jit=0 bench_utf8_regex_vs_u.php
empty_loop         0.004

utf8:

S regex            0.094    0.090
S //u              0.061    0.057
M regex            0.306    0.302
M //u              0.222    0.218
L regex            2.490    2.486
L //u              1.843    1.839

non-utf8:

S regex            0.079    0.076
S //u              0.023    0.019
M regex            0.189    0.186
M //u              0.107    0.103
L regex            1.271    1.267
L //u              0.915    0.911
------------------------
Total              7.604


$ php -n -v
PHP 7.3.8 (cli) (built: Jul 30 2019 09:26:16) ( NTS )
Copyright (c) 1997-2018 The PHP Group
Zend Engine v3.3.8, Copyright (c) 1998-2018 Zend Technologies
	<?php

	const UTF8_REGEX = '/\A(?:
	[\x00-\x7F]++ # ASCII
	\| [\xC2-\xDF][\x80-\xBF] # non-overlong 2-byte
	\| \xE0[\xA0-\xBF][\x80-\xBF] # excluding overlongs
	\| [\xE1-\xEC\xEE\xEF][\x80-\xBF]{2} # straight 3-byte
	\| \xED[\x80-\x9F][\x80-\xBF] # excluding surrogates
	\| \xF0[\x90-\xBF][\x80-\xBF]{2} # planes 1-3
	\| [\xF1-\xF3][\x80-\xBF]{3} # planes 4-15
	\| \xF4[\x80-\x8F][\x80-\xBF]{2} # plane 16
	)*+\z/x';

	const UTF8_U = '//u';

	function utf8_regex($n, $str)
	{
	for ($i = 0; $i < $n; ++$i) {
	\preg_match(UTF8_REGEX, $str);
	}
	}

	function utf8_u($n, $str)
	{
	for ($i = 0; $i < $n; ++$i) {
	\preg_match(UTF8_U, $str);
	}
	}

	function empty_loop($n, $str)
	{
	for ($i = 0; $i < $n; ++$i) {
	}
	}

	function getmicrotime()
	{
	$t = \gettimeofday();

	return $t['sec'] + $t['usec'] / 1000000;
	}

	function start_test()
	{
	\ob_start();

	return getmicrotime();
	}

	function end_test($start, $name, $overhead = null)
	{
	global $total;
	global $last_time;
	$end = getmicrotime();
	\ob_end_clean();
	$last_time = $end - $start;
	$total += $last_time;
	$num = \number_format($last_time, 3);
	$pad = \str_repeat(' ', 24 - \strlen($name) - \strlen($num));
	if (null === $overhead) {
	echo $name.$pad.$num."\n";
	} else {
	$num2 = \number_format($last_time - $overhead, 3);
	echo $name.$pad.$num.' '.$num2."\n";
	}
	\ob_start();

	return getmicrotime();
	}

	function total()
	{
	global $total;
	$pad = \str_repeat('-', 24);
	echo $pad."\n";
	$num = \number_format($total, 3);
	$pad = \str_repeat(' ', 24 - \strlen('Total') - \strlen($num));
	echo 'Total'.$pad.$num."\n";
	}

	const N = 200000;

	$str1_s = \str_repeat('c', 100);
	$str1_m = \str_repeat('c', 1000);
	$str1_l = \str_repeat('c', 10000);

	$str2_s = \str_repeat('c', 49)."\x80".\str_repeat('c', 50);
	$str2_m = \str_repeat('c', 499)."\x80".\str_repeat('c', 500);
	$str2_l = \str_repeat('c', 4999)."\x80".\str_repeat('c', 5000);

	$t0 = $t = start_test();
	empty_loop(N, $str);
	$t = end_test($t, 'empty_loop');
	$overhead = $last_time;

	\ob_end_clean();
	echo "\nutf8:\n\n";
	\ob_start();


	utf8_regex(N, $str1_s);
	$t = end_test($t, 'S regex', $overhead);
	utf8_u(N, $str1_s);
	$t = end_test($t, 'S //u', $overhead);

	utf8_regex(N, $str1_m);
	$t = end_test($t, 'M regex', $overhead);
	utf8_u(N, $str1_m);
	$t = end_test($t, 'M //u', $overhead);

	utf8_regex(N, $str1_l);
	$t = end_test($t, 'L regex', $overhead);
	utf8_u(N, $str1_l);
	$t = end_test($t, 'L //u', $overhead);


	\ob_end_clean();
	echo "\nnon-utf8:\n\n";
	\ob_start();

	utf8_regex(N, $str2_s);
	$t = end_test($t, 'S regex', $overhead);
	utf8_u(N, $str2_s);
	$t = end_test($t, 'S //u', $overhead);

	utf8_regex(N, $str2_m);
	$t = end_test($t, 'M regex', $overhead);
	utf8_u(N, $str2_m);
	$t = end_test($t, 'M //u', $overhead);

	utf8_regex(N, $str2_l);
	$t = end_test($t, 'L regex', $overhead);
	utf8_u(N, $str2_l);
	$t = end_test($t, 'L //u', $overhead);


	total();
	$ php -n -d pcre.jit=1 bench_utf8_regex_vs_u.php [40/1527]
	empty_loop 0.003

	utf8:

	S regex 0.036 0.033
	S //u 0.055 0.052
	M regex 0.120 0.117
	M //u 0.215 0.212
	L regex 0.928 0.925
	L //u 1.843 1.840

	non-utf8:

	S regex 0.030 0.027
	S //u 0.023 0.021
	M regex 0.072 0.069
	M //u 0.106 0.103
	L regex 0.475 0.472
	L //u 0.932 0.929
	------------------------
	Total 4.837


	$ php -n -d pcre.jit=0 bench_utf8_regex_vs_u.php
	empty_loop 0.004

	utf8:

	S regex 0.094 0.090
	S //u 0.061 0.057
	M regex 0.306 0.302
	M //u 0.222 0.218
	L regex 2.490 2.486
	L //u 1.843 1.839

	non-utf8:

	S regex 0.079 0.076
	S //u 0.023 0.019
	M regex 0.189 0.186
	M //u 0.107 0.103
	L regex 1.271 1.267
	L //u 0.915 0.911
	------------------------
	Total 7.604



	$ php -n -v
	PHP 7.3.8 (cli) (built: Jul 30 2019 09:26:16) ( NTS )
	Copyright (c) 1997-2018 The PHP Group
	Zend Engine v3.3.8, Copyright (c) 1998-2018 Zend Technologies