source code
<?php
/* The Computer Language Benchmarks Game
https://salsa.debian.org/benchmarksgame-team/benchmarksgame/
regex-dna program contributed by Danny Sauer
modified by Josh Goldfoot
modified by Sergey Khripunov
modified by Craig Russell
converted from regex-dna program
*/
$tok = ftok(__FILE__, chr(time() & 255));
$queue = msg_get_queue($tok);
$variants = array(
'agggtaaa|tttaccct',
'[cgt]gggtaaa|tttaccc[acg]',
'a[act]ggtaaa|tttacc[agt]t',
'ag[act]gtaaa|tttac[agt]ct',
'agg[act]taaa|ttta[agt]cct',
'aggg[acg]aaa|ttt[cgt]ccct',
'agggt[cgt]aa|tt[acg]accct',
'agggta[cgt]a|t[acg]taccct',
'agggtaa[cgt]|[acg]ttaccct',
);
// IUB replacement parallel arrays
$IUB = array(); $IUBnew = array();
$IUB[]='/tHa[Nt]/S'; $IUBnew[]='<4>';
$IUB[]='/aND|caN|Ha[DS]|WaS/S'; $IUBnew[]='<3>';
$IUB[]='/a[NSt]|BY/S'; $IUBnew[]='<2>';
$IUB[]='/<[^>]*>/S'; $IUBnew[]='|';
$IUB[]='/\\|[^|][^|]*\\|/S'; $IUBnew[]='-';
// read in file
$contents = file_get_contents('php://stdin');
$initialLength = strlen($contents);
// remove things
$contents = preg_replace('/^>.*$|\n/mS', '', $contents);
$codeLength = strlen($contents);
// do regexp counts
$messages = array_flip($variants);
$workers = $results = array();
foreach ($variants as $key => $regex){
if($key == 0 || $key == 2 || $key == 4 || $key == 6) {
$pid = pcntl_fork();
if($pid) $workers[] = $pid;
}
if($pid && $key > 7) {
$messages[$regex] =
preg_match_all('/' . $regex . '/iS', $contents, $discard);
}
else if(!$pid) {
$results[] = $regex . ',' .
preg_match_all('/' . $regex . '/iS', $contents, $discard);
if($key == 1 || $key == 3 || $key == 5 || $key == 7) {
msg_send($queue, 2, implode(';', $results), false, false, $errno);
exit;
}
}
}
// receive and output the counts
pcntl_wait($status);
foreach($workers as $worker) {
msg_receive($queue, 2, $msgtype, 4096, $message, false);
$message = explode(';', $message, 3);
foreach($message as $line) {
$tmp = explode(',', $line, 2);
$messages[$tmp[0]] = $tmp[1];
}
}
foreach($messages as $regex => $count) {
echo $regex, ' ', $count, "\n";
}
// do replacements
$contents = preg_replace($IUB, $IUBnew, $contents);
echo "\n",
$initialLength, "\n",
$codeLength, "\n",
strlen($contents), "\n";
notes, command-line, and program output
NOTES:
64-bit Ubuntu quad core
PHP 8.3.11 (cli)
(built: Sep 5 2024 12:34:23) (NTS)
Copyright (c) The PHP Group
Zend Engine v4.3.11,
with Zend OPcache v8.3.11,
Copyright (c) Zend Technologies
Thu, 05 Sep 2024 21:11:29 GMT
COMMAND LINE:
/opt/src/php-8.3.11/bin/php -dzend_extension=/opt/src/php-8.3.11/lib/php/extensions/no-debug-non-zts-20230831/opcache.so -dopcache.enable_cli=1 -dopcache.jit_buffer_size=64M -n -d memory_limit=512M regexredux.php 0 < regexredux-input5000000.txt
PROGRAM OUTPUT:
agggtaaa|tttaccct 356
[cgt]gggtaaa|tttaccc[acg] 1250
a[act]ggtaaa|tttacc[agt]t 4252
ag[act]gtaaa|tttac[agt]ct 2894
agg[act]taaa|ttta[agt]cct 5435
aggg[acg]aaa|ttt[cgt]ccct 1537
agggt[cgt]aa|tt[acg]accct 1431
agggta[cgt]a|t[acg]taccct 1608
agggtaa[cgt]|[acg]ttaccct 2178
50833411
50000000
27388361