includes/normal/UtfNormalMemStress.php

   1 <?php
   2 /**
   3  * Approximate benchmark for some basic operations.
   4  * Runs large chunks of text through cleanup with a lowish memory limit,
   5  * to test regression on mem usage (bug 28146)
   6  *
   7  * Copyright © 2004-2011 Brion Vibber <brion@wikimedia.org>
   8  * https://www.mediawiki.org/
   9  *
  10  * This program is free software; you can redistribute it and/or modify
  11  * it under the terms of the GNU General Public License as published by
  12  * the Free Software Foundation; either version 2 of the License, or
  13  * (at your option) any later version.
  14  *
  15  * This program is distributed in the hope that it will be useful,
  16  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  17  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  18  * GNU General Public License for more details.
  19  *
  20  * You should have received a copy of the GNU General Public License along
  21  * with this program; if not, write to the Free Software Foundation, Inc.,
  22  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  23  * http://www.gnu.org/copyleft/gpl.html
  24  *
  25  * @file
  26  * @ingroup UtfNormal
  27  */
  28
  29 if ( PHP_SAPI != 'cli' ) {
  30         die( "Run me from the command line please.\n" );
  31 }
  32
  33 if ( isset( $_SERVER['argv'] ) && in_array( '--icu', $_SERVER['argv'] ) ) {
  34         dl( 'php_utfnormal.so' );
  35 }
  36
  37 require_once 'UtfNormalDefines.php';
  38 require_once 'UtfNormalUtil.php';
  39 require_once 'UtfNormal.php';
  40
  41 define( 'BENCH_CYCLES', 1 );
  42 define( 'BIGSIZE', 1024 * 1024 * 10 ); // 10m
  43 ini_set( 'memory_limit', BIGSIZE + 120 * 1024 * 1024 );
  44
  45 $testfiles = array(
  46         'testdata/washington.txt' => 'English text',
  47         'testdata/berlin.txt' => 'German text',
  48         'testdata/bulgakov.txt' => 'Russian text',
  49         'testdata/tokyo.txt' => 'Japanese text',
  50         'testdata/young.txt' => 'Korean text'
  51 );
  52 $normalizer = new UtfNormal;
  53 UtfNormal::loadData();
  54 foreach ( $testfiles as $file => $desc ) {
  55         benchmarkTest( $normalizer, $file, $desc );
  56 }
  57
  58 # -------
  59
  60 function benchmarkTest( &$u, $filename, $desc ) {
  61         print "Testing $filename ($desc)...\n";
  62         $data = file_get_contents( $filename );
  63         $all = $data;
  64         while ( strlen( $all ) < BIGSIZE ) {
  65                 $all .= $all;
  66         }
  67         $data = $all;
  68         echo "Data is " . strlen( $data ) . " bytes.\n";
  69         $forms = array(
  70                 'quickIsNFCVerify',
  71                 'cleanUp',
  72         );
  73
  74         foreach ( $forms as $form ) {
  75                 if ( is_array( $form ) ) {
  76                         $str = $data;
  77                         foreach ( $form as $step ) {
  78                                 $str = benchmarkForm( $u, $str, $step );
  79                         }
  80                 } else {
  81                         benchmarkForm( $u, $data, $form );
  82                 }
  83         }
  84 }
  85
  86 function benchmarkForm( &$u, &$data, $form ) {
  87         #$start = microtime( true );
  88         for ( $i = 0; $i < BENCH_CYCLES; $i++ ) {
  89                 $start = microtime( true );
  90                 $out = $u->$form( $data, UtfNormal::$utfCanonicalDecomp );
  91                 $deltas[] = ( microtime( true ) - $start );
  92         }
  93         #$delta = (microtime( true ) - $start) / BENCH_CYCLES;
  94         sort( $deltas );
  95         $delta = $deltas[0]; # Take shortest time
  96
  97         $rate = intval( strlen( $data ) / $delta );
  98         $same = ( 0 == strcmp( $data, $out ) );
  99
 100         printf( " %20s %6.1fms %12s bytes/s (%s)\n",
 101                 $form,
 102                 $delta * 1000.0,
 103                 number_format( $rate ),
 104                 ( $same ? 'no change' : 'changed' ) );
 105
 106         return $out;
 107 }