tests/phpunit/includes/libs/normal/UtfNormalTest.php

   1 <?php
   2 /**
   3  * Implements the conformance test at:
   4  * http://www.unicode.org/Public/UNIDATA/NormalizationTest.txt
   5  *
   6  * Copyright © 2004 Brion Vibber <brion@pobox.com>
   7  * https://www.mediawiki.org/
   8  *
   9  * This program is free software; you can redistribute it and/or modify
  10  * it under the terms of the GNU General Public License as published by
  11  * the Free Software Foundation; either version 2 of the License, or
  12  * (at your option) any later version.
  13  *
  14  * This program is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  17  * GNU General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU General Public License along
  20  * with this program; if not, write to the Free Software Foundation, Inc.,
  21  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  22  * http://www.gnu.org/copyleft/gpl.html
  23  *
  24  * @file
  25  * @group UtfNormal
  26  * @group large
  27  */
  28
  29 class UtfNormalTest extends PHPUnit_Framework_TestCase {
  30
  31         protected static $testedChars = array();
  32
  33         public static function provideNormalizationTest() {
  34                 global $IP;
  35                 $in = fopen( "$IP/tests/phpunit/data/normal/NormalizationTest.txt", "rt" );
  36
  37                 $testCases = array();
  38                 while ( false !== ( $line = fgets( $in ) ) ) {
  39                         list( $data, $comment ) = explode( '#', $line );
  40                         if ( $data === '' ) continue;
  41                         $matches = array();
  42                         if ( preg_match( '/@Part([\d])/', $data, $matches ) ) {
  43                                 continue;
  44                         }
  45
  46                         $columns = array_map( "hexSequenceToUtf8", explode( ";", $data ) );
  47                         array_unshift( $columns, '' );
  48
  49                         self::$testedChars[$columns[1]] = true;
  50                         $testCases[] = array( $columns, $comment );
  51                 }
  52                 fclose( $in );
  53
  54                 return array( array( $testCases ) );
  55         }
  56
  57         function assertStringEquals( $a, $b, $desc ) {
  58                 $this->assertEquals( 0, strcmp( $a, $b ), $desc );
  59         }
  60
  61         function assertNFC( $c, $desc ) {
  62                 $this->assertStringEquals( $c[2], UtfNormal::toNFC( $c[1] ), $desc );
  63                 $this->assertStringEquals( $c[2], UtfNormal::toNFC( $c[2] ), $desc );
  64                 $this->assertStringEquals( $c[2], UtfNormal::toNFC( $c[3] ), $desc );
  65                 $this->assertStringEquals( $c[4], UtfNormal::toNFC( $c[4] ), $desc );
  66                 $this->assertStringEquals( $c[4], UtfNormal::toNFC( $c[5] ), $desc );
  67         }
  68
  69         function assertNFD( $c, $desc ) {
  70                 $this->assertStringEquals( $c[3], UtfNormal::toNFD( $c[1] ), $desc );
  71                 $this->assertStringEquals( $c[3], UtfNormal::toNFD( $c[2] ), $desc );
  72                 $this->assertStringEquals( $c[3], UtfNormal::toNFD( $c[3] ), $desc );
  73                 $this->assertStringEquals( $c[5], UtfNormal::toNFD( $c[4] ), $desc );
  74                 $this->assertStringEquals( $c[5], UtfNormal::toNFD( $c[5] ), $desc );
  75         }
  76
  77         function assertNFKC( $c, $desc ) {
  78                 $this->assertStringEquals( $c[4], UtfNormal::toNFKC( $c[1] ), $desc );
  79                 $this->assertStringEquals( $c[4], UtfNormal::toNFKC( $c[2] ), $desc );
  80                 $this->assertStringEquals( $c[4], UtfNormal::toNFKC( $c[3] ), $desc );
  81                 $this->assertStringEquals( $c[4], UtfNormal::toNFKC( $c[4] ), $desc );
  82                 $this->assertStringEquals( $c[4], UtfNormal::toNFKC( $c[5] ), $desc );
  83         }
  84
  85         function assertNFKD( $c, $desc ) {
  86                 $this->assertStringEquals( $c[5], UtfNormal::toNFKD( $c[1] ), $desc );
  87                 $this->assertStringEquals( $c[5], UtfNormal::toNFKD( $c[2] ), $desc );
  88                 $this->assertStringEquals( $c[5], UtfNormal::toNFKD( $c[3] ), $desc );
  89                 $this->assertStringEquals( $c[5], UtfNormal::toNFKD( $c[4] ), $desc );
  90                 $this->assertStringEquals( $c[5], UtfNormal::toNFKD( $c[5] ), $desc );
  91         }
  92
  93         function assertCleanUp( $c, $desc ) {
  94                 $this->assertStringEquals( $c[2], UtfNormal::cleanUp( $c[1] ), $desc );
  95                 $this->assertStringEquals( $c[2], UtfNormal::cleanUp( $c[2] ), $desc );
  96                 $this->assertStringEquals( $c[2], UtfNormal::cleanUp( $c[3] ), $desc );
  97                 $this->assertStringEquals( $c[4], UtfNormal::cleanUp( $c[4] ), $desc );
  98                 $this->assertStringEquals( $c[4], UtfNormal::cleanUp( $c[5] ), $desc );
  99         }
 100
 101         /**
 102          * The data provider for this intentionally returns all the
 103          * test cases as one since PHPUnit is too slow otherwise
 104          *
 105          * @dataProvider provideNormalizationTest
 106          */
 107         function testNormals( $testCases ) {
 108                 foreach ( $testCases as $case ) {
 109                         $c = $case[0];
 110                         $desc = $case[1];
 111                         $this->assertNFC( $c, $desc );
 112                         $this->assertNFD( $c, $desc );
 113                         $this->assertNFKC( $c, $desc );
 114                         $this->assertNFKD( $c, $desc );
 115                         $this->assertCleanUp( $c, $desc );
 116                 }
 117         }
 118
 119         public static function provideUnicodeData() {
 120                 global $IP;
 121                 $in = fopen( "$IP/tests/phpunit/data/normal/UnicodeData.txt", "rt" );
 122                 $testCases = array();
 123                 while ( false !== ( $line = fgets( $in ) ) ) {
 124                         $cols = explode( ';', $line );
 125                         $char = codepointToUtf8( hexdec( $cols[0] ) );
 126                         $desc = $cols[0] . ": " . $cols[1];
 127                         if ( $char < "\x20" || $char >= UTF8_SURROGATE_FIRST && $char <= UTF8_SURROGATE_LAST ) {
 128                                 # Can't check NULL with the ICU plugin, as null bytes fail in C land.
 129                                 # Skip other control characters, as we strip them for XML safety.
 130                                 # Surrogates are illegal on their own or in UTF-8, ignore.
 131                                 continue;
 132                         }
 133                         if ( empty( self::$testedChars[$char] ) ) {
 134                                 $testCases[] = array( $char, $desc );
 135                         }
 136                 }
 137                 fclose( $in );
 138
 139                 return array( array( $testCases ) );
 140         }
 141
 142         /**
 143          * The data provider for this intentionally returns all the
 144          * test cases as one since PHPUnit is too slow otherwise
 145          *
 146          * @depends testNormals
 147          * @dataProvider provideUnicodeData
 148          */
 149         public function testInvariant( $testCases ) {
 150                 foreach ( $testCases as $case ) {
 151                         $char = $case[0];
 152                         $desc = $case[1];
 153                         $this->assertStringEquals( $char, UtfNormal::toNFC( $char ), $desc );
 154                         $this->assertStringEquals( $char, UtfNormal::toNFD( $char ), $desc );
 155                         $this->assertStringEquals( $char, UtfNormal::toNFKC( $char ), $desc );
 156                         $this->assertStringEquals( $char, UtfNormal::toNFKD( $char ), $desc );
 157                         $this->assertStringEquals( $char, UtfNormal::cleanUp( $char ), $desc );
 158                 }
 159         }
 160 }