tests/phpunit/includes/title/MediaWikiTitleCodecTest.php

   1 <?php
   2 /**
   3  * This program is free software; you can redistribute it and/or modify
   4  * it under the terms of the GNU General Public License as published by
   5  * the Free Software Foundation; either version 2 of the License, or
   6  * (at your option) any later version.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  11  * GNU General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public License along
  14  * with this program; if not, write to the Free Software Foundation, Inc.,
  15  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  16  * http://www.gnu.org/copyleft/gpl.html
  17  *
  18  * @file
  19  * @license GPL 2+
  20  * @author Daniel Kinzler
  21  */
  22
  23 /**
  24  * @covers MediaWikiTitleCodec
  25  *
  26  * @group Title
  27  * @group Database
  28  *        ^--- needed because of global state in
  29  */
  30 class MediaWikiTitleCodecTest extends MediaWikiTestCase {
  31
  32         public function setUp() {
  33                 parent::setUp();
  34
  35                 $this->setMwGlobals( array(
  36                         'wgLanguageCode' => 'en',
  37                         'wgContLang' => Language::factory( 'en' ),
  38                         // User language
  39                         'wgLang' => Language::factory( 'en' ),
  40                         'wgAllowUserJs' => false,
  41                         'wgDefaultLanguageVariant' => false,
  42                         'wgLocalInterwikis' => array( 'localtestiw' ),
  43
  44                         // NOTE: this is why global state is evil.
  45                         // TODO: refactor access to the interwiki codes so it can be injected.
  46                         'wgHooks' => array(
  47                                 'InterwikiLoadPrefix' => array(
  48                                         function ( $prefix, &$data ) {
  49                                                 if ( $prefix === 'localtestiw' ) {
  50                                                         $data = array( 'iw_url' => 'localtestiw' );
  51                                                 } elseif ( $prefix === 'remotetestiw' ) {
  52                                                         $data = array( 'iw_url' => 'remotetestiw' );
  53                                                 }
  54                                                 return false;
  55                                         }
  56                                 )
  57                         )
  58                 ) );
  59         }
  60
  61         /**
  62          * Returns a mock GenderCache that will consider a user "female" if the
  63          * first part of the user name ends with "a".
  64          *
  65          * @return GenderCache
  66          */
  67         private function getGenderCache() {
  68                 $genderCache = $this->getMockBuilder( 'GenderCache' )
  69                         ->disableOriginalConstructor()
  70                         ->getMock();
  71
  72                 $genderCache->expects( $this->any() )
  73                         ->method( 'getGenderOf' )
  74                         ->will( $this->returnCallback( function( $userName ) {
  75                                 return preg_match( '/^[^- _]+a( |_|$)/u', $userName ) ? 'female' : 'male';
  76                         } ) );
  77
  78                 return $genderCache;
  79         }
  80
  81         protected function makeCodec( $lang ) {
  82                 $gender = $this->getGenderCache();
  83                 $lang = Language::factory( $lang );
  84                 return new MediaWikiTitleCodec( $lang, $gender );
  85         }
  86
  87         public function provideFormat() {
  88                 return array(
  89                         array( NS_MAIN, 'Foo_Bar', '', 'en', 'Foo Bar' ),
  90                         array( NS_USER, 'Hansi_Maier', 'stuff_and_so_on', 'en', 'User:Hansi Maier#stuff and so on' ),
  91                         array( false, 'Hansi_Maier', '', 'en', 'Hansi Maier' ),
  92                         array( NS_USER_TALK, 'hansi__maier', '', 'en', 'User talk:hansi  maier', 'User talk:Hansi maier' ),
  93
  94                         // getGenderCache() provides a mock that considers first
  95                         // names ending in "a" to be female.
  96                         array( NS_USER, 'Lisa_Müller', '', 'de', 'Benutzerin:Lisa Müller' ),
  97                 );
  98         }
  99
 100         /**
 101          * @dataProvider provideFormat
 102          */
 103         public function testFormat( $namespace, $text, $fragment, $lang, $expected, $normalized = null ) {
 104                 if ( $normalized === null ) {
 105                         $normalized = $expected;
 106                 }
 107
 108                 $codec = $this->makeCodec( $lang );
 109                 $actual = $codec->formatTitle( $namespace, $text, $fragment );
 110
 111                 $this->assertEquals( $expected, $actual, 'formatted' );
 112
 113                 // test round trip
 114                 $parsed = $codec->parseTitle( $actual, NS_MAIN );
 115                 $actual2 = $codec->formatTitle( $parsed->getNamespace(), $parsed->getText(), $parsed->getFragment() );
 116
 117                 $this->assertEquals( $normalized, $actual2, 'normalized after round trip' );
 118         }
 119
 120         public function provideGetText() {
 121                 return array(
 122                         array( NS_MAIN, 'Foo_Bar', '', 'en', 'Foo Bar' ),
 123                         array( NS_USER, 'Hansi_Maier', 'stuff_and_so_on', 'en', 'Hansi Maier' ),
 124                 );
 125         }
 126
 127         /**
 128          * @dataProvider provideGetText
 129          */
 130         public function testGetText( $namespace, $dbkey, $fragment, $lang, $expected ) {
 131                 $codec = $this->makeCodec( $lang );
 132                 $title = new TitleValue( $namespace, $dbkey, $fragment );
 133
 134                 $actual = $codec->getText( $title );
 135
 136                 $this->assertEquals( $expected, $actual );
 137         }
 138
 139         public function provideGetPrefixedText() {
 140                 return array(
 141                         array( NS_MAIN, 'Foo_Bar', '', 'en', 'Foo Bar' ),
 142                         array( NS_USER, 'Hansi_Maier', 'stuff_and_so_on', 'en', 'User:Hansi Maier' ),
 143
 144                         // No capitalization or normalization is applied while formatting!
 145                         array( NS_USER_TALK, 'hansi__maier', '', 'en', 'User talk:hansi  maier' ),
 146
 147                         // getGenderCache() provides a mock that considers first
 148                         // names ending in "a" to be female.
 149                         array( NS_USER, 'Lisa_Müller', '', 'de', 'Benutzerin:Lisa Müller' ),
 150                 );
 151         }
 152
 153         /**
 154          * @dataProvider provideGetPrefixedText
 155          */
 156         public function testGetPrefixedText( $namespace, $dbkey, $fragment, $lang, $expected ) {
 157                 $codec = $this->makeCodec( $lang );
 158                 $title = new TitleValue( $namespace, $dbkey, $fragment );
 159
 160                 $actual = $codec->getPrefixedText( $title );
 161
 162                 $this->assertEquals( $expected, $actual );
 163         }
 164
 165         public function provideGetFullText() {
 166                 return array(
 167                         array( NS_MAIN, 'Foo_Bar', '', 'en', 'Foo Bar' ),
 168                         array( NS_USER, 'Hansi_Maier', 'stuff_and_so_on', 'en', 'User:Hansi Maier#stuff and so on' ),
 169
 170                         // No capitalization or normalization is applied while formatting!
 171                         array( NS_USER_TALK, 'hansi__maier', '', 'en', 'User talk:hansi  maier' ),
 172                 );
 173         }
 174
 175         /**
 176          * @dataProvider provideGetFullText
 177          */
 178         public function testGetFullText( $namespace, $dbkey, $fragment, $lang, $expected ) {
 179                 $codec = $this->makeCodec( $lang );
 180                 $title = new TitleValue( $namespace, $dbkey, $fragment );
 181
 182                 $actual = $codec->getFullText( $title );
 183
 184                 $this->assertEquals( $expected, $actual );
 185         }
 186
 187         public function provideParseTitle() {
 188                 //TODO: test capitalization and trimming
 189                 //TODO: test unicode normalization
 190
 191                 return array(
 192                         array( '  : Hansi_Maier _ ', NS_MAIN, 'en',
 193                                 new TitleValue( NS_MAIN, 'Hansi_Maier', '' ) ),
 194                         array( 'User:::1', NS_MAIN, 'de',
 195                                 new TitleValue( NS_USER, '0:0:0:0:0:0:0:1', '' ) ),
 196                         array( ' lisa Müller', NS_USER, 'de',
 197                                 new TitleValue( NS_USER, 'Lisa_Müller', '' ) ),
 198                         array( 'benutzerin:lisa Müller#stuff', NS_MAIN, 'de',
 199                                 new TitleValue( NS_USER, 'Lisa_Müller', 'stuff' ) ),
 200
 201                         array( ':Category:Quux', NS_MAIN, 'en',
 202                                 new TitleValue( NS_CATEGORY, 'Quux', '' ) ),
 203                         array( 'Category:Quux', NS_MAIN, 'en',
 204                                 new TitleValue( NS_CATEGORY, 'Quux', '' ) ),
 205                         array( 'Category:Quux', NS_CATEGORY, 'en',
 206                                 new TitleValue( NS_CATEGORY, 'Quux', '' ) ),
 207                         array( 'Quux', NS_CATEGORY, 'en',
 208                                 new TitleValue( NS_CATEGORY, 'Quux', '' ) ),
 209                         array( ':Quux', NS_CATEGORY, 'en',
 210                                 new TitleValue( NS_MAIN, 'Quux', '' ) ),
 211
 212                         // getGenderCache() provides a mock that considers first
 213                         // names ending in "a" to be female.
 214
 215                         array( 'a b c', NS_MAIN, 'en',
 216                                 new TitleValue( NS_MAIN, 'A_b_c' ) ),
 217                         array( ' a  b  c ', NS_MAIN, 'en',
 218                                 new TitleValue( NS_MAIN, 'A_b_c' ) ),
 219                         array( ' _ Foo __ Bar_ _', NS_MAIN, 'en',
 220                                 new TitleValue( NS_MAIN, 'Foo_Bar' ) ),
 221
 222                         //NOTE: cases copied from TitleTest::testSecureAndSplit. Keep in sync.
 223                         array( 'Sandbox', NS_MAIN, 'en', ),
 224                         array( 'A "B"', NS_MAIN, 'en', ),
 225                         array( 'A \'B\'', NS_MAIN, 'en', ),
 226                         array( '.com', NS_MAIN, 'en', ),
 227                         array( '~', NS_MAIN, 'en', ),
 228                         array( '"', NS_MAIN, 'en', ),
 229                         array( '\'', NS_MAIN, 'en', ),
 230
 231                         array( 'Talk:Sandbox', NS_MAIN, 'en',
 232                                 new TitleValue( NS_TALK, 'Sandbox' ) ),
 233                         array( 'Talk:Foo:Sandbox', NS_MAIN, 'en',
 234                                 new TitleValue( NS_TALK, 'Foo:Sandbox' ) ),
 235                         array( 'File:Example.svg', NS_MAIN, 'en',
 236                                 new TitleValue( NS_FILE, 'Example.svg' ) ),
 237                         array( 'File_talk:Example.svg', NS_MAIN, 'en',
 238                                 new TitleValue( NS_FILE_TALK, 'Example.svg' ) ),
 239                         array( 'Foo/.../Sandbox', NS_MAIN, 'en',
 240                                 'Foo/.../Sandbox' ),
 241                         array( 'Sandbox/...', NS_MAIN, 'en',
 242                                 'Sandbox/...' ),
 243                         array( 'A~~', NS_MAIN, 'en',
 244                                 'A~~' ),
 245                         // Length is 256 total, but only title part matters
 246                         array( 'Category:' . str_repeat( 'x', 248 ), NS_MAIN, 'en',
 247                                 new TitleValue( NS_CATEGORY,
 248                                         'X' . str_repeat( 'x', 247 ) ) ),
 249                         array( str_repeat( 'x', 252 ), NS_MAIN, 'en',
 250                                 'X' . str_repeat( 'x', 251 ) )
 251                 );
 252         }
 253
 254         /**
 255          * @dataProvider provideParseTitle
 256          */
 257         public function testParseTitle( $text, $ns, $lang, $title = null ) {
 258                 if ( $title === null ) {
 259                         $title = str_replace( ' ', '_', trim( $text ) );
 260                 }
 261
 262                 if ( is_string( $title ) ) {
 263                         $title = new TitleValue( NS_MAIN, $title, '' );
 264                 }
 265
 266                 $codec = $this->makeCodec( $lang );
 267                 $actual = $codec->parseTitle( $text, $ns );
 268
 269                 $this->assertEquals( $title, $actual );
 270         }
 271
 272         public function provideParseTitle_invalid() {
 273                 //TODO: test unicode errors
 274
 275                 return array(
 276                         array( '#' ),
 277                         array( '::' ),
 278                         array( '::xx' ),
 279                         array( '::##' ),
 280                         array( ' :: x' ),
 281
 282                         array( 'Talk:File:Foo.jpg' ),
 283                         array( 'Talk:localtestiw:Foo' ),
 284                         array( 'remotetestiw:Foo' ),
 285                         array( '::1' ), // only valid in user namespace
 286                         array( 'User::x' ), // leading ":" in a user name is only valid of IPv6 addresses
 287
 288                         //NOTE: cases copied from TitleTest::testSecureAndSplit. Keep in sync.
 289                         array( '' ),
 290                         array( ':' ),
 291                         array( '__  __' ),
 292                         array( '  __  ' ),
 293                         // Bad characters forbidden regardless of wgLegalTitleChars
 294                         array( 'A [ B' ),
 295                         array( 'A ] B' ),
 296                         array( 'A { B' ),
 297                         array( 'A } B' ),
 298                         array( 'A < B' ),
 299                         array( 'A > B' ),
 300                         array( 'A | B' ),
 301                         // URL encoding
 302                         array( 'A%20B' ),
 303                         array( 'A%23B' ),
 304                         array( 'A%2523B' ),
 305                         // XML/HTML character entity references
 306                         // Note: Commented out because they are not marked invalid by the PHP test as
 307                         // Title::newFromText runs Sanitizer::decodeCharReferencesAndNormalize first.
 308                         //array( 'A &eacute; B' ),
 309                         //array( 'A &#233; B' ),
 310                         //array( 'A &#x00E9; B' ),
 311                         // Subject of NS_TALK does not roundtrip to NS_MAIN
 312                         array( 'Talk:File:Example.svg' ),
 313                         // Directory navigation
 314                         array( '.' ),
 315                         array( '..' ),
 316                         array( './Sandbox' ),
 317                         array( '../Sandbox' ),
 318                         array( 'Foo/./Sandbox' ),
 319                         array( 'Foo/../Sandbox' ),
 320                         array( 'Sandbox/.' ),
 321                         array( 'Sandbox/..' ),
 322                         // Tilde
 323                         array( 'A ~~~ Name' ),
 324                         array( 'A ~~~~ Signature' ),
 325                         array( 'A ~~~~~ Timestamp' ),
 326                         array( str_repeat( 'x', 256 ) ),
 327                         // Namespace prefix without actual title
 328                         array( 'Talk:' ),
 329                         array( 'Category: ' ),
 330                         array( 'Category: #bar' )
 331                 );
 332         }
 333
 334         /**
 335          * @dataProvider provideParseTitle_invalid
 336          */
 337         public function testParseTitle_invalid( $text ) {
 338                 $this->setExpectedException( 'MalformedTitleException' );
 339
 340                 $codec = $this->makeCodec( 'en' );
 341                 $codec->parseTitle( $text, NS_MAIN );
 342         }
 343
 344         public function provideGetNamespaceName() {
 345                 return array(
 346                         array( NS_MAIN, 'Foo', 'en', '' ),
 347                         array( NS_USER, 'Foo', 'en', 'User' ),
 348                         array( NS_USER, 'Hansi Maier', 'de', 'Benutzer' ),
 349
 350                         // getGenderCache() provides a mock that considers first
 351                         // names ending in "a" to be female.
 352                         array( NS_USER, 'Lisa Müller', 'de', 'Benutzerin' ),
 353                 );
 354         }
 355
 356         /**
 357          * @dataProvider provideGetNamespaceName
 358          *
 359          * @param $namespace
 360          * @param $text
 361          * @param $lang
 362          * @param $expected
 363          *
 364          * @internal param \TitleValue $title
 365          */
 366         public function testGetNamespaceName( $namespace, $text, $lang, $expected ) {
 367                 $codec = $this->makeCodec( $lang );
 368                 $name = $codec->getNamespaceName( $namespace, $text );
 369
 370                 $this->assertEquals( $expected, $name );
 371         }
 372 }