3 * Tests for UtfNormal::cleanUp() function.
5 * Copyright © 2004 Brion Vibber <brion@pobox.com>
6 * https://www.mediawiki.org/
8 * This program is free software; you can redistribute it and/or modify
9 * it under the terms of the GNU General Public License as published by
10 * the Free Software Foundation; either version 2 of the License, or
11 * (at your option) any later version.
13 * This program is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 * GNU General Public License for more details.
18 * You should have received a copy of the GNU General Public License along
19 * with this program; if not, write to the Free Software Foundation, Inc.,
20 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
21 * http://www.gnu.org/copyleft/gpl.html
27 * Additional tests for UtfNormal::cleanUp() function, inclusion
28 * regression checks for known problems.
34 * @todo covers tags, will be UtfNormal::cleanUp once the below is resolved
35 * @todo split me into test methods and providers per the below comment
36 * @todo Document individual tests
38 * We ignore code coverage for this test suite until they are rewritten
39 * to use data providers (bug 46561).
42 class CleanUpTest
extends PHPUnit_Framework_TestCase
{
43 public function testAscii() {
44 $text = 'This is plain ASCII text.';
45 $this->assertEquals( $text, UtfNormal
::cleanUp( $text ) );
48 public function testNull() {
49 $text = "a \x00 null";
50 $expect = "a \xef\xbf\xbd null";
53 bin2hex( UtfNormal
::cleanUp( $text ) ) );
56 public function testLatin() {
57 $text = "L'\xc3\xa9cole";
58 $this->assertEquals( $text, UtfNormal
::cleanUp( $text ) );
61 public function testLatinNormal() {
62 $text = "L'e\xcc\x81cole";
63 $expect = "L'\xc3\xa9cole";
64 $this->assertEquals( $expect, UtfNormal
::cleanUp( $text ) );
68 * This test is *very* expensive!
70 function XtestAllChars() {
71 $rep = UTF8_REPLACEMENT
;
72 for ( $i = 0x0; $i < UNICODE_MAX
; $i++
) {
73 $char = codepointToUtf8( $i );
74 $clean = UtfNormal
::cleanUp( $char );
75 $x = sprintf( "%04X", $i );
77 if ( $i %
0x1000 == 0 ) {
84 ( $i > 0x001f && $i < UNICODE_SURROGATE_FIRST
) ||
85 ( $i > UNICODE_SURROGATE_LAST
&& $i < 0xfffe ) ||
86 ( $i > 0xffff && $i <= UNICODE_MAX
)
88 if ( isset( UtfNormal
::$utfCanonicalComp[$char] )
89 ||
isset( UtfNormal
::$utfCanonicalDecomp[$char] )
91 $comp = UtfNormal
::NFC( $char );
95 "U+$x should be decomposed" );
100 "U+$x should be intact" );
103 $this->assertEquals( bin2hex( $rep ), bin2hex( $clean ), $x );
108 public static function provideAllBytes() {
118 * @dataProvider provideAllBytes
120 function testBytes( $head, $tail ) {
121 for ( $i = 0x0; $i < 256; $i++
) {
122 $char = $head . chr( $i ) . $tail;
123 $clean = UtfNormal
::cleanUp( $char );
124 $x = sprintf( "%02X", $i );
129 ( $i > 0x001f && $i < 0x80 )
134 "ASCII byte $x should be intact" );
135 if ( $char != $clean ) {
139 $norm = $head . UTF8_REPLACEMENT
. $tail;
143 "Forbidden byte $x should be rejected" );
144 if ( $norm != $clean ) {
152 * @dataProvider provideAllBytes
154 function testDoubleBytes( $head, $tail ) {
155 for ( $first = 0xc0; $first < 0x100; $first +
= 2 ) {
156 for ( $second = 0x80; $second < 0x100; $second +
= 2 ) {
157 $char = $head . chr( $first ) . chr( $second ) . $tail;
158 $clean = UtfNormal
::cleanUp( $char );
159 $x = sprintf( "%02X,%02X", $first, $second );
160 if ( $first > 0xc1 &&
164 $norm = UtfNormal
::NFC( $char );
168 "Pair $x should be intact" );
169 if ( $norm != $clean ) {
172 } elseif ( $first > 0xfd ||
$second > 0xbf ) {
173 # fe and ff are not legal head bytes -- expect two replacement chars
174 $norm = $head . UTF8_REPLACEMENT
. UTF8_REPLACEMENT
. $tail;
178 "Forbidden pair $x should be rejected" );
179 if ( $norm != $clean ) {
183 $norm = $head . UTF8_REPLACEMENT
. $tail;
187 "Forbidden pair $x should be rejected" );
188 if ( $norm != $clean ) {
197 * @dataProvider provideAllBytes
199 function testTripleBytes( $head, $tail ) {
200 for ( $first = 0xc0; $first < 0x100; $first +
= 2 ) {
201 for ( $second = 0x80; $second < 0x100; $second +
= 2 ) {
202 #for( $third = 0x80; $third < 0x100; $third++ ) {
203 for ( $third = 0x80; $third < 0x81; $third++
) {
204 $char = $head . chr( $first ) . chr( $second ) . chr( $third ) . $tail;
205 $clean = UtfNormal
::cleanUp( $char );
206 $x = sprintf( "%02X,%02X,%02X", $first, $second, $third );
208 if ( $first >= 0xe0 &&
213 if ( $first == 0xe0 && $second < 0xa0 ) {
215 bin2hex( $head . UTF8_REPLACEMENT
. $tail ),
217 "Overlong triplet $x should be rejected" );
218 } elseif ( $first == 0xed &&
219 ( chr( $first ) . chr( $second ) . chr( $third ) ) >= UTF8_SURROGATE_FIRST
222 bin2hex( $head . UTF8_REPLACEMENT
. $tail ),
224 "Surrogate triplet $x should be rejected" );
227 bin2hex( UtfNormal
::NFC( $char ) ),
229 "Triplet $x should be intact" );
231 } elseif ( $first > 0xc1 && $first < 0xe0 && $second < 0xc0 ) {
233 bin2hex( UtfNormal
::NFC( $head . chr( $first ) .
234 chr( $second ) ) . UTF8_REPLACEMENT
. $tail ),
236 "Valid 2-byte $x + broken tail" );
237 } elseif ( $second > 0xc1 && $second < 0xe0 && $third < 0xc0 ) {
239 bin2hex( $head . UTF8_REPLACEMENT
.
240 UtfNormal
::NFC( chr( $second ) . chr( $third ) . $tail ) ),
242 "Broken head + valid 2-byte $x" );
243 } elseif ( ( $first > 0xfd ||
$second > 0xfd ) &&
244 ( ( $second > 0xbf && $third > 0xbf ) ||
245 ( $second < 0xc0 && $third < 0xc0 ) ||
246 ( $second > 0xfd ) ||
249 # fe and ff are not legal head bytes -- expect three replacement chars
251 bin2hex( $head . UTF8_REPLACEMENT
. UTF8_REPLACEMENT
. UTF8_REPLACEMENT
. $tail ),
253 "Forbidden triplet $x should be rejected" );
254 } elseif ( $first > 0xc2 && $second < 0xc0 && $third < 0xc0 ) {
256 bin2hex( $head . UTF8_REPLACEMENT
. $tail ),
258 "Forbidden triplet $x should be rejected" );
261 bin2hex( $head . UTF8_REPLACEMENT
. UTF8_REPLACEMENT
. $tail ),
263 "Forbidden triplet $x should be rejected" );
270 public function testChunkRegression() {
271 # Check for regression against a chunking bug
272 $text = "\x46\x55\xb8" .
279 $expect = "\x46\x55\xef\xbf\xbd" .
289 bin2hex( UtfNormal
::cleanUp( $text ) ) );
292 public function testInterposeRegression() {
307 $expect = "\x4e\x30" .
323 bin2hex( UtfNormal
::cleanUp( $text ) ) );
326 public function testOverlongRegression() {
328 "\x1a" . # forbidden ascii
330 "\xc1\xa6" . # overlong sequence
332 "\x1c" . # forbidden ascii
347 bin2hex( UtfNormal
::cleanUp( $text ) ) );
350 public function testSurrogateRegression() {
351 $text = "\xed\xb4\x96" . # surrogate 0xDD16
355 $expect = "\xef\xbf\xbd" .
361 bin2hex( UtfNormal
::cleanUp( $text ) ) );
364 public function testBomRegression() {
365 $text = "\xef\xbf\xbe" . # U+FFFE, illegal char
369 $expect = "\xef\xbf\xbd" .
375 bin2hex( UtfNormal
::cleanUp( $text ) ) );
378 public function testForbiddenRegression() {
379 $text = "\xef\xbf\xbf"; # U+FFFF, illegal char
380 $expect = "\xef\xbf\xbd";
383 bin2hex( UtfNormal
::cleanUp( $text ) ) );
386 public function testHangulRegression() {
387 $text = "\xed\x9c\xaf" . # Hangul char
388 "\xe1\x87\x81"; # followed by another final jamo
389 $expect = $text; # Should *not* change.
392 bin2hex( UtfNormal
::cleanUp( $text ) ) );