Merge "Fix various docs and version numbers from 84a2f570"
[lhc/web/wiklou.git] / tests / phpunit / includes / normal / CleanUpTest.php
1 <?php
2 /**
3 * Tests for UtfNormal::cleanUp() function.
4 *
5 * Copyright © 2004 Brion Vibber <brion@pobox.com>
6 * http://www.mediawiki.org/
7 *
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.
12 *
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.
17 *
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
22 *
23 * @file
24 */
25
26 /**
27 * Additional tests for UtfNormal::cleanUp() function, inclusion
28 * regression checks for known problems.
29 * Requires PHPUnit.
30 *
31 * @ingroup UtfNormal
32 * @group Large
33 *
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 *
37 * We ignore code coverage for this test suite until they are rewritten
38 * to use data providers (bug 46561).
39 * @codeCoverageIgnore
40 */
41 class CleanUpTest extends MediaWikiTestCase {
42 /** @todo document */
43 public function testAscii() {
44 $text = 'This is plain ASCII text.';
45 $this->assertEquals( $text, UtfNormal::cleanUp( $text ) );
46 }
47
48 /** @todo document */
49 public function testNull() {
50 $text = "a \x00 null";
51 $expect = "a \xef\xbf\xbd null";
52 $this->assertEquals(
53 bin2hex( $expect ),
54 bin2hex( UtfNormal::cleanUp( $text ) ) );
55 }
56
57 /** @todo document */
58 public function testLatin() {
59 $text = "L'\xc3\xa9cole";
60 $this->assertEquals( $text, UtfNormal::cleanUp( $text ) );
61 }
62
63 /** @todo document */
64 public function testLatinNormal() {
65 $text = "L'e\xcc\x81cole";
66 $expect = "L'\xc3\xa9cole";
67 $this->assertEquals( $expect, UtfNormal::cleanUp( $text ) );
68 }
69
70 /**
71 * This test is *very* expensive!
72 * @todo document
73 */
74 function XtestAllChars() {
75 $rep = UTF8_REPLACEMENT;
76 for ( $i = 0x0; $i < UNICODE_MAX; $i++ ) {
77 $char = codepointToUtf8( $i );
78 $clean = UtfNormal::cleanUp( $char );
79 $x = sprintf( "%04X", $i );
80
81 if ( $i % 0x1000 == 0 ) {
82 echo "U+$x\n";
83 }
84
85 if ( $i == 0x0009 ||
86 $i == 0x000a ||
87 $i == 0x000d ||
88 ( $i > 0x001f && $i < UNICODE_SURROGATE_FIRST ) ||
89 ( $i > UNICODE_SURROGATE_LAST && $i < 0xfffe ) ||
90 ( $i > 0xffff && $i <= UNICODE_MAX )
91 ) {
92 if ( isset( UtfNormal::$utfCanonicalComp[$char] ) || isset( UtfNormal::$utfCanonicalDecomp[$char] ) ) {
93 $comp = UtfNormal::NFC( $char );
94 $this->assertEquals(
95 bin2hex( $comp ),
96 bin2hex( $clean ),
97 "U+$x should be decomposed" );
98 } else {
99 $this->assertEquals(
100 bin2hex( $char ),
101 bin2hex( $clean ),
102 "U+$x should be intact" );
103 }
104 } else {
105 $this->assertEquals( bin2hex( $rep ), bin2hex( $clean ), $x );
106 }
107 }
108 }
109
110 /** @todo document */
111 public function testAllBytes() {
112 $this->doTestBytes( '', '' );
113 $this->doTestBytes( 'x', '' );
114 $this->doTestBytes( '', 'x' );
115 $this->doTestBytes( 'x', 'x' );
116 }
117
118 /** @todo document */
119 function doTestBytes( $head, $tail ) {
120 for ( $i = 0x0; $i < 256; $i++ ) {
121 $char = $head . chr( $i ) . $tail;
122 $clean = UtfNormal::cleanUp( $char );
123 $x = sprintf( "%02X", $i );
124
125 if ( $i == 0x0009 ||
126 $i == 0x000a ||
127 $i == 0x000d ||
128 ( $i > 0x001f && $i < 0x80 )
129 ) {
130 $this->assertEquals(
131 bin2hex( $char ),
132 bin2hex( $clean ),
133 "ASCII byte $x should be intact" );
134 if ( $char != $clean ) {
135 return;
136 }
137 } else {
138 $norm = $head . UTF8_REPLACEMENT . $tail;
139 $this->assertEquals(
140 bin2hex( $norm ),
141 bin2hex( $clean ),
142 "Forbidden byte $x should be rejected" );
143 if ( $norm != $clean ) {
144 return;
145 }
146 }
147 }
148 }
149
150 /** @todo document */
151 public function testDoubleBytes() {
152 $this->doTestDoubleBytes( '', '' );
153 $this->doTestDoubleBytes( 'x', '' );
154 $this->doTestDoubleBytes( '', 'x' );
155 $this->doTestDoubleBytes( 'x', 'x' );
156 }
157
158 /**
159 * @todo document
160 */
161 function doTestDoubleBytes( $head, $tail ) {
162 for ( $first = 0xc0; $first < 0x100; $first += 2 ) {
163 for ( $second = 0x80; $second < 0x100; $second += 2 ) {
164 $char = $head . chr( $first ) . chr( $second ) . $tail;
165 $clean = UtfNormal::cleanUp( $char );
166 $x = sprintf( "%02X,%02X", $first, $second );
167 if ( $first > 0xc1 &&
168 $first < 0xe0 &&
169 $second < 0xc0
170 ) {
171 $norm = UtfNormal::NFC( $char );
172 $this->assertEquals(
173 bin2hex( $norm ),
174 bin2hex( $clean ),
175 "Pair $x should be intact" );
176 if ( $norm != $clean ) {
177 return;
178 }
179 } elseif ( $first > 0xfd || $second > 0xbf ) {
180 # fe and ff are not legal head bytes -- expect two replacement chars
181 $norm = $head . UTF8_REPLACEMENT . UTF8_REPLACEMENT . $tail;
182 $this->assertEquals(
183 bin2hex( $norm ),
184 bin2hex( $clean ),
185 "Forbidden pair $x should be rejected" );
186 if ( $norm != $clean ) {
187 return;
188 }
189 } else {
190 $norm = $head . UTF8_REPLACEMENT . $tail;
191 $this->assertEquals(
192 bin2hex( $norm ),
193 bin2hex( $clean ),
194 "Forbidden pair $x should be rejected" );
195 if ( $norm != $clean ) {
196 return;
197 }
198 }
199 }
200 }
201 }
202
203 /** @todo document */
204 public function testTripleBytes() {
205 $this->doTestTripleBytes( '', '' );
206 $this->doTestTripleBytes( 'x', '' );
207 $this->doTestTripleBytes( '', 'x' );
208 $this->doTestTripleBytes( 'x', 'x' );
209 }
210
211 /** @todo document */
212 function doTestTripleBytes( $head, $tail ) {
213 for ( $first = 0xc0; $first < 0x100; $first += 2 ) {
214 for ( $second = 0x80; $second < 0x100; $second += 2 ) {
215 #for( $third = 0x80; $third < 0x100; $third++ ) {
216 for ( $third = 0x80; $third < 0x81; $third++ ) {
217 $char = $head . chr( $first ) . chr( $second ) . chr( $third ) . $tail;
218 $clean = UtfNormal::cleanUp( $char );
219 $x = sprintf( "%02X,%02X,%02X", $first, $second, $third );
220
221 if ( $first >= 0xe0 &&
222 $first < 0xf0 &&
223 $second < 0xc0 &&
224 $third < 0xc0
225 ) {
226 if ( $first == 0xe0 && $second < 0xa0 ) {
227 $this->assertEquals(
228 bin2hex( $head . UTF8_REPLACEMENT . $tail ),
229 bin2hex( $clean ),
230 "Overlong triplet $x should be rejected" );
231 } elseif ( $first == 0xed &&
232 ( chr( $first ) . chr( $second ) . chr( $third ) ) >= UTF8_SURROGATE_FIRST
233 ) {
234 $this->assertEquals(
235 bin2hex( $head . UTF8_REPLACEMENT . $tail ),
236 bin2hex( $clean ),
237 "Surrogate triplet $x should be rejected" );
238 } else {
239 $this->assertEquals(
240 bin2hex( UtfNormal::NFC( $char ) ),
241 bin2hex( $clean ),
242 "Triplet $x should be intact" );
243 }
244 } elseif ( $first > 0xc1 && $first < 0xe0 && $second < 0xc0 ) {
245 $this->assertEquals(
246 bin2hex( UtfNormal::NFC( $head . chr( $first ) . chr( $second ) ) . UTF8_REPLACEMENT . $tail ),
247 bin2hex( $clean ),
248 "Valid 2-byte $x + broken tail" );
249 } elseif ( $second > 0xc1 && $second < 0xe0 && $third < 0xc0 ) {
250 $this->assertEquals(
251 bin2hex( $head . UTF8_REPLACEMENT . UtfNormal::NFC( chr( $second ) . chr( $third ) . $tail ) ),
252 bin2hex( $clean ),
253 "Broken head + valid 2-byte $x" );
254 } elseif ( ( $first > 0xfd || $second > 0xfd ) &&
255 ( ( $second > 0xbf && $third > 0xbf ) ||
256 ( $second < 0xc0 && $third < 0xc0 ) ||
257 ( $second > 0xfd ) ||
258 ( $third > 0xfd ) )
259 ) {
260 # fe and ff are not legal head bytes -- expect three replacement chars
261 $this->assertEquals(
262 bin2hex( $head . UTF8_REPLACEMENT . UTF8_REPLACEMENT . UTF8_REPLACEMENT . $tail ),
263 bin2hex( $clean ),
264 "Forbidden triplet $x should be rejected" );
265 } elseif ( $first > 0xc2 && $second < 0xc0 && $third < 0xc0 ) {
266 $this->assertEquals(
267 bin2hex( $head . UTF8_REPLACEMENT . $tail ),
268 bin2hex( $clean ),
269 "Forbidden triplet $x should be rejected" );
270 } else {
271 $this->assertEquals(
272 bin2hex( $head . UTF8_REPLACEMENT . UTF8_REPLACEMENT . $tail ),
273 bin2hex( $clean ),
274 "Forbidden triplet $x should be rejected" );
275 }
276 }
277 }
278 }
279 }
280
281 /** @todo document */
282 public function testChunkRegression() {
283 # Check for regression against a chunking bug
284 $text = "\x46\x55\xb8" .
285 "\xdc\x96" .
286 "\xee" .
287 "\xe7" .
288 "\x44" .
289 "\xaa" .
290 "\x2f\x25";
291 $expect = "\x46\x55\xef\xbf\xbd" .
292 "\xdc\x96" .
293 "\xef\xbf\xbd" .
294 "\xef\xbf\xbd" .
295 "\x44" .
296 "\xef\xbf\xbd" .
297 "\x2f\x25";
298
299 $this->assertEquals(
300 bin2hex( $expect ),
301 bin2hex( UtfNormal::cleanUp( $text ) ) );
302 }
303
304 /** @todo document */
305 public function testInterposeRegression() {
306 $text = "\x4e\x30" .
307 "\xb1" . # bad tail
308 "\x3a" .
309 "\x92" . # bad tail
310 "\x62\x3a" .
311 "\x84" . # bad tail
312 "\x43" .
313 "\xc6" . # bad head
314 "\x3f" .
315 "\x92" . # bad tail
316 "\xad" . # bad tail
317 "\x7d" .
318 "\xd9\x95";
319
320 $expect = "\x4e\x30" .
321 "\xef\xbf\xbd" .
322 "\x3a" .
323 "\xef\xbf\xbd" .
324 "\x62\x3a" .
325 "\xef\xbf\xbd" .
326 "\x43" .
327 "\xef\xbf\xbd" .
328 "\x3f" .
329 "\xef\xbf\xbd" .
330 "\xef\xbf\xbd" .
331 "\x7d" .
332 "\xd9\x95";
333
334 $this->assertEquals(
335 bin2hex( $expect ),
336 bin2hex( UtfNormal::cleanUp( $text ) ) );
337 }
338
339 /** @todo document */
340 public function testOverlongRegression() {
341 $text = "\x67" .
342 "\x1a" . # forbidden ascii
343 "\xea" . # bad head
344 "\xc1\xa6" . # overlong sequence
345 "\xad" . # bad tail
346 "\x1c" . # forbidden ascii
347 "\xb0" . # bad tail
348 "\x3c" .
349 "\x9e"; # bad tail
350 $expect = "\x67" .
351 "\xef\xbf\xbd" .
352 "\xef\xbf\xbd" .
353 "\xef\xbf\xbd" .
354 "\xef\xbf\xbd" .
355 "\xef\xbf\xbd" .
356 "\xef\xbf\xbd" .
357 "\x3c" .
358 "\xef\xbf\xbd";
359 $this->assertEquals(
360 bin2hex( $expect ),
361 bin2hex( UtfNormal::cleanUp( $text ) ) );
362 }
363
364 /** @todo document */
365 public function testSurrogateRegression() {
366 $text = "\xed\xb4\x96" . # surrogate 0xDD16
367 "\x83" . # bad tail
368 "\xb4" . # bad tail
369 "\xac"; # bad head
370 $expect = "\xef\xbf\xbd" .
371 "\xef\xbf\xbd" .
372 "\xef\xbf\xbd" .
373 "\xef\xbf\xbd";
374 $this->assertEquals(
375 bin2hex( $expect ),
376 bin2hex( UtfNormal::cleanUp( $text ) ) );
377 }
378
379 /** @todo document */
380 public function testBomRegression() {
381 $text = "\xef\xbf\xbe" . # U+FFFE, illegal char
382 "\xb2" . # bad tail
383 "\xef" . # bad head
384 "\x59";
385 $expect = "\xef\xbf\xbd" .
386 "\xef\xbf\xbd" .
387 "\xef\xbf\xbd" .
388 "\x59";
389 $this->assertEquals(
390 bin2hex( $expect ),
391 bin2hex( UtfNormal::cleanUp( $text ) ) );
392 }
393
394 /** @todo document */
395 public function testForbiddenRegression() {
396 $text = "\xef\xbf\xbf"; # U+FFFF, illegal char
397 $expect = "\xef\xbf\xbd";
398 $this->assertEquals(
399 bin2hex( $expect ),
400 bin2hex( UtfNormal::cleanUp( $text ) ) );
401 }
402
403 /** @todo document */
404 public function testHangulRegression() {
405 $text = "\xed\x9c\xaf" . # Hangul char
406 "\xe1\x87\x81"; # followed by another final jamo
407 $expect = $text; # Should *not* change.
408 $this->assertEquals(
409 bin2hex( $expect ),
410 bin2hex( UtfNormal::cleanUp( $text ) ) );
411 }
412 }