1 /*
2 * Copyright © 2011 Codethink Limited
3 * Copyright © 2011 Google, Inc.
4 *
5 * This is part of HarfBuzz, a text shaping library.
6 *
7 * Permission is hereby granted, without written agreement and without
8 * license or royalty fees, to use, copy, modify, and distribute this
9 * software and its documentation for any purpose, provided that the
10 * above copyright notice and the following two paragraphs appear in
11 * all copies of this software.
12 *
13 * IN NO EVENT SHALL THE COPYRIGHT HOLDER BE LIABLE TO ANY PARTY FOR
14 * DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES
15 * ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN
16 * IF THE COPYRIGHT HOLDER HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
17 * DAMAGE.
18 *
19 * THE COPYRIGHT HOLDER SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING,
20 * BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
21 * FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS
22 * ON AN "AS IS" BASIS, AND THE COPYRIGHT HOLDER HAS NO OBLIGATION TO
23 * PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.
24 *
25 * Codethink Author(s): Ryan Lortie
26 * Google Author(s): Behdad Esfahbod
27 */
28
29 #include "hb-test.h"
30
31 /* Unit tests for hb-unicode.h */
32 /* Unit tests for hb-glib.h */
33 /* Unit tests for hb-icu.h */
34
35
36 #ifdef HAVE_GLIB
37 #include <hb-glib.h>
38 #endif
39 #ifdef HAVE_ICU
40 #include <hb-icu.h>
41 #endif
42
43
44 /* Some useful stuff */
45
46 #define MAGIC0 0x12345678
47 #define MAGIC1 0x76543210
48
49 typedef struct {
50 int value;
51 gboolean freed;
52 } data_t;
53
free_up(void * p)54 static void free_up (void *p)
55 {
56 data_t *data = (data_t *) p;
57
58 g_assert (data->value == MAGIC0 || data->value == MAGIC1);
59 g_assert (!data->freed);
60 data->freed = TRUE;
61 }
62
63 static hb_script_t
simple_get_script(hb_unicode_funcs_t * ufuncs,hb_codepoint_t codepoint,void * user_data)64 simple_get_script (hb_unicode_funcs_t *ufuncs,
65 hb_codepoint_t codepoint,
66 void *user_data)
67 {
68 data_t *data = (data_t *) user_data;
69
70 g_assert (hb_unicode_funcs_get_parent (ufuncs) != NULL);
71 g_assert_cmphex (data->value, ==, MAGIC0);
72 g_assert (!data->freed);
73
74 if ('a' <= codepoint && codepoint <= 'z')
75 return HB_SCRIPT_LATIN;
76 else
77 return HB_SCRIPT_UNKNOWN;
78 }
79
80 static hb_script_t
a_is_for_arabic_get_script(hb_unicode_funcs_t * ufuncs,hb_codepoint_t codepoint,void * user_data)81 a_is_for_arabic_get_script (hb_unicode_funcs_t *ufuncs,
82 hb_codepoint_t codepoint,
83 void *user_data)
84 {
85 data_t *data = (data_t *) user_data;
86
87 g_assert (hb_unicode_funcs_get_parent (ufuncs) != NULL);
88 g_assert_cmphex (data->value, ==, MAGIC1);
89 g_assert (!data->freed);
90
91 if (codepoint == 'a') {
92 return HB_SCRIPT_ARABIC;
93 } else {
94 hb_unicode_funcs_t *parent = hb_unicode_funcs_get_parent (ufuncs);
95
96 return hb_unicode_script (parent, codepoint);
97 }
98 }
99
100
101
102 /* Check all properties */
103
104 /* Some of the following tables where adapted from glib/glib/tests/utf8-misc.c.
105 * The license is compatible. */
106
107 typedef struct {
108 hb_codepoint_t unicode;
109 unsigned int value;
110 } test_pair_t;
111
112 static const test_pair_t combining_class_tests[] =
113 {
114 { 0x0020, 0 },
115 { 0x0334, 1 },
116 { 0x093C, 7 },
117 { 0x3099, 8 },
118 { 0x094D, 9 },
119 { 0x05B0, 10 },
120 { 0x05B1, 11 },
121 { 0x05B2, 12 },
122 { 0x05B3, 13 },
123 { 0x05B4, 14 },
124 { 0x05B5, 15 },
125 { 0x05B6, 16 },
126 { 0x05B7, 17 },
127 { 0x05B8, 18 },
128 { 0x05B9, 19 },
129 { 0x05BB, 20 },
130 { 0x05BC, 21 },
131 { 0x05BD, 22 },
132 { 0x05BF, 23 },
133 { 0x05C1, 24 },
134 { 0x05C2, 25 },
135 { 0xFB1E, 26 },
136 { 0x064B, 27 },
137 { 0x064C, 28 },
138 { 0x064D, 29 },
139 /* ... */
140 { 0x05AE, 228 },
141 { 0x0300, 230 },
142 { 0x302C, 232 },
143 { 0x0362, 233 },
144 { 0x0360, 234 },
145 { 0x0345, 240 },
146
147 { 0x111111, 0 }
148 };
149 static const test_pair_t combining_class_tests_more[] =
150 {
151 /* Unicode-5.1 character additions */
152 { 0x1DCD, 234 },
153
154 /* Unicode-5.2 character additions */
155 { 0xA8E0, 230 },
156
157 /* Unicode-6.0 character additions */
158 { 0x135D, 230 },
159
160 { 0x111111, 0 }
161 };
162
163
164 static const test_pair_t general_category_tests[] =
165 {
166 { 0x000D, HB_UNICODE_GENERAL_CATEGORY_CONTROL },
167 { 0x200E, HB_UNICODE_GENERAL_CATEGORY_FORMAT },
168 { 0x0378, HB_UNICODE_GENERAL_CATEGORY_UNASSIGNED },
169 { 0xE000, HB_UNICODE_GENERAL_CATEGORY_PRIVATE_USE },
170 { 0xD800, HB_UNICODE_GENERAL_CATEGORY_SURROGATE },
171 { 0x0061, HB_UNICODE_GENERAL_CATEGORY_LOWERCASE_LETTER },
172 { 0x02B0, HB_UNICODE_GENERAL_CATEGORY_MODIFIER_LETTER },
173 { 0x3400, HB_UNICODE_GENERAL_CATEGORY_OTHER_LETTER },
174 { 0x01C5, HB_UNICODE_GENERAL_CATEGORY_TITLECASE_LETTER },
175 { 0xFF21, HB_UNICODE_GENERAL_CATEGORY_UPPERCASE_LETTER },
176 { 0x0903, HB_UNICODE_GENERAL_CATEGORY_SPACING_MARK },
177 { 0x20DD, HB_UNICODE_GENERAL_CATEGORY_ENCLOSING_MARK },
178 { 0xA806, HB_UNICODE_GENERAL_CATEGORY_NON_SPACING_MARK },
179 { 0xFF10, HB_UNICODE_GENERAL_CATEGORY_DECIMAL_NUMBER },
180 { 0x16EE, HB_UNICODE_GENERAL_CATEGORY_LETTER_NUMBER },
181 { 0x17F0, HB_UNICODE_GENERAL_CATEGORY_OTHER_NUMBER },
182 { 0x005F, HB_UNICODE_GENERAL_CATEGORY_CONNECT_PUNCTUATION },
183 { 0x058A, HB_UNICODE_GENERAL_CATEGORY_DASH_PUNCTUATION },
184 { 0x0F3B, HB_UNICODE_GENERAL_CATEGORY_CLOSE_PUNCTUATION },
185 { 0x2019, HB_UNICODE_GENERAL_CATEGORY_FINAL_PUNCTUATION },
186 { 0x2018, HB_UNICODE_GENERAL_CATEGORY_INITIAL_PUNCTUATION },
187 { 0x2016, HB_UNICODE_GENERAL_CATEGORY_OTHER_PUNCTUATION },
188 { 0x0F3A, HB_UNICODE_GENERAL_CATEGORY_OPEN_PUNCTUATION },
189 { 0x20A0, HB_UNICODE_GENERAL_CATEGORY_CURRENCY_SYMBOL },
190 { 0x309B, HB_UNICODE_GENERAL_CATEGORY_MODIFIER_SYMBOL },
191 { 0xFB29, HB_UNICODE_GENERAL_CATEGORY_MATH_SYMBOL },
192 { 0x00A6, HB_UNICODE_GENERAL_CATEGORY_OTHER_SYMBOL },
193 { 0x2028, HB_UNICODE_GENERAL_CATEGORY_LINE_SEPARATOR },
194 { 0x2029, HB_UNICODE_GENERAL_CATEGORY_PARAGRAPH_SEPARATOR },
195 { 0x202F, HB_UNICODE_GENERAL_CATEGORY_SPACE_SEPARATOR },
196
197 { 0x111111, HB_UNICODE_GENERAL_CATEGORY_UNASSIGNED }
198 };
199 static const test_pair_t general_category_tests_more[] =
200 {
201 /* Unicode-5.2 character additions */
202 { 0x1F131, HB_UNICODE_GENERAL_CATEGORY_OTHER_SYMBOL },
203
204 /* Unicode-6.0 character additions */
205 { 0x0620, HB_UNICODE_GENERAL_CATEGORY_OTHER_LETTER },
206
207 { 0x111111, HB_UNICODE_GENERAL_CATEGORY_UNASSIGNED }
208 };
209
210 static const test_pair_t mirroring_tests[] =
211 {
212 /* Some characters that do NOT mirror */
213 { 0x0020, 0x0020 },
214 { 0x0041, 0x0041 },
215 { 0x00F0, 0x00F0 },
216 { 0x27CC, 0x27CC },
217 { 0xE01EF, 0xE01EF },
218 { 0x1D7C3, 0x1D7C3 },
219 { 0x100000, 0x100000 },
220
221 /* Some characters that do mirror */
222 { 0x0029, 0x0028 },
223 { 0x0028, 0x0029 },
224 { 0x003E, 0x003C },
225 { 0x003C, 0x003E },
226 { 0x005D, 0x005B },
227 { 0x005B, 0x005D },
228 { 0x007D, 0x007B },
229 { 0x007B, 0x007D },
230 { 0x00BB, 0x00AB },
231 { 0x00AB, 0x00BB },
232 { 0x226B, 0x226A },
233 { 0x226A, 0x226B },
234 { 0x22F1, 0x22F0 },
235 { 0x22F0, 0x22F1 },
236 { 0xFF60, 0xFF5F },
237 { 0xFF5F, 0xFF60 },
238 { 0xFF63, 0xFF62 },
239 { 0xFF62, 0xFF63 },
240
241 { 0x111111, 0x111111 },
242 };
243 static const test_pair_t mirroring_tests_more[] =
244 {
245 /* No new mirroring characters have been encoded in recent Unicode versions. */
246 { 0x111111, 0x111111 }
247 };
248
249 static const test_pair_t script_tests[] =
250 {
251 { 0x002A, HB_SCRIPT_COMMON },
252 { 0x0670, HB_SCRIPT_INHERITED },
253 { 0x060D, HB_SCRIPT_ARABIC },
254 { 0x0559, HB_SCRIPT_ARMENIAN },
255 { 0x09CD, HB_SCRIPT_BENGALI },
256 { 0x31B6, HB_SCRIPT_BOPOMOFO },
257 { 0x13A2, HB_SCRIPT_CHEROKEE },
258 { 0x2CFD, HB_SCRIPT_COPTIC },
259 { 0x0482, HB_SCRIPT_CYRILLIC },
260 { 0x10401, HB_SCRIPT_DESERET },
261 { 0x094D, HB_SCRIPT_DEVANAGARI },
262 { 0x1258, HB_SCRIPT_ETHIOPIC },
263 { 0x10FC, HB_SCRIPT_GEORGIAN },
264 { 0x10341, HB_SCRIPT_GOTHIC },
265 { 0x0375, HB_SCRIPT_GREEK },
266 { 0x0A83, HB_SCRIPT_GUJARATI },
267 { 0x0A3C, HB_SCRIPT_GURMUKHI },
268 { 0x3005, HB_SCRIPT_HAN },
269 { 0x1100, HB_SCRIPT_HANGUL },
270 { 0x05BF, HB_SCRIPT_HEBREW },
271 { 0x309F, HB_SCRIPT_HIRAGANA },
272 { 0x0CBC, HB_SCRIPT_KANNADA },
273 { 0x30FF, HB_SCRIPT_KATAKANA },
274 { 0x17DD, HB_SCRIPT_KHMER },
275 { 0x0EDD, HB_SCRIPT_LAO },
276 { 0x0061, HB_SCRIPT_LATIN },
277 { 0x0D3D, HB_SCRIPT_MALAYALAM },
278 { 0x1843, HB_SCRIPT_MONGOLIAN },
279 { 0x1031, HB_SCRIPT_MYANMAR },
280 { 0x169C, HB_SCRIPT_OGHAM },
281 { 0x10322, HB_SCRIPT_OLD_ITALIC },
282 { 0x0B3C, HB_SCRIPT_ORIYA },
283 { 0x16EF, HB_SCRIPT_RUNIC },
284 { 0x0DBD, HB_SCRIPT_SINHALA },
285 { 0x0711, HB_SCRIPT_SYRIAC },
286 { 0x0B82, HB_SCRIPT_TAMIL },
287 { 0x0C03, HB_SCRIPT_TELUGU },
288 { 0x07B1, HB_SCRIPT_THAANA },
289 { 0x0E31, HB_SCRIPT_THAI },
290 { 0x0FD4, HB_SCRIPT_TIBETAN },
291 { 0x1401, HB_SCRIPT_CANADIAN_SYLLABICS },
292 { 0xA015, HB_SCRIPT_YI },
293 { 0x1700, HB_SCRIPT_TAGALOG },
294 { 0x1720, HB_SCRIPT_HANUNOO },
295 { 0x1740, HB_SCRIPT_BUHID },
296 { 0x1760, HB_SCRIPT_TAGBANWA },
297
298 /* Unicode-4.0 additions */
299 { 0x2800, HB_SCRIPT_BRAILLE },
300 { 0x10808, HB_SCRIPT_CYPRIOT },
301 { 0x1932, HB_SCRIPT_LIMBU },
302 { 0x10480, HB_SCRIPT_OSMANYA },
303 { 0x10450, HB_SCRIPT_SHAVIAN },
304 { 0x10000, HB_SCRIPT_LINEAR_B },
305 { 0x1950, HB_SCRIPT_TAI_LE },
306 { 0x1039F, HB_SCRIPT_UGARITIC },
307
308 /* Unicode-4.1 additions */
309 { 0x1980, HB_SCRIPT_NEW_TAI_LUE },
310 { 0x1A1F, HB_SCRIPT_BUGINESE },
311 { 0x2C00, HB_SCRIPT_GLAGOLITIC },
312 { 0x2D6F, HB_SCRIPT_TIFINAGH },
313 { 0xA800, HB_SCRIPT_SYLOTI_NAGRI },
314 { 0x103D0, HB_SCRIPT_OLD_PERSIAN },
315 { 0x10A3F, HB_SCRIPT_KHAROSHTHI },
316
317 /* Unicode-5.0 additions */
318 { 0x0378, HB_SCRIPT_UNKNOWN },
319 { 0x1B04, HB_SCRIPT_BALINESE },
320 { 0x12000, HB_SCRIPT_CUNEIFORM },
321 { 0x10900, HB_SCRIPT_PHOENICIAN },
322 { 0xA840, HB_SCRIPT_PHAGS_PA },
323 { 0x07C0, HB_SCRIPT_NKO },
324
325 /* Unicode-5.1 additions */
326 { 0xA900, HB_SCRIPT_KAYAH_LI },
327 { 0x1C00, HB_SCRIPT_LEPCHA },
328 { 0xA930, HB_SCRIPT_REJANG },
329 { 0x1B80, HB_SCRIPT_SUNDANESE },
330 { 0xA880, HB_SCRIPT_SAURASHTRA },
331 { 0xAA00, HB_SCRIPT_CHAM },
332 { 0x1C50, HB_SCRIPT_OL_CHIKI },
333 { 0xA500, HB_SCRIPT_VAI },
334 { 0x102A0, HB_SCRIPT_CARIAN },
335 { 0x10280, HB_SCRIPT_LYCIAN },
336 { 0x1093F, HB_SCRIPT_LYDIAN },
337
338 { 0x111111, HB_SCRIPT_UNKNOWN }
339 };
340 static const test_pair_t script_tests_more[] =
341 {
342 /* Unicode-5.2 additions */
343 { 0x10B00, HB_SCRIPT_AVESTAN },
344 { 0xA6A0, HB_SCRIPT_BAMUM },
345 { 0x13000, HB_SCRIPT_EGYPTIAN_HIEROGLYPHS },
346 { 0x10840, HB_SCRIPT_IMPERIAL_ARAMAIC },
347 { 0x10B60, HB_SCRIPT_INSCRIPTIONAL_PAHLAVI },
348 { 0x10B40, HB_SCRIPT_INSCRIPTIONAL_PARTHIAN },
349 { 0xA980, HB_SCRIPT_JAVANESE },
350 { 0x11082, HB_SCRIPT_KAITHI },
351 { 0xA4D0, HB_SCRIPT_LISU },
352 { 0xABE5, HB_SCRIPT_MEETEI_MAYEK },
353 { 0x10A60, HB_SCRIPT_OLD_SOUTH_ARABIAN },
354 { 0x10C00, HB_SCRIPT_OLD_TURKIC },
355 { 0x0800, HB_SCRIPT_SAMARITAN },
356 { 0x1A20, HB_SCRIPT_TAI_THAM },
357 { 0xAA80, HB_SCRIPT_TAI_VIET },
358
359 /* Unicode-6.0 additions */
360 { 0x1BC0, HB_SCRIPT_BATAK },
361 { 0x11000, HB_SCRIPT_BRAHMI },
362 { 0x0840, HB_SCRIPT_MANDAIC },
363
364 /* Unicode-5.2 character additions */
365 { 0x1CED, HB_SCRIPT_INHERITED },
366 { 0x1400, HB_SCRIPT_CANADIAN_ABORIGINAL },
367
368 { 0x111111, HB_SCRIPT_UNKNOWN }
369 };
370
371
372 typedef unsigned int (*get_func_t) (hb_unicode_funcs_t *ufuncs,
373 hb_codepoint_t unicode,
374 void *user_data);
375 typedef unsigned int (*func_setter_func_t) (hb_unicode_funcs_t *ufuncs,
376 get_func_t func,
377 void *user_data,
378 hb_destroy_func_t destroy);
379 typedef unsigned int (*getter_func_t) (hb_unicode_funcs_t *ufuncs,
380 hb_codepoint_t unicode);
381
382 typedef struct {
383 const char *name;
384 func_setter_func_t func_setter;
385 getter_func_t getter;
386 const test_pair_t *tests;
387 unsigned int num_tests;
388 const test_pair_t *tests_more;
389 unsigned int num_tests_more;
390 unsigned int default_value;
391 } property_t;
392
393 #define RETURNS_UNICODE_ITSELF ((unsigned int) -1)
394
395 #define PROPERTY(name, DEFAULT) \
396 { \
397 #name, \
398 (func_setter_func_t) hb_unicode_funcs_set_##name##_func, \
399 (getter_func_t) hb_unicode_##name, \
400 name##_tests, \
401 G_N_ELEMENTS (name##_tests), \
402 name##_tests_more, \
403 G_N_ELEMENTS (name##_tests_more), \
404 DEFAULT \
405 }
406 static const property_t properties[] =
407 {
408 PROPERTY (combining_class, 0),
409 PROPERTY (general_category, (unsigned int) HB_UNICODE_GENERAL_CATEGORY_OTHER_LETTER),
410 PROPERTY (mirroring, RETURNS_UNICODE_ITSELF),
411 PROPERTY (script, (unsigned int) HB_SCRIPT_UNKNOWN)
412 };
413 #undef PROPERTY
414
415 static void
test_unicode_properties(gconstpointer user_data)416 test_unicode_properties (gconstpointer user_data)
417 {
418 hb_unicode_funcs_t *uf = (hb_unicode_funcs_t *) user_data;
419 unsigned int i, j;
420 gboolean failed = TRUE;
421
422 g_assert (hb_unicode_funcs_is_immutable (uf));
423 g_assert (hb_unicode_funcs_get_parent (uf));
424
425 for (i = 0; i < G_N_ELEMENTS (properties); i++) {
426 const property_t *p = &properties[i];
427 const test_pair_t *tests;
428
429 g_test_message ("Testing property %s", p->name);
430 tests = p->tests;
431 for (j = 0; j < p->num_tests; j++) {
432 g_test_message ("Test %s #%d: U+%04X", p->name, j, tests[j].unicode);
433 g_assert_cmphex (p->getter (uf, tests[j].unicode), ==, tests[j].value);
434 }
435 /* These tests are from Unicode 5.2 onward and older glib/ICU
436 * don't get them right. Just warn instead of assert. */
437 tests = p->tests_more;
438 for (j = 0; j < p->num_tests_more; j++) {
439 g_test_message ("Test %s more #%d: U+%04X", p->name, j, tests[j].unicode);
440 if (p->getter (uf, tests[j].unicode) != tests[j].value) {
441 g_test_message ("Soft fail: Received %x, expected %x", p->getter (uf, tests[j].unicode), tests[j].value);
442 failed = TRUE;
443 }
444 }
445 }
446
447 if (failed)
448 g_test_message ("Some property tests failed. You probably have an old version of one of the libraries used.");
449 }
450
451 static hb_codepoint_t
default_value(hb_codepoint_t _default_value,hb_codepoint_t unicode)452 default_value (hb_codepoint_t _default_value, hb_codepoint_t unicode)
453 {
454 return _default_value == RETURNS_UNICODE_ITSELF ? unicode : _default_value;
455 }
456
457 static void
_test_unicode_properties_nil(hb_unicode_funcs_t * uf)458 _test_unicode_properties_nil (hb_unicode_funcs_t *uf)
459 {
460 unsigned int i, j;
461
462 for (i = 0; i < G_N_ELEMENTS (properties); i++) {
463 const property_t *p = &properties[i];
464 const test_pair_t *tests;
465
466 g_test_message ("Testing property %s", p->name);
467 tests = p->tests;
468 for (j = 0; j < p->num_tests; j++) {
469 g_test_message ("Test %s #%d: U+%04X", p->name, j, tests[j].unicode);
470 g_assert_cmphex (p->getter (uf, tests[j].unicode), ==, default_value (p->default_value, tests[j].unicode));
471 }
472 tests = p->tests_more;
473 for (j = 0; j < p->num_tests_more; j++) {
474 g_test_message ("Test %s more #%d: U+%04X", p->name, j, tests[j].unicode);
475 g_assert_cmphex (p->getter (uf, tests[j].unicode), ==, default_value (p->default_value, tests[j].unicode));
476 }
477 }
478 }
479
480 static void
test_unicode_properties_nil(void)481 test_unicode_properties_nil (void)
482 {
483 hb_unicode_funcs_t *uf = hb_unicode_funcs_create (NULL);
484
485 g_assert (!hb_unicode_funcs_is_immutable (uf));
486 _test_unicode_properties_nil (uf);
487
488 hb_unicode_funcs_destroy (uf);
489 }
490
491 static void
test_unicode_properties_empty(void)492 test_unicode_properties_empty (void)
493 {
494 hb_unicode_funcs_t *uf = hb_unicode_funcs_get_empty ();
495
496 g_assert (uf);
497 g_assert (hb_unicode_funcs_is_immutable (uf));
498 _test_unicode_properties_nil (uf);
499 }
500
501
502 static void
test_unicode_chainup(void)503 test_unicode_chainup (void)
504 {
505 hb_unicode_funcs_t *uf, *uf2;
506
507 /* Chain-up to nil */
508
509 uf = hb_unicode_funcs_create (NULL);
510 g_assert (!hb_unicode_funcs_is_immutable (uf));
511
512 uf2 = hb_unicode_funcs_create (uf);
513 g_assert (hb_unicode_funcs_is_immutable (uf));
514 hb_unicode_funcs_destroy (uf);
515
516 g_assert (!hb_unicode_funcs_is_immutable (uf2));
517 _test_unicode_properties_nil (uf2);
518
519 hb_unicode_funcs_destroy (uf2);
520
521 /* Chain-up to default */
522
523 uf = hb_unicode_funcs_create (hb_unicode_funcs_get_default ());
524 g_assert (!hb_unicode_funcs_is_immutable (uf));
525
526 uf2 = hb_unicode_funcs_create (uf);
527 g_assert (hb_unicode_funcs_is_immutable (uf));
528 hb_unicode_funcs_destroy (uf);
529
530 g_assert (!hb_unicode_funcs_is_immutable (uf2));
531 hb_unicode_funcs_make_immutable (uf2);
532 test_unicode_properties (uf2);
533
534 hb_unicode_funcs_destroy (uf2);
535
536 }
537
538 static void
test_unicode_setters(void)539 test_unicode_setters (void)
540 {
541 hb_unicode_funcs_t *uf;
542 unsigned int i;
543
544 /* This is cruel: we use script-returning functions to test all properties,
545 * but it works. */
546
547 for (i = 0; i < G_N_ELEMENTS (properties); i++) {
548 const property_t *p = &properties[i];
549 data_t data[2] = {{MAGIC0, FALSE}, {MAGIC1, FALSE}};
550
551 g_test_message ("Testing property %s", p->name);
552
553 uf = hb_unicode_funcs_create (NULL);
554 g_assert (!hb_unicode_funcs_is_immutable (uf));
555
556 p->func_setter (uf, (get_func_t) simple_get_script, &data[0], free_up);
557
558 g_assert_cmphex (p->getter (uf, 'a'), ==, HB_SCRIPT_LATIN);
559 g_assert_cmphex (p->getter (uf, '0'), ==, HB_SCRIPT_UNKNOWN);
560
561 p->func_setter (uf, (get_func_t) NULL, NULL, NULL);
562 g_assert (data[0].freed && !data[1].freed);
563
564 g_assert (!hb_unicode_funcs_is_immutable (uf));
565 hb_unicode_funcs_make_immutable (uf);
566 g_assert (hb_unicode_funcs_is_immutable (uf));
567
568 /* Since uf is immutable now, the following setter should do nothing. */
569 p->func_setter (uf, (get_func_t) a_is_for_arabic_get_script, &data[1], free_up);
570
571 g_assert (data[0].freed && !data[1].freed);
572 hb_unicode_funcs_destroy (uf);
573 g_assert (data[0].freed && !data[1].freed);
574 }
575 }
576
577
578
579 typedef struct {
580 data_t data[2];
581 } data_fixture_t;
582
583 static void
data_fixture_init(data_fixture_t * f,gconstpointer user_data HB_UNUSED)584 data_fixture_init (data_fixture_t *f, gconstpointer user_data HB_UNUSED)
585 {
586 f->data[0].value = MAGIC0;
587 f->data[1].value = MAGIC1;
588 }
589 static void
data_fixture_finish(data_fixture_t * f HB_UNUSED,gconstpointer user_data HB_UNUSED)590 data_fixture_finish (data_fixture_t *f HB_UNUSED, gconstpointer user_data HB_UNUSED)
591 {
592 }
593
594 static void
test_unicode_subclassing_nil(data_fixture_t * f,gconstpointer user_data HB_UNUSED)595 test_unicode_subclassing_nil (data_fixture_t *f, gconstpointer user_data HB_UNUSED)
596 {
597 hb_unicode_funcs_t *uf, *aa;
598
599 uf = hb_unicode_funcs_create (NULL);
600
601 aa = hb_unicode_funcs_create (uf);
602
603 hb_unicode_funcs_destroy (uf);
604
605 hb_unicode_funcs_set_script_func (aa, a_is_for_arabic_get_script,
606 &f->data[1], free_up);
607
608 g_assert_cmphex (hb_unicode_script (aa, 'a'), ==, HB_SCRIPT_ARABIC);
609 g_assert_cmphex (hb_unicode_script (aa, 'b'), ==, HB_SCRIPT_UNKNOWN);
610
611 g_assert (!f->data[0].freed && !f->data[1].freed);
612 hb_unicode_funcs_destroy (aa);
613 g_assert (!f->data[0].freed && f->data[1].freed);
614 }
615
616 static void
test_unicode_subclassing_default(data_fixture_t * f,gconstpointer user_data HB_UNUSED)617 test_unicode_subclassing_default (data_fixture_t *f, gconstpointer user_data HB_UNUSED)
618 {
619 hb_unicode_funcs_t *uf, *aa;
620
621 uf = hb_unicode_funcs_get_default ();
622 aa = hb_unicode_funcs_create (uf);
623
624 hb_unicode_funcs_set_script_func (aa, a_is_for_arabic_get_script,
625 &f->data[1], free_up);
626
627 g_assert_cmphex (hb_unicode_script (aa, 'a'), ==, HB_SCRIPT_ARABIC);
628 g_assert_cmphex (hb_unicode_script (aa, 'b'), ==, HB_SCRIPT_LATIN);
629
630 g_assert (!f->data[0].freed && !f->data[1].freed);
631 hb_unicode_funcs_destroy (aa);
632 g_assert (!f->data[0].freed && f->data[1].freed);
633 }
634
635 static void
test_unicode_subclassing_deep(data_fixture_t * f,gconstpointer user_data HB_UNUSED)636 test_unicode_subclassing_deep (data_fixture_t *f, gconstpointer user_data HB_UNUSED)
637 {
638 hb_unicode_funcs_t *uf, *aa;
639
640 uf = hb_unicode_funcs_create (NULL);
641
642 hb_unicode_funcs_set_script_func (uf, simple_get_script,
643 &f->data[0], free_up);
644
645 aa = hb_unicode_funcs_create (uf);
646
647 hb_unicode_funcs_destroy (uf);
648
649 /* make sure the 'uf' didn't get freed, since 'aa' holds a ref */
650 g_assert (!f->data[0].freed);
651
652 hb_unicode_funcs_set_script_func (aa, a_is_for_arabic_get_script,
653 &f->data[1], free_up);
654
655 g_assert_cmphex (hb_unicode_script (aa, 'a'), ==, HB_SCRIPT_ARABIC);
656 g_assert_cmphex (hb_unicode_script (aa, 'b'), ==, HB_SCRIPT_LATIN);
657 g_assert_cmphex (hb_unicode_script (aa, '0'), ==, HB_SCRIPT_UNKNOWN);
658
659 g_assert (!f->data[0].freed && !f->data[1].freed);
660 hb_unicode_funcs_destroy (aa);
661 g_assert (f->data[0].freed && f->data[1].freed);
662 }
663
664
665 static hb_script_t
script_roundtrip_default(hb_script_t script)666 script_roundtrip_default (hb_script_t script)
667 {
668 return hb_script_from_iso15924_tag (hb_script_to_iso15924_tag (script));
669 }
670
671 #ifdef HAVE_GLIB
672 static hb_script_t
script_roundtrip_glib(hb_script_t script)673 script_roundtrip_glib (hb_script_t script)
674 {
675 return hb_glib_script_to_script (hb_glib_script_from_script (script));
676 }
677 #endif
678
679 #ifdef HAVE_ICU
680 static hb_script_t
script_roundtrip_icu(hb_script_t script)681 script_roundtrip_icu (hb_script_t script)
682 {
683 return hb_icu_script_to_script (hb_icu_script_from_script (script));
684 }
685 #endif
686
687 static void
test_unicode_script_roundtrip(gconstpointer user_data)688 test_unicode_script_roundtrip (gconstpointer user_data)
689 {
690 typedef hb_script_t (*roundtrip_func_t) (hb_script_t);
691 roundtrip_func_t roundtrip_func = (roundtrip_func_t) user_data;
692 unsigned int i;
693 gboolean failed = FALSE;
694
695 for (i = 0; i < G_N_ELEMENTS (script_tests); i++) {
696 const test_pair_t *test = &script_tests[i];
697 hb_script_t script = test->value;
698
699 g_test_message ("Test script roundtrip #%d: %x", i, script);
700 g_assert_cmphex (script, ==, roundtrip_func (script));
701 }
702 for (i = 0; i < G_N_ELEMENTS (script_tests_more); i++) {
703 const test_pair_t *test = &script_tests_more[i];
704 hb_script_t script = test->value;
705
706 g_test_message ("Test script roundtrip more #%d: %x", i, script);
707 if (script != roundtrip_func (script)) {
708 g_test_message ("Soft fail: Received %x, expected %x", roundtrip_func (script), script);
709 failed = TRUE;
710 }
711 }
712
713 g_assert_cmphex (HB_SCRIPT_INVALID, ==, roundtrip_func (HB_SCRIPT_INVALID));
714
715 if (failed)
716 g_test_message ("Some script roundtrip tests failed. You probably have an old version of one of the libraries used.");
717 }
718
719
720 static void
test_unicode_normalization(gconstpointer user_data)721 test_unicode_normalization (gconstpointer user_data)
722 {
723 hb_unicode_funcs_t *uf = (hb_unicode_funcs_t *) user_data;
724 gunichar a, b, ab;
725
726
727 /* Test compose() */
728
729 /* Not composable */
730 g_assert (!hb_unicode_compose (uf, 0x0041, 0x0042, &ab) && ab == 0);
731 g_assert (!hb_unicode_compose (uf, 0x0041, 0, &ab) && ab == 0);
732 g_assert (!hb_unicode_compose (uf, 0x0066, 0x0069, &ab) && ab == 0);
733
734 /* Singletons should not compose */
735 g_assert (!hb_unicode_compose (uf, 0x212B, 0, &ab) && ab == 0);
736 g_assert (!hb_unicode_compose (uf, 0x00C5, 0, &ab) && ab == 0);
737 g_assert (!hb_unicode_compose (uf, 0x2126, 0, &ab) && ab == 0);
738 g_assert (!hb_unicode_compose (uf, 0x03A9, 0, &ab) && ab == 0);
739
740 /* Non-starter pairs should not compose */
741 g_assert (!hb_unicode_compose (uf, 0x0308, 0x0301, &ab) && ab == 0); /* !0x0344 */
742 g_assert (!hb_unicode_compose (uf, 0x0F71, 0x0F72, &ab) && ab == 0); /* !0x0F73 */
743
744 /* Pairs */
745 g_assert (hb_unicode_compose (uf, 0x0041, 0x030A, &ab) && ab == 0x00C5);
746 g_assert (hb_unicode_compose (uf, 0x006F, 0x0302, &ab) && ab == 0x00F4);
747 g_assert (hb_unicode_compose (uf, 0x1E63, 0x0307, &ab) && ab == 0x1E69);
748 g_assert (hb_unicode_compose (uf, 0x0073, 0x0323, &ab) && ab == 0x1E63);
749 g_assert (hb_unicode_compose (uf, 0x0064, 0x0307, &ab) && ab == 0x1E0B);
750 g_assert (hb_unicode_compose (uf, 0x0064, 0x0323, &ab) && ab == 0x1E0D);
751
752 /* Hangul */
753 g_assert (hb_unicode_compose (uf, 0xD4CC, 0x11B6, &ab) && ab == 0xD4DB);
754 g_assert (hb_unicode_compose (uf, 0x1111, 0x1171, &ab) && ab == 0xD4CC);
755 g_assert (hb_unicode_compose (uf, 0xCE20, 0x11B8, &ab) && ab == 0xCE31);
756 g_assert (hb_unicode_compose (uf, 0x110E, 0x1173, &ab) && ab == 0xCE20);
757
758
759 /* Test decompose() */
760
761 /* Not decomposable */
762 g_assert (!hb_unicode_decompose (uf, 0x0041, &a, &b) && a == 0x0041 && b == 0);
763 g_assert (!hb_unicode_decompose (uf, 0xFB01, &a, &b) && a == 0xFB01 && b == 0);
764 g_assert (!hb_unicode_decompose (uf, 0x1F1EF, &a, &b) && a == 0x1F1EF && b == 0);
765
766 /* Singletons */
767 g_assert (hb_unicode_decompose (uf, 0x212B, &a, &b) && a == 0x00C5 && b == 0);
768 g_assert (hb_unicode_decompose (uf, 0x2126, &a, &b) && a == 0x03A9 && b == 0);
769
770 /* Non-starter pairs decompose, but not compose */
771 g_assert (hb_unicode_decompose (uf, 0x0344, &a, &b) && a == 0x0308 && b == 0x0301);
772 g_assert (hb_unicode_decompose (uf, 0x0F73, &a, &b) && a == 0x0F71 && b == 0x0F72);
773
774 /* Pairs */
775 g_assert (hb_unicode_decompose (uf, 0x00C5, &a, &b) && a == 0x0041 && b == 0x030A);
776 g_assert (hb_unicode_decompose (uf, 0x00F4, &a, &b) && a == 0x006F && b == 0x0302);
777 g_assert (hb_unicode_decompose (uf, 0x1E69, &a, &b) && a == 0x1E63 && b == 0x0307);
778 g_assert (hb_unicode_decompose (uf, 0x1E63, &a, &b) && a == 0x0073 && b == 0x0323);
779 g_assert (hb_unicode_decompose (uf, 0x1E0B, &a, &b) && a == 0x0064 && b == 0x0307);
780 g_assert (hb_unicode_decompose (uf, 0x1E0D, &a, &b) && a == 0x0064 && b == 0x0323);
781
782 /* Hangul */
783 g_assert (hb_unicode_decompose (uf, 0xD4DB, &a, &b) && a == 0xD4CC && b == 0x11B6);
784 g_assert (hb_unicode_decompose (uf, 0xD4CC, &a, &b) && a == 0x1111 && b == 0x1171);
785 g_assert (hb_unicode_decompose (uf, 0xCE31, &a, &b) && a == 0xCE20 && b == 0x11B8);
786 g_assert (hb_unicode_decompose (uf, 0xCE20, &a, &b) && a == 0x110E && b == 0x1173);
787 }
788
789
790
791 int
main(int argc,char ** argv)792 main (int argc, char **argv)
793 {
794 hb_test_init (&argc, &argv);
795
796 hb_test_add (test_unicode_properties_nil);
797 hb_test_add (test_unicode_properties_empty);
798
799 hb_test_add_data_flavor (hb_unicode_funcs_get_default (), "default", test_unicode_properties);
800 hb_test_add_data_flavor (hb_unicode_funcs_get_default (), "default", test_unicode_normalization);
801 hb_test_add_data_flavor ((gconstpointer) script_roundtrip_default, "default", test_unicode_script_roundtrip);
802 #ifdef HAVE_GLIB
803 hb_test_add_data_flavor (hb_glib_get_unicode_funcs (), "glib", test_unicode_properties);
804 hb_test_add_data_flavor (hb_glib_get_unicode_funcs (), "glib", test_unicode_normalization);
805 hb_test_add_data_flavor ((gconstpointer) script_roundtrip_glib, "glib", test_unicode_script_roundtrip);
806 #endif
807 #ifdef HAVE_ICU
808 hb_test_add_data_flavor (hb_icu_get_unicode_funcs (), "icu", test_unicode_properties);
809 hb_test_add_data_flavor (hb_icu_get_unicode_funcs (), "icu", test_unicode_normalization);
810 hb_test_add_data_flavor ((gconstpointer) script_roundtrip_icu, "icu", test_unicode_script_roundtrip);
811 #endif
812
813 hb_test_add (test_unicode_chainup);
814
815 hb_test_add (test_unicode_setters);
816
817 hb_test_add_fixture (data_fixture, NULL, test_unicode_subclassing_nil);
818 hb_test_add_fixture (data_fixture, NULL, test_unicode_subclassing_default);
819 hb_test_add_fixture (data_fixture, NULL, test_unicode_subclassing_deep);
820
821 return hb_test_run ();
822 }
823