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 /* Unicode-6.1 character additions */
161 { 0xA674, 230 },
162
163 /* Unicode-7.0 character additions */
164 { 0x1AB0, 230 },
165
166 /* Unicode-8.0 character additions */
167 { 0xA69E, 230 },
168
169 /* Unicode-9.0 character additions */
170 { 0x1E000, 230 },
171
172 /* Unicode-10.0 character additions */
173 { 0x1DF6, 232 },
174
175 /* Unicode-11.0 character additions */
176 { 0x07FD, 220 },
177
178 /* Unicode-12.0 character additions */
179 { 0x0EBA, 9 },
180
181 /* Unicode-13.0 character additions */
182 { 0x1ABF, 220 },
183
184 { 0x111111, 0 }
185 };
186
187
188 static const test_pair_t general_category_tests[] =
189 {
190 { 0x000D, HB_UNICODE_GENERAL_CATEGORY_CONTROL },
191 { 0x200E, HB_UNICODE_GENERAL_CATEGORY_FORMAT },
192 { 0x0378, HB_UNICODE_GENERAL_CATEGORY_UNASSIGNED },
193 { 0xE000, HB_UNICODE_GENERAL_CATEGORY_PRIVATE_USE },
194 { 0xD800, HB_UNICODE_GENERAL_CATEGORY_SURROGATE },
195 { 0x0061, HB_UNICODE_GENERAL_CATEGORY_LOWERCASE_LETTER },
196 { 0x02B0, HB_UNICODE_GENERAL_CATEGORY_MODIFIER_LETTER },
197 { 0x3400, HB_UNICODE_GENERAL_CATEGORY_OTHER_LETTER },
198 { 0x01C5, HB_UNICODE_GENERAL_CATEGORY_TITLECASE_LETTER },
199 { 0xFF21, HB_UNICODE_GENERAL_CATEGORY_UPPERCASE_LETTER },
200 { 0x0903, HB_UNICODE_GENERAL_CATEGORY_SPACING_MARK },
201 { 0x20DD, HB_UNICODE_GENERAL_CATEGORY_ENCLOSING_MARK },
202 { 0xA806, HB_UNICODE_GENERAL_CATEGORY_NON_SPACING_MARK },
203 { 0xFF10, HB_UNICODE_GENERAL_CATEGORY_DECIMAL_NUMBER },
204 { 0x16EE, HB_UNICODE_GENERAL_CATEGORY_LETTER_NUMBER },
205 { 0x17F0, HB_UNICODE_GENERAL_CATEGORY_OTHER_NUMBER },
206 { 0x005F, HB_UNICODE_GENERAL_CATEGORY_CONNECT_PUNCTUATION },
207 { 0x058A, HB_UNICODE_GENERAL_CATEGORY_DASH_PUNCTUATION },
208 { 0x0F3B, HB_UNICODE_GENERAL_CATEGORY_CLOSE_PUNCTUATION },
209 { 0x2019, HB_UNICODE_GENERAL_CATEGORY_FINAL_PUNCTUATION },
210 { 0x2018, HB_UNICODE_GENERAL_CATEGORY_INITIAL_PUNCTUATION },
211 { 0x2016, HB_UNICODE_GENERAL_CATEGORY_OTHER_PUNCTUATION },
212 { 0x0F3A, HB_UNICODE_GENERAL_CATEGORY_OPEN_PUNCTUATION },
213 { 0x20A0, HB_UNICODE_GENERAL_CATEGORY_CURRENCY_SYMBOL },
214 { 0x309B, HB_UNICODE_GENERAL_CATEGORY_MODIFIER_SYMBOL },
215 { 0xFB29, HB_UNICODE_GENERAL_CATEGORY_MATH_SYMBOL },
216 { 0x00A6, HB_UNICODE_GENERAL_CATEGORY_OTHER_SYMBOL },
217 { 0x2028, HB_UNICODE_GENERAL_CATEGORY_LINE_SEPARATOR },
218 { 0x2029, HB_UNICODE_GENERAL_CATEGORY_PARAGRAPH_SEPARATOR },
219 { 0x202F, HB_UNICODE_GENERAL_CATEGORY_SPACE_SEPARATOR },
220
221 { 0x111111, HB_UNICODE_GENERAL_CATEGORY_UNASSIGNED }
222 };
223 static const test_pair_t general_category_tests_more[] =
224 {
225 /* Unicode-5.2 character additions */
226 { 0x1F131, HB_UNICODE_GENERAL_CATEGORY_OTHER_SYMBOL },
227
228 /* Unicode-6.0 character additions */
229 { 0x0620, HB_UNICODE_GENERAL_CATEGORY_OTHER_LETTER },
230
231 /* Unicode-6.1 character additions */
232 { 0x058F, HB_UNICODE_GENERAL_CATEGORY_CURRENCY_SYMBOL },
233
234 /* Unicode-6.2 character additions */
235 { 0x20BA, HB_UNICODE_GENERAL_CATEGORY_CURRENCY_SYMBOL },
236
237 /* Unicode-6.3 character additions */
238 { 0x061C, HB_UNICODE_GENERAL_CATEGORY_FORMAT },
239
240 /* Unicode-7.0 character additions */
241 { 0x058D, HB_UNICODE_GENERAL_CATEGORY_OTHER_SYMBOL },
242
243 /* Unicode-8.0 character additions */
244 { 0x08E3, HB_UNICODE_GENERAL_CATEGORY_NON_SPACING_MARK },
245
246 /* Unicode-9.0 character additions */
247 { 0x08D4, HB_UNICODE_GENERAL_CATEGORY_NON_SPACING_MARK },
248
249 /* Unicode-10.0 character additions */
250 { 0x09FD, HB_UNICODE_GENERAL_CATEGORY_OTHER_PUNCTUATION },
251
252 /* Unicode-11.0 character additions */
253 { 0x0560, HB_UNICODE_GENERAL_CATEGORY_LOWERCASE_LETTER },
254
255 /* Unicode-12.0 character additions */
256 { 0x0C77, HB_UNICODE_GENERAL_CATEGORY_OTHER_PUNCTUATION },
257
258 /* Unicode-12.1 character additions */
259 { 0x32FF, HB_UNICODE_GENERAL_CATEGORY_OTHER_SYMBOL },
260
261 /* Unicode-13.0 character additions */
262 { 0x08BE, HB_UNICODE_GENERAL_CATEGORY_OTHER_LETTER },
263
264 { 0x111111, HB_UNICODE_GENERAL_CATEGORY_UNASSIGNED }
265 };
266
267 static const test_pair_t mirroring_tests[] =
268 {
269 /* Some characters that do NOT mirror */
270 { 0x0020, 0x0020 },
271 { 0x0041, 0x0041 },
272 { 0x00F0, 0x00F0 },
273 { 0x27CC, 0x27CC },
274 { 0xE01EF, 0xE01EF },
275 { 0x1D7C3, 0x1D7C3 },
276 { 0x100000, 0x100000 },
277
278 /* Some characters that do mirror */
279 { 0x0029, 0x0028 },
280 { 0x0028, 0x0029 },
281 { 0x003E, 0x003C },
282 { 0x003C, 0x003E },
283 { 0x005D, 0x005B },
284 { 0x005B, 0x005D },
285 { 0x007D, 0x007B },
286 { 0x007B, 0x007D },
287 { 0x00BB, 0x00AB },
288 { 0x00AB, 0x00BB },
289 { 0x226B, 0x226A },
290 { 0x226A, 0x226B },
291 { 0x22F1, 0x22F0 },
292 { 0x22F0, 0x22F1 },
293 { 0xFF60, 0xFF5F },
294 { 0xFF5F, 0xFF60 },
295 { 0xFF63, 0xFF62 },
296 { 0xFF62, 0xFF63 },
297
298 { 0x111111, 0x111111 },
299 };
300 static const test_pair_t mirroring_tests_more[] =
301 {
302 /* Unicode-6.1 character additions */
303 { 0x27CB, 0x27CD },
304
305 /* Unicode-11.0 character additions */
306 { 0x2BFE, 0x221F },
307
308 { 0x111111, 0x111111 }
309 };
310
311 static const test_pair_t script_tests[] =
312 {
313 { 0x002A, HB_SCRIPT_COMMON },
314 { 0x0670, HB_SCRIPT_INHERITED },
315 { 0x060D, HB_SCRIPT_ARABIC },
316 { 0x0559, HB_SCRIPT_ARMENIAN },
317 { 0x09CD, HB_SCRIPT_BENGALI },
318 { 0x31B6, HB_SCRIPT_BOPOMOFO },
319 { 0x13A2, HB_SCRIPT_CHEROKEE },
320 { 0x2CFD, HB_SCRIPT_COPTIC },
321 { 0x0482, HB_SCRIPT_CYRILLIC },
322 { 0x10401, HB_SCRIPT_DESERET },
323 { 0x094D, HB_SCRIPT_DEVANAGARI },
324 { 0x1258, HB_SCRIPT_ETHIOPIC },
325 { 0x10FC, HB_SCRIPT_GEORGIAN },
326 { 0x10341, HB_SCRIPT_GOTHIC },
327 { 0x0375, HB_SCRIPT_GREEK },
328 { 0x0A83, HB_SCRIPT_GUJARATI },
329 { 0x0A3C, HB_SCRIPT_GURMUKHI },
330 { 0x3005, HB_SCRIPT_HAN },
331 { 0x1100, HB_SCRIPT_HANGUL },
332 { 0x05BF, HB_SCRIPT_HEBREW },
333 { 0x309F, HB_SCRIPT_HIRAGANA },
334 { 0x0CBC, HB_SCRIPT_KANNADA },
335 { 0x30FF, HB_SCRIPT_KATAKANA },
336 { 0x17DD, HB_SCRIPT_KHMER },
337 { 0x0EDD, HB_SCRIPT_LAO },
338 { 0x0061, HB_SCRIPT_LATIN },
339 { 0x0D3D, HB_SCRIPT_MALAYALAM },
340 { 0x1843, HB_SCRIPT_MONGOLIAN },
341 { 0x1031, HB_SCRIPT_MYANMAR },
342 { 0x169C, HB_SCRIPT_OGHAM },
343 { 0x10322, HB_SCRIPT_OLD_ITALIC },
344 { 0x0B3C, HB_SCRIPT_ORIYA },
345 { 0x16EF, HB_SCRIPT_RUNIC },
346 { 0x0DBD, HB_SCRIPT_SINHALA },
347 { 0x0711, HB_SCRIPT_SYRIAC },
348 { 0x0B82, HB_SCRIPT_TAMIL },
349 { 0x0C03, HB_SCRIPT_TELUGU },
350 { 0x07B1, HB_SCRIPT_THAANA },
351 { 0x0E31, HB_SCRIPT_THAI },
352 { 0x0FD4, HB_SCRIPT_TIBETAN },
353 { 0x1401, HB_SCRIPT_CANADIAN_SYLLABICS },
354 { 0xA015, HB_SCRIPT_YI },
355 { 0x1700, HB_SCRIPT_TAGALOG },
356 { 0x1720, HB_SCRIPT_HANUNOO },
357 { 0x1740, HB_SCRIPT_BUHID },
358 { 0x1760, HB_SCRIPT_TAGBANWA },
359
360 /* Unicode-4.0 additions */
361 { 0x2800, HB_SCRIPT_BRAILLE },
362 { 0x10808, HB_SCRIPT_CYPRIOT },
363 { 0x1932, HB_SCRIPT_LIMBU },
364 { 0x10480, HB_SCRIPT_OSMANYA },
365 { 0x10450, HB_SCRIPT_SHAVIAN },
366 { 0x10000, HB_SCRIPT_LINEAR_B },
367 { 0x1950, HB_SCRIPT_TAI_LE },
368 { 0x1039F, HB_SCRIPT_UGARITIC },
369
370 /* Unicode-4.1 additions */
371 { 0x1980, HB_SCRIPT_NEW_TAI_LUE },
372 { 0x1A1F, HB_SCRIPT_BUGINESE },
373 { 0x2C00, HB_SCRIPT_GLAGOLITIC },
374 { 0x2D6F, HB_SCRIPT_TIFINAGH },
375 { 0xA800, HB_SCRIPT_SYLOTI_NAGRI },
376 { 0x103D0, HB_SCRIPT_OLD_PERSIAN },
377 { 0x10A3F, HB_SCRIPT_KHAROSHTHI },
378
379 /* Unicode-5.0 additions */
380 { 0x0378, HB_SCRIPT_UNKNOWN },
381 { 0x1B04, HB_SCRIPT_BALINESE },
382 { 0x12000, HB_SCRIPT_CUNEIFORM },
383 { 0x10900, HB_SCRIPT_PHOENICIAN },
384 { 0xA840, HB_SCRIPT_PHAGS_PA },
385 { 0x07C0, HB_SCRIPT_NKO },
386
387 /* Unicode-5.1 additions */
388 { 0xA900, HB_SCRIPT_KAYAH_LI },
389 { 0x1C00, HB_SCRIPT_LEPCHA },
390 { 0xA930, HB_SCRIPT_REJANG },
391 { 0x1B80, HB_SCRIPT_SUNDANESE },
392 { 0xA880, HB_SCRIPT_SAURASHTRA },
393 { 0xAA00, HB_SCRIPT_CHAM },
394 { 0x1C50, HB_SCRIPT_OL_CHIKI },
395 { 0xA500, HB_SCRIPT_VAI },
396 { 0x102A0, HB_SCRIPT_CARIAN },
397 { 0x10280, HB_SCRIPT_LYCIAN },
398 { 0x1093F, HB_SCRIPT_LYDIAN },
399
400 { 0x111111, HB_SCRIPT_UNKNOWN }
401 };
402 static const test_pair_t script_tests_more[] =
403 {
404 /* Unicode-5.2 additions */
405 { 0x10B00, HB_SCRIPT_AVESTAN },
406 { 0xA6A0, HB_SCRIPT_BAMUM },
407 { 0x1400, HB_SCRIPT_CANADIAN_ABORIGINAL },
408 { 0x13000, HB_SCRIPT_EGYPTIAN_HIEROGLYPHS },
409 { 0x10840, HB_SCRIPT_IMPERIAL_ARAMAIC },
410 { 0x1CED, HB_SCRIPT_INHERITED },
411 { 0x10B60, HB_SCRIPT_INSCRIPTIONAL_PAHLAVI },
412 { 0x10B40, HB_SCRIPT_INSCRIPTIONAL_PARTHIAN },
413 { 0xA980, HB_SCRIPT_JAVANESE },
414 { 0x11082, HB_SCRIPT_KAITHI },
415 { 0xA4D0, HB_SCRIPT_LISU },
416 { 0xABE5, HB_SCRIPT_MEETEI_MAYEK },
417 { 0x10A60, HB_SCRIPT_OLD_SOUTH_ARABIAN },
418 { 0x10C00, HB_SCRIPT_OLD_TURKIC },
419 { 0x0800, HB_SCRIPT_SAMARITAN },
420 { 0x1A20, HB_SCRIPT_TAI_THAM },
421 { 0xAA80, HB_SCRIPT_TAI_VIET },
422
423 /* Unicode-6.0 additions */
424 { 0x1BC0, HB_SCRIPT_BATAK },
425 { 0x11000, HB_SCRIPT_BRAHMI },
426 { 0x0840, HB_SCRIPT_MANDAIC },
427
428 /* Unicode-6.1 additions */
429 { 0x10980, HB_SCRIPT_MEROITIC_HIEROGLYPHS },
430 { 0x109A0, HB_SCRIPT_MEROITIC_CURSIVE },
431 { 0x110D0, HB_SCRIPT_SORA_SOMPENG },
432 { 0x11100, HB_SCRIPT_CHAKMA },
433 { 0x11180, HB_SCRIPT_SHARADA },
434 { 0x11680, HB_SCRIPT_TAKRI },
435 { 0x16F00, HB_SCRIPT_MIAO },
436
437 /* Unicode-6.2 additions */
438 { 0x20BA, HB_SCRIPT_COMMON },
439
440 /* Unicode-6.3 additions */
441 { 0x2066, HB_SCRIPT_COMMON },
442
443 /* Unicode-7.0 additions */
444 { 0x10350, HB_SCRIPT_OLD_PERMIC },
445 { 0x10500, HB_SCRIPT_ELBASAN },
446 { 0x10530, HB_SCRIPT_CAUCASIAN_ALBANIAN },
447 { 0x10600, HB_SCRIPT_LINEAR_A },
448 { 0x10860, HB_SCRIPT_PALMYRENE },
449 { 0x10880, HB_SCRIPT_NABATAEAN },
450 { 0x10A80, HB_SCRIPT_OLD_NORTH_ARABIAN },
451 { 0x10AC0, HB_SCRIPT_MANICHAEAN },
452 { 0x10B80, HB_SCRIPT_PSALTER_PAHLAVI },
453 { 0x11150, HB_SCRIPT_MAHAJANI },
454 { 0x11200, HB_SCRIPT_KHOJKI },
455 { 0x112B0, HB_SCRIPT_KHUDAWADI },
456 { 0x11300, HB_SCRIPT_GRANTHA },
457 { 0x11480, HB_SCRIPT_TIRHUTA },
458 { 0x11580, HB_SCRIPT_SIDDHAM },
459 { 0x11600, HB_SCRIPT_MODI },
460 { 0x118A0, HB_SCRIPT_WARANG_CITI },
461 { 0x11AC0, HB_SCRIPT_PAU_CIN_HAU },
462 { 0x16A40, HB_SCRIPT_MRO },
463 { 0x16AD0, HB_SCRIPT_BASSA_VAH },
464 { 0x16B00, HB_SCRIPT_PAHAWH_HMONG },
465 { 0x1BC00, HB_SCRIPT_DUPLOYAN },
466 { 0x1E800, HB_SCRIPT_MENDE_KIKAKUI },
467
468 /* Unicode-8.0 additions */
469 { 0x108E0, HB_SCRIPT_HATRAN },
470 { 0x10C80, HB_SCRIPT_OLD_HUNGARIAN },
471 { 0x11280, HB_SCRIPT_MULTANI },
472 { 0x11700, HB_SCRIPT_AHOM },
473 { 0x14400, HB_SCRIPT_ANATOLIAN_HIEROGLYPHS },
474 { 0x1D800, HB_SCRIPT_SIGNWRITING },
475
476 /* Unicode-9.0 additions */
477 { 0x104B0, HB_SCRIPT_OSAGE },
478 { 0x11400, HB_SCRIPT_NEWA },
479 { 0x11C00, HB_SCRIPT_BHAIKSUKI },
480 { 0x11C70, HB_SCRIPT_MARCHEN },
481 { 0x17000, HB_SCRIPT_TANGUT },
482 { 0x1E900, HB_SCRIPT_ADLAM },
483
484 /* Unicode-10.0 additions */
485 { 0x11A00, HB_SCRIPT_ZANABAZAR_SQUARE },
486 { 0x11A50, HB_SCRIPT_SOYOMBO },
487 { 0x11D00, HB_SCRIPT_MASARAM_GONDI },
488 { 0x1B170, HB_SCRIPT_NUSHU },
489
490 /* Unicode-11.0 additions */
491 { 0x10D00, HB_SCRIPT_HANIFI_ROHINGYA },
492 { 0x10F00, HB_SCRIPT_OLD_SOGDIAN },
493 { 0x10F30, HB_SCRIPT_SOGDIAN },
494 { 0x11800, HB_SCRIPT_DOGRA },
495 { 0x11D60, HB_SCRIPT_GUNJALA_GONDI },
496 { 0x11EE0, HB_SCRIPT_MAKASAR },
497 { 0x16E40, HB_SCRIPT_MEDEFAIDRIN },
498
499 /* Unicode-12.0 additions */
500 { 0x10FE0, HB_SCRIPT_ELYMAIC },
501 { 0x119A0, HB_SCRIPT_NANDINAGARI },
502 { 0x1E100, HB_SCRIPT_NYIAKENG_PUACHUE_HMONG },
503 { 0x1E2C0, HB_SCRIPT_WANCHO },
504
505 /* Unicode-12.1 additions */
506 { 0x32FF, HB_SCRIPT_COMMON },
507
508 /* Unicode-13.0 additions */
509 { 0x10E80, HB_SCRIPT_YEZIDI },
510 { 0x10FB0, HB_SCRIPT_CHORASMIAN },
511 { 0x11900, HB_SCRIPT_DIVES_AKURU },
512 { 0x18B00, HB_SCRIPT_KHITAN_SMALL_SCRIPT },
513
514 { 0x111111, HB_SCRIPT_UNKNOWN }
515 };
516
517
518 typedef unsigned int (*get_func_t) (hb_unicode_funcs_t *ufuncs,
519 hb_codepoint_t unicode,
520 void *user_data);
521 typedef unsigned int (*func_setter_func_t) (hb_unicode_funcs_t *ufuncs,
522 get_func_t func,
523 void *user_data,
524 hb_destroy_func_t destroy);
525 typedef unsigned int (*getter_func_t) (hb_unicode_funcs_t *ufuncs,
526 hb_codepoint_t unicode);
527
528 typedef struct {
529 const char *name;
530 func_setter_func_t func_setter;
531 getter_func_t getter;
532 const test_pair_t *tests;
533 unsigned int num_tests;
534 const test_pair_t *tests_more;
535 unsigned int num_tests_more;
536 unsigned int default_value;
537 } property_t;
538
539 #define RETURNS_UNICODE_ITSELF ((unsigned int) -1)
540
541 #define PROPERTY(name, DEFAULT) \
542 { \
543 #name, \
544 (func_setter_func_t) hb_unicode_funcs_set_##name##_func, \
545 (getter_func_t) hb_unicode_##name, \
546 name##_tests, \
547 G_N_ELEMENTS (name##_tests), \
548 name##_tests_more, \
549 G_N_ELEMENTS (name##_tests_more), \
550 DEFAULT \
551 }
552 #pragma GCC diagnostic push
553 #pragma GCC diagnostic ignored "-Wcast-function-type"
554 static const property_t properties[] =
555 {
556 PROPERTY (combining_class, 0),
557 PROPERTY (general_category, (unsigned int) HB_UNICODE_GENERAL_CATEGORY_OTHER_LETTER),
558 PROPERTY (mirroring, RETURNS_UNICODE_ITSELF),
559 PROPERTY (script, (unsigned int) HB_SCRIPT_UNKNOWN)
560 };
561 #pragma GCC diagnostic pop
562 #undef PROPERTY
563
564 static void
test_unicode_properties(gconstpointer user_data,hb_bool_t lenient)565 test_unicode_properties (gconstpointer user_data, hb_bool_t lenient)
566 {
567 hb_unicode_funcs_t *uf = (hb_unicode_funcs_t *) user_data;
568 unsigned int i, j;
569 gboolean failed = TRUE;
570
571 g_assert (hb_unicode_funcs_is_immutable (uf));
572 g_assert (hb_unicode_funcs_get_parent (uf));
573
574 for (i = 0; i < G_N_ELEMENTS (properties); i++) {
575 const property_t *p = &properties[i];
576 const test_pair_t *tests;
577
578 g_test_message ("Testing property %s", p->name);
579 tests = p->tests;
580 for (j = 0; j < p->num_tests; j++) {
581 g_test_message ("Test %s #%d: U+%04X", p->name, j, tests[j].unicode);
582 g_assert_cmphex (p->getter (uf, tests[j].unicode), ==, tests[j].value);
583 }
584 /* These tests are from Unicode 5.2 onward and older glib/ICU
585 * don't get them right. Just warn instead of assert. */
586 tests = p->tests_more;
587 for (j = 0; j < p->num_tests_more; j++) {
588 g_test_message ("Test %s more #%d: U+%04X", p->name, j, tests[j].unicode);
589 if (lenient) {
590 if (p->getter (uf, tests[j].unicode) != tests[j].value) {
591 g_test_message ("Soft fail: Received %x, expected %x", p->getter (uf, tests[j].unicode), tests[j].value);
592 failed = TRUE;
593 }
594 }
595 else
596 g_assert_cmphex (p->getter (uf, tests[j].unicode), ==, tests[j].value);
597 }
598 }
599
600 if (failed)
601 g_test_message ("Some property tests failed. You probably have an old version of one of the libraries used.");
602 }
603 static void
test_unicode_properties_lenient(gconstpointer user_data)604 test_unicode_properties_lenient (gconstpointer user_data)
605 {
606 test_unicode_properties (user_data, TRUE);
607 }
608 static void
test_unicode_properties_strict(gconstpointer user_data)609 test_unicode_properties_strict (gconstpointer user_data)
610 {
611 test_unicode_properties (user_data, FALSE);
612 }
613
614 static hb_codepoint_t
default_value(hb_codepoint_t _default_value,hb_codepoint_t unicode)615 default_value (hb_codepoint_t _default_value, hb_codepoint_t unicode)
616 {
617 return _default_value == RETURNS_UNICODE_ITSELF ? unicode : _default_value;
618 }
619
620 static void
_test_unicode_properties_nil(hb_unicode_funcs_t * uf)621 _test_unicode_properties_nil (hb_unicode_funcs_t *uf)
622 {
623 unsigned int i, j;
624
625 for (i = 0; i < G_N_ELEMENTS (properties); i++) {
626 const property_t *p = &properties[i];
627 const test_pair_t *tests;
628
629 g_test_message ("Testing property %s", p->name);
630 tests = p->tests;
631 for (j = 0; j < p->num_tests; j++) {
632 g_test_message ("Test %s #%d: U+%04X", p->name, j, tests[j].unicode);
633 g_assert_cmphex (p->getter (uf, tests[j].unicode), ==, default_value (p->default_value, tests[j].unicode));
634 }
635 tests = p->tests_more;
636 for (j = 0; j < p->num_tests_more; j++) {
637 g_test_message ("Test %s more #%d: U+%04X", p->name, j, tests[j].unicode);
638 g_assert_cmphex (p->getter (uf, tests[j].unicode), ==, default_value (p->default_value, tests[j].unicode));
639 }
640 }
641 }
642
643 static void
test_unicode_properties_nil(void)644 test_unicode_properties_nil (void)
645 {
646 hb_unicode_funcs_t *uf = hb_unicode_funcs_create (NULL);
647
648 g_assert (!hb_unicode_funcs_is_immutable (uf));
649 _test_unicode_properties_nil (uf);
650
651 hb_unicode_funcs_destroy (uf);
652 }
653
654 static void
test_unicode_properties_empty(void)655 test_unicode_properties_empty (void)
656 {
657 hb_unicode_funcs_t *uf = hb_unicode_funcs_get_empty ();
658
659 g_assert (uf);
660 g_assert (hb_unicode_funcs_is_immutable (uf));
661 _test_unicode_properties_nil (uf);
662 }
663
664
665 static void
test_unicode_chainup(void)666 test_unicode_chainup (void)
667 {
668 hb_unicode_funcs_t *uf, *uf2;
669
670 /* Chain-up to nil */
671
672 uf = hb_unicode_funcs_create (NULL);
673 g_assert (!hb_unicode_funcs_is_immutable (uf));
674
675 uf2 = hb_unicode_funcs_create (uf);
676 g_assert (hb_unicode_funcs_is_immutable (uf));
677 hb_unicode_funcs_destroy (uf);
678
679 g_assert (!hb_unicode_funcs_is_immutable (uf2));
680 _test_unicode_properties_nil (uf2);
681
682 hb_unicode_funcs_destroy (uf2);
683
684 /* Chain-up to default */
685
686 uf = hb_unicode_funcs_create (hb_unicode_funcs_get_default ());
687 g_assert (!hb_unicode_funcs_is_immutable (uf));
688
689 uf2 = hb_unicode_funcs_create (uf);
690 g_assert (hb_unicode_funcs_is_immutable (uf));
691 hb_unicode_funcs_destroy (uf);
692
693 g_assert (!hb_unicode_funcs_is_immutable (uf2));
694 hb_unicode_funcs_make_immutable (uf2);
695 test_unicode_properties_strict (uf2);
696
697 hb_unicode_funcs_destroy (uf2);
698
699 }
700
701 static void
test_unicode_setters(void)702 test_unicode_setters (void)
703 {
704 hb_unicode_funcs_t *uf;
705 unsigned int i;
706
707 /* This is cruel: we use script-returning functions to test all properties,
708 * but it works. */
709
710 for (i = 0; i < G_N_ELEMENTS (properties); i++) {
711 const property_t *p = &properties[i];
712 data_t data[2] = {{MAGIC0, FALSE}, {MAGIC1, FALSE}};
713
714 g_test_message ("Testing property %s", p->name);
715
716 uf = hb_unicode_funcs_create (NULL);
717 g_assert (!hb_unicode_funcs_is_immutable (uf));
718
719 p->func_setter (uf, (get_func_t) simple_get_script, &data[0], free_up);
720
721 g_assert_cmphex (p->getter (uf, 'a'), ==, HB_SCRIPT_LATIN);
722 g_assert_cmphex (p->getter (uf, '0'), ==, HB_SCRIPT_UNKNOWN);
723
724 p->func_setter (uf, (get_func_t) NULL, NULL, NULL);
725 g_assert (data[0].freed && !data[1].freed);
726
727 g_assert (!hb_unicode_funcs_is_immutable (uf));
728 hb_unicode_funcs_make_immutable (uf);
729 g_assert (hb_unicode_funcs_is_immutable (uf));
730
731 /* Since uf is immutable now, the following setter should do nothing. */
732 p->func_setter (uf, (get_func_t) a_is_for_arabic_get_script, &data[1], free_up);
733
734 g_assert (data[0].freed && !data[1].freed);
735 hb_unicode_funcs_destroy (uf);
736 g_assert (data[0].freed && !data[1].freed);
737 }
738 }
739
740
741
742 typedef struct {
743 data_t data[2];
744 } data_fixture_t;
745
746 static void
data_fixture_init(data_fixture_t * f,gconstpointer user_data HB_UNUSED)747 data_fixture_init (data_fixture_t *f, gconstpointer user_data HB_UNUSED)
748 {
749 f->data[0].value = MAGIC0;
750 f->data[1].value = MAGIC1;
751 }
752 static void
data_fixture_finish(data_fixture_t * f HB_UNUSED,gconstpointer user_data HB_UNUSED)753 data_fixture_finish (data_fixture_t *f HB_UNUSED, gconstpointer user_data HB_UNUSED)
754 {
755 }
756
757 static void
test_unicode_subclassing_nil(data_fixture_t * f,gconstpointer user_data HB_UNUSED)758 test_unicode_subclassing_nil (data_fixture_t *f, gconstpointer user_data HB_UNUSED)
759 {
760 hb_unicode_funcs_t *uf, *aa;
761
762 uf = hb_unicode_funcs_create (NULL);
763
764 aa = hb_unicode_funcs_create (uf);
765
766 hb_unicode_funcs_destroy (uf);
767
768 hb_unicode_funcs_set_script_func (aa, a_is_for_arabic_get_script,
769 &f->data[1], free_up);
770
771 g_assert_cmphex (hb_unicode_script (aa, 'a'), ==, HB_SCRIPT_ARABIC);
772 g_assert_cmphex (hb_unicode_script (aa, 'b'), ==, HB_SCRIPT_UNKNOWN);
773
774 g_assert (!f->data[0].freed && !f->data[1].freed);
775 hb_unicode_funcs_destroy (aa);
776 g_assert (!f->data[0].freed && f->data[1].freed);
777 }
778
779 static void
test_unicode_subclassing_default(data_fixture_t * f,gconstpointer user_data HB_UNUSED)780 test_unicode_subclassing_default (data_fixture_t *f, gconstpointer user_data HB_UNUSED)
781 {
782 hb_unicode_funcs_t *uf, *aa;
783
784 uf = hb_unicode_funcs_get_default ();
785 aa = hb_unicode_funcs_create (uf);
786
787 hb_unicode_funcs_set_script_func (aa, a_is_for_arabic_get_script,
788 &f->data[1], free_up);
789
790 g_assert_cmphex (hb_unicode_script (aa, 'a'), ==, HB_SCRIPT_ARABIC);
791 g_assert_cmphex (hb_unicode_script (aa, 'b'), ==, HB_SCRIPT_LATIN);
792
793 g_assert (!f->data[0].freed && !f->data[1].freed);
794 hb_unicode_funcs_destroy (aa);
795 g_assert (!f->data[0].freed && f->data[1].freed);
796 }
797
798 static void
test_unicode_subclassing_deep(data_fixture_t * f,gconstpointer user_data HB_UNUSED)799 test_unicode_subclassing_deep (data_fixture_t *f, gconstpointer user_data HB_UNUSED)
800 {
801 hb_unicode_funcs_t *uf, *aa;
802
803 uf = hb_unicode_funcs_create (NULL);
804
805 hb_unicode_funcs_set_script_func (uf, simple_get_script,
806 &f->data[0], free_up);
807
808 aa = hb_unicode_funcs_create (uf);
809
810 hb_unicode_funcs_destroy (uf);
811
812 /* make sure the 'uf' didn't get freed, since 'aa' holds a ref */
813 g_assert (!f->data[0].freed);
814
815 hb_unicode_funcs_set_script_func (aa, a_is_for_arabic_get_script,
816 &f->data[1], free_up);
817
818 g_assert_cmphex (hb_unicode_script (aa, 'a'), ==, HB_SCRIPT_ARABIC);
819 g_assert_cmphex (hb_unicode_script (aa, 'b'), ==, HB_SCRIPT_LATIN);
820 g_assert_cmphex (hb_unicode_script (aa, '0'), ==, HB_SCRIPT_UNKNOWN);
821
822 g_assert (!f->data[0].freed && !f->data[1].freed);
823 hb_unicode_funcs_destroy (aa);
824 g_assert (f->data[0].freed && f->data[1].freed);
825 }
826
827
828 static hb_script_t
script_roundtrip_default(hb_script_t script)829 script_roundtrip_default (hb_script_t script)
830 {
831 return hb_script_from_iso15924_tag (hb_script_to_iso15924_tag (script));
832 }
833
834 #ifdef HAVE_GLIB
835 static hb_script_t
script_roundtrip_glib(hb_script_t script)836 script_roundtrip_glib (hb_script_t script)
837 {
838 return hb_glib_script_to_script (hb_glib_script_from_script (script));
839 }
840 #endif
841
842 #ifdef HAVE_ICU
843 static hb_script_t
script_roundtrip_icu(hb_script_t script)844 script_roundtrip_icu (hb_script_t script)
845 {
846 return hb_icu_script_to_script (hb_icu_script_from_script (script));
847 }
848 #endif
849
850 static void
test_unicode_script_roundtrip(gconstpointer user_data)851 test_unicode_script_roundtrip (gconstpointer user_data)
852 {
853 typedef hb_script_t (*roundtrip_func_t) (hb_script_t);
854 roundtrip_func_t roundtrip_func = (roundtrip_func_t) user_data;
855 unsigned int i;
856 gboolean failed = FALSE;
857
858 for (i = 0; i < G_N_ELEMENTS (script_tests); i++) {
859 const test_pair_t *test = &script_tests[i];
860 hb_script_t script = test->value;
861
862 g_test_message ("Test script roundtrip #%d: %x", i, script);
863 g_assert_cmphex (script, ==, roundtrip_func (script));
864 }
865 for (i = 0; i < G_N_ELEMENTS (script_tests_more); i++) {
866 const test_pair_t *test = &script_tests_more[i];
867 hb_script_t script = test->value;
868
869 g_test_message ("Test script roundtrip more #%d: %x", i, script);
870 if (script != roundtrip_func (script)) {
871 g_test_message ("Soft fail: Received %x, expected %x", roundtrip_func (script), script);
872 failed = TRUE;
873 }
874 }
875
876 g_assert_cmphex (HB_SCRIPT_INVALID, ==, roundtrip_func (HB_SCRIPT_INVALID));
877
878 if (failed)
879 g_test_message ("Some script roundtrip tests failed. You probably have an old version of one of the libraries used.");
880 }
881
882
883 static void
test_unicode_normalization(gconstpointer user_data)884 test_unicode_normalization (gconstpointer user_data)
885 {
886 hb_unicode_funcs_t *uf = (hb_unicode_funcs_t *) user_data;
887 gunichar a, b, ab;
888
889
890 /* Test compose() */
891
892 /* Not composable */
893 g_assert (!hb_unicode_compose (uf, 0x0041, 0x0042, &ab) && ab == 0);
894 g_assert (!hb_unicode_compose (uf, 0x0041, 0, &ab) && ab == 0);
895 g_assert (!hb_unicode_compose (uf, 0x0066, 0x0069, &ab) && ab == 0);
896
897 /* Singletons should not compose */
898 g_assert (!hb_unicode_compose (uf, 0x212B, 0, &ab) && ab == 0);
899 g_assert (!hb_unicode_compose (uf, 0x00C5, 0, &ab) && ab == 0);
900 g_assert (!hb_unicode_compose (uf, 0x2126, 0, &ab) && ab == 0);
901 g_assert (!hb_unicode_compose (uf, 0x03A9, 0, &ab) && ab == 0);
902
903 /* Non-starter pairs should not compose */
904 g_assert (!hb_unicode_compose (uf, 0x0308, 0x0301, &ab) && ab == 0); /* !0x0344 */
905 g_assert (!hb_unicode_compose (uf, 0x0F71, 0x0F72, &ab) && ab == 0); /* !0x0F73 */
906
907 /* Pairs */
908 g_assert (hb_unicode_compose (uf, 0x0041, 0x030A, &ab) && ab == 0x00C5);
909 g_assert (hb_unicode_compose (uf, 0x006F, 0x0302, &ab) && ab == 0x00F4);
910 g_assert (hb_unicode_compose (uf, 0x1E63, 0x0307, &ab) && ab == 0x1E69);
911 g_assert (hb_unicode_compose (uf, 0x0073, 0x0323, &ab) && ab == 0x1E63);
912 g_assert (hb_unicode_compose (uf, 0x0064, 0x0307, &ab) && ab == 0x1E0B);
913 g_assert (hb_unicode_compose (uf, 0x0064, 0x0323, &ab) && ab == 0x1E0D);
914
915 /* Hangul */
916 g_assert (hb_unicode_compose (uf, 0xD4CC, 0x11B6, &ab) && ab == 0xD4DB);
917 g_assert (hb_unicode_compose (uf, 0x1111, 0x1171, &ab) && ab == 0xD4CC);
918 g_assert (hb_unicode_compose (uf, 0xCE20, 0x11B8, &ab) && ab == 0xCE31);
919 g_assert (hb_unicode_compose (uf, 0x110E, 0x1173, &ab) && ab == 0xCE20);
920
921 g_assert (!hb_unicode_compose (uf, 0xAC00, 0x11A7, &ab));
922 g_assert (hb_unicode_compose (uf, 0xAC00, 0x11A8, &ab) && ab == 0xAC01);
923 g_assert (!hb_unicode_compose (uf, 0xAC01, 0x11A8, &ab));
924
925
926 /* Test decompose() */
927
928 /* Not decomposable */
929 g_assert (!hb_unicode_decompose (uf, 0x0041, &a, &b) && a == 0x0041 && b == 0);
930 g_assert (!hb_unicode_decompose (uf, 0xFB01, &a, &b) && a == 0xFB01 && b == 0);
931 g_assert (!hb_unicode_decompose (uf, 0x1F1EF, &a, &b) && a == 0x1F1EF && b == 0);
932
933 /* Singletons */
934 g_assert (hb_unicode_decompose (uf, 0x212B, &a, &b) && a == 0x00C5 && b == 0);
935 g_assert (hb_unicode_decompose (uf, 0x2126, &a, &b) && a == 0x03A9 && b == 0);
936
937 /* Non-starter pairs decompose, but not compose */
938 g_assert (hb_unicode_decompose (uf, 0x0344, &a, &b) && a == 0x0308 && b == 0x0301);
939 g_assert (hb_unicode_decompose (uf, 0x0F73, &a, &b) && a == 0x0F71 && b == 0x0F72);
940
941 /* Pairs */
942 g_assert (hb_unicode_decompose (uf, 0x00C5, &a, &b) && a == 0x0041 && b == 0x030A);
943 g_assert (hb_unicode_decompose (uf, 0x00F4, &a, &b) && a == 0x006F && b == 0x0302);
944 g_assert (hb_unicode_decompose (uf, 0x1E69, &a, &b) && a == 0x1E63 && b == 0x0307);
945 g_assert (hb_unicode_decompose (uf, 0x1E63, &a, &b) && a == 0x0073 && b == 0x0323);
946 g_assert (hb_unicode_decompose (uf, 0x1E0B, &a, &b) && a == 0x0064 && b == 0x0307);
947 g_assert (hb_unicode_decompose (uf, 0x1E0D, &a, &b) && a == 0x0064 && b == 0x0323);
948
949 /* Hangul */
950 g_assert (hb_unicode_decompose (uf, 0xD4DB, &a, &b) && a == 0xD4CC && b == 0x11B6);
951 g_assert (hb_unicode_decompose (uf, 0xD4CC, &a, &b) && a == 0x1111 && b == 0x1171);
952 g_assert (hb_unicode_decompose (uf, 0xCE31, &a, &b) && a == 0xCE20 && b == 0x11B8);
953 g_assert (hb_unicode_decompose (uf, 0xCE20, &a, &b) && a == 0x110E && b == 0x1173);
954 }
955
956
957
958 int
main(int argc,char ** argv)959 main (int argc, char **argv)
960 {
961 hb_test_init (&argc, &argv);
962
963 hb_test_add (test_unicode_properties_nil);
964 hb_test_add (test_unicode_properties_empty);
965
966 hb_test_add_data_flavor (hb_unicode_funcs_get_default (), "default", test_unicode_properties_strict);
967 hb_test_add_data_flavor (hb_unicode_funcs_get_default (), "default", test_unicode_normalization);
968 hb_test_add_data_flavor ((gconstpointer) script_roundtrip_default, "default", test_unicode_script_roundtrip);
969 #ifdef HAVE_GLIB
970 hb_test_add_data_flavor (hb_glib_get_unicode_funcs (), "glib", test_unicode_properties_lenient);
971 hb_test_add_data_flavor (hb_glib_get_unicode_funcs (), "glib", test_unicode_normalization);
972 hb_test_add_data_flavor ((gconstpointer) script_roundtrip_glib, "glib", test_unicode_script_roundtrip);
973 #endif
974 #ifdef HAVE_ICU
975 hb_test_add_data_flavor (hb_icu_get_unicode_funcs (), "icu", test_unicode_properties_lenient);
976 hb_test_add_data_flavor (hb_icu_get_unicode_funcs (), "icu", test_unicode_normalization);
977 hb_test_add_data_flavor ((gconstpointer) script_roundtrip_icu, "icu", test_unicode_script_roundtrip);
978 #endif
979
980 hb_test_add (test_unicode_chainup);
981
982 hb_test_add (test_unicode_setters);
983
984 hb_test_add_fixture (data_fixture, NULL, test_unicode_subclassing_nil);
985 hb_test_add_fixture (data_fixture, NULL, test_unicode_subclassing_default);
986 hb_test_add_fixture (data_fixture, NULL, test_unicode_subclassing_deep);
987
988 return hb_test_run ();
989 }
990