utf8-selftest.c (fbc59d65059ecfea8b746715e920c325cc16cede) utf8-selftest.c (6ca99ce756c27852d1ea1e555045de1c920f30ed)
1// SPDX-License-Identifier: GPL-2.0-only
2/*
3 * Kernel module for testing utf-8 support.
4 *
5 * Copyright 2017 Collabora Ltd.
6 */
7
8#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

--- 4 unchanged lines hidden (view full) ---

13#include <linux/dcache.h>
14
15#include "utf8n.h"
16
17unsigned int failed_tests;
18unsigned int total_tests;
19
20/* Tests will be based on this version. */
1// SPDX-License-Identifier: GPL-2.0-only
2/*
3 * Kernel module for testing utf-8 support.
4 *
5 * Copyright 2017 Collabora Ltd.
6 */
7
8#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

--- 4 unchanged lines hidden (view full) ---

13#include <linux/dcache.h>
14
15#include "utf8n.h"
16
17unsigned int failed_tests;
18unsigned int total_tests;
19
20/* Tests will be based on this version. */
21#define latest_maj 12
22#define latest_min 1
23#define latest_rev 0
21#define UTF8_LATEST UNICODE_AGE(12, 1, 0)
24
25#define _test(cond, func, line, fmt, ...) do { \
26 total_tests++; \
27 if (!cond) { \
28 failed_tests++; \
29 pr_err("test %s:%d Failed: %s%s", \
30 func, line, #cond, (fmt?":":".")); \
31 if (fmt) \

--- 123 unchanged lines hidden (view full) ---

155 {
156 /* GEORGIAN SMALL LETTER AN folds to GEORGIAN MTAVRULI
157 CAPITAL LETTER AN */
158 .str = {0xe1, 0xb2, 0x90, 0x00},
159 .ncf = {0xe1, 0x83, 0x90, 0x00},
160 }
161};
162
22
23#define _test(cond, func, line, fmt, ...) do { \
24 total_tests++; \
25 if (!cond) { \
26 failed_tests++; \
27 pr_err("test %s:%d Failed: %s%s", \
28 func, line, #cond, (fmt?":":".")); \
29 if (fmt) \

--- 123 unchanged lines hidden (view full) ---

153 {
154 /* GEORGIAN SMALL LETTER AN folds to GEORGIAN MTAVRULI
155 CAPITAL LETTER AN */
156 .str = {0xe1, 0xb2, 0x90, 0x00},
157 .ncf = {0xe1, 0x83, 0x90, 0x00},
158 }
159};
160
163static ssize_t utf8len(const struct utf8data *data, const char *s)
161static ssize_t utf8len(const struct unicode_map *um, enum utf8_normalization n,
162 const char *s)
164{
163{
165 return utf8nlen(data, s, (size_t)-1);
164 return utf8nlen(um, n, s, (size_t)-1);
166}
167
165}
166
168static int utf8cursor(struct utf8cursor *u8c, const struct utf8data *data,
169 const char *s)
167static int utf8cursor(struct utf8cursor *u8c, const struct unicode_map *um,
168 enum utf8_normalization n, const char *s)
170{
169{
171 return utf8ncursor(u8c, data, s, (unsigned int)-1);
170 return utf8ncursor(u8c, um, n, s, (unsigned int)-1);
172}
173
171}
172
174static void check_utf8_nfdi(void)
173static void check_utf8_nfdi(struct unicode_map *um)
175{
176 int i;
177 struct utf8cursor u8c;
174{
175 int i;
176 struct utf8cursor u8c;
178 const struct utf8data *data;
179
177
180 data = utf8nfdi(UNICODE_AGE(latest_maj, latest_min, latest_rev));
181 if (!data) {
182 pr_err("%s: Unable to load utf8-%d.%d.%d. Skipping.\n",
183 __func__, latest_maj, latest_min, latest_rev);
184 return;
185 }
186
187 for (i = 0; i < ARRAY_SIZE(nfdi_test_data); i++) {
188 int len = strlen(nfdi_test_data[i].str);
189 int nlen = strlen(nfdi_test_data[i].dec);
190 int j = 0;
191 unsigned char c;
192
178 for (i = 0; i < ARRAY_SIZE(nfdi_test_data); i++) {
179 int len = strlen(nfdi_test_data[i].str);
180 int nlen = strlen(nfdi_test_data[i].dec);
181 int j = 0;
182 unsigned char c;
183
193 test((utf8len(data, nfdi_test_data[i].str) == nlen));
194 test((utf8nlen(data, nfdi_test_data[i].str, len) == nlen));
184 test((utf8len(um, UTF8_NFDI, nfdi_test_data[i].str) == nlen));
185 test((utf8nlen(um, UTF8_NFDI, nfdi_test_data[i].str, len) ==
186 nlen));
195
187
196 if (utf8cursor(&u8c, data, nfdi_test_data[i].str) < 0)
188 if (utf8cursor(&u8c, um, UTF8_NFDI, nfdi_test_data[i].str) < 0)
197 pr_err("can't create cursor\n");
198
199 while ((c = utf8byte(&u8c)) > 0) {
200 test_f((c == nfdi_test_data[i].dec[j]),
201 "Unexpected byte 0x%x should be 0x%x\n",
202 c, nfdi_test_data[i].dec[j]);
203 j++;
204 }
205
206 test((j == nlen));
207 }
208}
209
189 pr_err("can't create cursor\n");
190
191 while ((c = utf8byte(&u8c)) > 0) {
192 test_f((c == nfdi_test_data[i].dec[j]),
193 "Unexpected byte 0x%x should be 0x%x\n",
194 c, nfdi_test_data[i].dec[j]);
195 j++;
196 }
197
198 test((j == nlen));
199 }
200}
201
210static void check_utf8_nfdicf(void)
202static void check_utf8_nfdicf(struct unicode_map *um)
211{
212 int i;
213 struct utf8cursor u8c;
203{
204 int i;
205 struct utf8cursor u8c;
214 const struct utf8data *data;
215
206
216 data = utf8nfdicf(UNICODE_AGE(latest_maj, latest_min, latest_rev));
217 if (!data) {
218 pr_err("%s: Unable to load utf8-%d.%d.%d. Skipping.\n",
219 __func__, latest_maj, latest_min, latest_rev);
220 return;
221 }
222
223 for (i = 0; i < ARRAY_SIZE(nfdicf_test_data); i++) {
224 int len = strlen(nfdicf_test_data[i].str);
225 int nlen = strlen(nfdicf_test_data[i].ncf);
226 int j = 0;
227 unsigned char c;
228
207 for (i = 0; i < ARRAY_SIZE(nfdicf_test_data); i++) {
208 int len = strlen(nfdicf_test_data[i].str);
209 int nlen = strlen(nfdicf_test_data[i].ncf);
210 int j = 0;
211 unsigned char c;
212
229 test((utf8len(data, nfdicf_test_data[i].str) == nlen));
230 test((utf8nlen(data, nfdicf_test_data[i].str, len) == nlen));
213 test((utf8len(um, UTF8_NFDICF, nfdicf_test_data[i].str) ==
214 nlen));
215 test((utf8nlen(um, UTF8_NFDICF, nfdicf_test_data[i].str, len) ==
216 nlen));
231
217
232 if (utf8cursor(&u8c, data, nfdicf_test_data[i].str) < 0)
218 if (utf8cursor(&u8c, um, UTF8_NFDICF,
219 nfdicf_test_data[i].str) < 0)
233 pr_err("can't create cursor\n");
234
235 while ((c = utf8byte(&u8c)) > 0) {
236 test_f((c == nfdicf_test_data[i].ncf[j]),
237 "Unexpected byte 0x%x should be 0x%x\n",
238 c, nfdicf_test_data[i].ncf[j]);
239 j++;
240 }
241
242 test((j == nlen));
243 }
244}
245
220 pr_err("can't create cursor\n");
221
222 while ((c = utf8byte(&u8c)) > 0) {
223 test_f((c == nfdicf_test_data[i].ncf[j]),
224 "Unexpected byte 0x%x should be 0x%x\n",
225 c, nfdicf_test_data[i].ncf[j]);
226 j++;
227 }
228
229 test((j == nlen));
230 }
231}
232
246static void check_utf8_comparisons(void)
233static void check_utf8_comparisons(struct unicode_map *table)
247{
248 int i;
234{
235 int i;
249 struct unicode_map *table = utf8_load(UNICODE_AGE(12, 1, 0));
250
236
251 if (IS_ERR(table)) {
252 pr_err("%s: Unable to load utf8 %d.%d.%d. Skipping.\n",
253 __func__, latest_maj, latest_min, latest_rev);
254 return;
255 }
256
257 for (i = 0; i < ARRAY_SIZE(nfdi_test_data); i++) {
258 const struct qstr s1 = {.name = nfdi_test_data[i].str,
259 .len = sizeof(nfdi_test_data[i].str)};
260 const struct qstr s2 = {.name = nfdi_test_data[i].dec,
261 .len = sizeof(nfdi_test_data[i].dec)};
262
263 test_f(!utf8_strncmp(table, &s1, &s2),
264 "%s %s comparison mismatch\n", s1.name, s2.name);
265 }
266
267 for (i = 0; i < ARRAY_SIZE(nfdicf_test_data); i++) {
268 const struct qstr s1 = {.name = nfdicf_test_data[i].str,
269 .len = sizeof(nfdicf_test_data[i].str)};
270 const struct qstr s2 = {.name = nfdicf_test_data[i].ncf,
271 .len = sizeof(nfdicf_test_data[i].ncf)};
272
273 test_f(!utf8_strncasecmp(table, &s1, &s2),
274 "%s %s comparison mismatch\n", s1.name, s2.name);
275 }
237 for (i = 0; i < ARRAY_SIZE(nfdi_test_data); i++) {
238 const struct qstr s1 = {.name = nfdi_test_data[i].str,
239 .len = sizeof(nfdi_test_data[i].str)};
240 const struct qstr s2 = {.name = nfdi_test_data[i].dec,
241 .len = sizeof(nfdi_test_data[i].dec)};
242
243 test_f(!utf8_strncmp(table, &s1, &s2),
244 "%s %s comparison mismatch\n", s1.name, s2.name);
245 }
246
247 for (i = 0; i < ARRAY_SIZE(nfdicf_test_data); i++) {
248 const struct qstr s1 = {.name = nfdicf_test_data[i].str,
249 .len = sizeof(nfdicf_test_data[i].str)};
250 const struct qstr s2 = {.name = nfdicf_test_data[i].ncf,
251 .len = sizeof(nfdicf_test_data[i].ncf)};
252
253 test_f(!utf8_strncasecmp(table, &s1, &s2),
254 "%s %s comparison mismatch\n", s1.name, s2.name);
255 }
276
277 utf8_unload(table);
278}
279
280static void check_supported_versions(void)
281{
282 /* Unicode 7.0.0 should be supported. */
283 test(utf8version_is_supported(UNICODE_AGE(7, 0, 0)));
284
285 /* Unicode 9.0.0 should be supported. */
286 test(utf8version_is_supported(UNICODE_AGE(9, 0, 0)));
287
288 /* Unicode 1x.0.0 (the latest version) should be supported. */
256}
257
258static void check_supported_versions(void)
259{
260 /* Unicode 7.0.0 should be supported. */
261 test(utf8version_is_supported(UNICODE_AGE(7, 0, 0)));
262
263 /* Unicode 9.0.0 should be supported. */
264 test(utf8version_is_supported(UNICODE_AGE(9, 0, 0)));
265
266 /* Unicode 1x.0.0 (the latest version) should be supported. */
289 test(utf8version_is_supported(
290 UNICODE_AGE(latest_maj, latest_min, latest_rev)));
267 test(utf8version_is_supported(UTF8_LATEST));
291
292 /* Next versions don't exist. */
293 test(!utf8version_is_supported(UNICODE_AGE(13, 0, 0)));
294 test(!utf8version_is_supported(UNICODE_AGE(0, 0, 0)));
295 test(!utf8version_is_supported(UNICODE_AGE(-1, -1, -1)));
296}
297
298static int __init init_test_ucd(void)
299{
268
269 /* Next versions don't exist. */
270 test(!utf8version_is_supported(UNICODE_AGE(13, 0, 0)));
271 test(!utf8version_is_supported(UNICODE_AGE(0, 0, 0)));
272 test(!utf8version_is_supported(UNICODE_AGE(-1, -1, -1)));
273}
274
275static int __init init_test_ucd(void)
276{
277 struct unicode_map *um;
278
300 failed_tests = 0;
301 total_tests = 0;
302
279 failed_tests = 0;
280 total_tests = 0;
281
282 um = utf8_load(UTF8_LATEST);
283 if (IS_ERR(um)) {
284 pr_err("%s: Unable to load utf8 table.\n", __func__);
285 return PTR_ERR(um);
286 }
287
303 check_supported_versions();
288 check_supported_versions();
304 check_utf8_nfdi();
305 check_utf8_nfdicf();
306 check_utf8_comparisons();
289 check_utf8_nfdi(um);
290 check_utf8_nfdicf(um);
291 check_utf8_comparisons(um);
307
308 if (!failed_tests)
309 pr_info("All %u tests passed\n", total_tests);
310 else
311 pr_err("%u out of %u tests failed\n", failed_tests,
312 total_tests);
292
293 if (!failed_tests)
294 pr_info("All %u tests passed\n", total_tests);
295 else
296 pr_err("%u out of %u tests failed\n", failed_tests,
297 total_tests);
298 utf8_unload(um);
313 return 0;
314}
315
316static void __exit exit_test_ucd(void)
317{
318}
319
320module_init(init_test_ucd);
321module_exit(exit_test_ucd);
322
323MODULE_AUTHOR("Gabriel Krisman Bertazi <krisman@collabora.co.uk>");
324MODULE_LICENSE("GPL");
299 return 0;
300}
301
302static void __exit exit_test_ucd(void)
303{
304}
305
306module_init(init_test_ucd);
307module_exit(exit_test_ucd);
308
309MODULE_AUTHOR("Gabriel Krisman Bertazi <krisman@collabora.co.uk>");
310MODULE_LICENSE("GPL");