~mdw / sgt / charset / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
Bring utf8.c's internal tests up to date in the (somewhat belated)
[sgt/charset] / utf8.c
1 /*
2 * utf8.c - routines to handle UTF-8.
3 */
4
5 #ifndef ENUM_CHARSETS
6
7 #include "charset.h"
8 #include "internal.h"
9
10 /*
11 * UTF-8 has no associated data, so `charset' may be ignored.
12 */
13
14 static void read_utf8(charset_spec const *charset, long int input_chr,
15 charset_state *state,
16 void (*emit)(void *ctx, long int output), void *emitctx)
17 {
18 UNUSEDARG(charset);
19
20 /*
21 * For reading UTF-8, the `state' word contains:
22 *
23 * - in bits 29-31, the number of bytes expected to be in the
24 * current multibyte character (which we can tell instantly
25 * from the first byte, of course).
26 *
27 * - in bits 26-28, the number of bytes _seen so far_ in the
28 * current multibyte character.
29 *
30 * - in the remainder of the word, the current value of the
31 * character, which is shifted upwards by 6 bits to
32 * accommodate each new byte.
33 *
34 * As required, the state is zero when we are not in the middle
35 * of a multibyte character at all.
36 *
37 * For example, when reading E9 8D 8B, starting at state=0:
38 *
39 * - after E9, the state is 0x64000009
40 * - after 8D, the state is 0x6800024d
41 * - after 8B, the state conceptually becomes 0x6c00934b, at
42 * which point we notice we've got as many characters as we
43 * were expecting, output U+934B, and reset the state to
44 * zero.
45 *
46 * Note that the maximum number of bits we might need to store
47 * in the character value field is 25 (U+7FFFFFFF contains 31
48 * bits, but we will never actually store its full value
49 * because when we receive the last 6 bits in the final
50 * continuation byte we will output it and revert the state to
51 * zero). Hence the character value field never collides with
52 * the byte counts.
53 */
54
55 if (input_chr < 0x80) {
56 /*
57 * Single-byte character. If the state is nonzero before
58 * coming here, output an error for an incomplete sequence.
59 * Then output the character.
60 */
61 if (state->s0 != 0) {
62 emit(emitctx, ERROR);
63 state->s0 = 0;
64 }
65 emit(emitctx, input_chr);
66 } else if (input_chr == 0xFE || input_chr == 0xFF) {
67 /*
68 * FE and FF bytes should _never_ occur in UTF-8. They are
69 * automatic errors; if the state was nonzero to start
70 * with, output a further error for an incomplete sequence.
71 */
72 if (state->s0 != 0) {
73 emit(emitctx, ERROR);
74 state->s0 = 0;
75 }
76 emit(emitctx, ERROR);
77 } else if (input_chr >= 0x80 && input_chr < 0xC0) {
78 /*
79 * Continuation byte. Output an error for an unexpected
80 * continuation byte, if the state is zero.
81 */
82 if (state->s0 == 0) {
83 emit(emitctx, ERROR);
84 } else {
85 unsigned long charval;
86 unsigned long topstuff;
87 int bytes;
88
89 /*
90 * Otherwise, accumulate more of the character value.
91 */
92 charval = state->s0 & 0x03ffffffL;
93 charval = (charval << 6) | (input_chr & 0x3F);
94
95 /*
96 * Check the byte counts; if we have not reached the
97 * end of the character, update the state and return.
98 */
99 topstuff = state->s0 & 0xfc000000L;
100 topstuff += 0x04000000L; /* add one to the byte count */
101 if (((topstuff << 3) ^ topstuff) & 0xe0000000L) {
102 state->s0 = topstuff | charval;
103 return;
104 }
105
106 /*
107 * Now we know we've reached the end of the character.
108 * `charval' is the Unicode value. We should check for
109 * various invalid things, and then either output
110 * charval or an error. In all cases we reset the state
111 * to zero.
112 */
113 bytes = topstuff >> 29;
114 state->s0 = 0;
115
116 if (charval >= 0xD800 && charval < 0xE000) {
117 /*
118 * Surrogates (0xD800-0xDFFF) may never be encoded
119 * in UTF-8. A surrogate pair in Unicode should
120 * have been encoded as a single UTF-8 character
121 * occupying more than three bytes.
122 */
123 emit(emitctx, ERROR);
124 } else if (charval == 0xFFFE || charval == 0xFFFF) {
125 /*
126 * U+FFFE and U+FFFF are invalid Unicode characters
127 * and may never be encoded in UTF-8. (This is one
128 * reason why U+FFFF is our way of signalling an
129 * error to our `emit' function :-)
130 */
131 emit(emitctx, ERROR);
132 } else if ((charval <= 0x7FL /* && bytes > 1 */) ||
133 (charval <= 0x7FFL && bytes > 2) ||
134 (charval <= 0xFFFFL && bytes > 3) ||
135 (charval <= 0x1FFFFFL && bytes > 4) ||
136 (charval <= 0x3FFFFFFL && bytes > 5)) {
137 /*
138 * Overlong sequences are not to be tolerated,
139 * under any circumstances.
140 */
141 emit(emitctx, ERROR);
142 } else {
143 /*
144 * Oh, all right. We'll let this one off.
145 */
146 emit(emitctx, charval);
147 }
148 }
149
150 } else {
151 /*
152 * Lead byte. First output an error for an incomplete
153 * sequence, if the state is nonzero.
154 */
155 if (state->s0 != 0)
156 emit(emitctx, ERROR);
157
158 /*
159 * Now deal with the lead byte: work out the number of
160 * bytes we expect to see in this character, and extract
161 * the initial bits of it too.
162 */
163 if (input_chr >= 0xC0 && input_chr < 0xE0) {
164 state->s0 = 0x44000000L | (input_chr & 0x1F);
165 } else if (input_chr >= 0xE0 && input_chr < 0xF0) {
166 state->s0 = 0x64000000L | (input_chr & 0x0F);
167 } else if (input_chr >= 0xF0 && input_chr < 0xF8) {
168 state->s0 = 0x84000000L | (input_chr & 0x07);
169 } else if (input_chr >= 0xF8 && input_chr < 0xFC) {
170 state->s0 = 0xa4000000L | (input_chr & 0x03);
171 } else if (input_chr >= 0xFC && input_chr < 0xFE) {
172 state->s0 = 0xc4000000L | (input_chr & 0x01);
173 }
174 }
175 }
176
177 /*
178 * UTF-8 is a stateless multi-byte encoding (in the sense that just
179 * after any character has been completed, the state is always the
180 * same); hence when writing it, there is no need to use the
181 * charset_state.
182 */
183
184 static int write_utf8(charset_spec const *charset, long int input_chr,
185 charset_state *state,
186 void (*emit)(void *ctx, long int output),
187 void *emitctx)
188 {
189 UNUSEDARG(charset);
190 UNUSEDARG(state);
191
192 if (input_chr == -1)
193 return TRUE; /* stateless; no cleanup required */
194
195 /*
196 * Refuse to output any illegal code points.
197 */
198 if (input_chr == 0xFFFE || input_chr == 0xFFFF ||
199 (input_chr >= 0xD800 && input_chr < 0xE000)) {
200 return FALSE;
201 } else if (input_chr < 0x80) { /* one-byte character */
202 emit(emitctx, input_chr);
203 return TRUE;
204 } else if (input_chr < 0x800) { /* two-byte character */
205 emit(emitctx, 0xC0 | (0x1F & (input_chr >> 6)));
206 emit(emitctx, 0x80 | (0x3F & (input_chr )));
207 return TRUE;
208 } else if (input_chr < 0x10000) { /* three-byte character */
209 emit(emitctx, 0xE0 | (0x0F & (input_chr >> 12)));
210 emit(emitctx, 0x80 | (0x3F & (input_chr >> 6)));
211 emit(emitctx, 0x80 | (0x3F & (input_chr )));
212 return TRUE;
213 } else if (input_chr < 0x200000) { /* four-byte character */
214 emit(emitctx, 0xF0 | (0x07 & (input_chr >> 18)));
215 emit(emitctx, 0x80 | (0x3F & (input_chr >> 12)));
216 emit(emitctx, 0x80 | (0x3F & (input_chr >> 6)));
217 emit(emitctx, 0x80 | (0x3F & (input_chr )));
218 return TRUE;
219 } else if (input_chr < 0x4000000) {/* five-byte character */
220 emit(emitctx, 0xF8 | (0x03 & (input_chr >> 24)));
221 emit(emitctx, 0x80 | (0x3F & (input_chr >> 18)));
222 emit(emitctx, 0x80 | (0x3F & (input_chr >> 12)));
223 emit(emitctx, 0x80 | (0x3F & (input_chr >> 6)));
224 emit(emitctx, 0x80 | (0x3F & (input_chr )));
225 return TRUE;
226 } else { /* six-byte character */
227 emit(emitctx, 0xFC | (0x01 & (input_chr >> 30)));
228 emit(emitctx, 0x80 | (0x3F & (input_chr >> 24)));
229 emit(emitctx, 0x80 | (0x3F & (input_chr >> 18)));
230 emit(emitctx, 0x80 | (0x3F & (input_chr >> 12)));
231 emit(emitctx, 0x80 | (0x3F & (input_chr >> 6)));
232 emit(emitctx, 0x80 | (0x3F & (input_chr )));
233 return TRUE;
234 }
235 }
236
237 #ifdef TESTMODE
238
239 #include <stdio.h>
240 #include <stdarg.h>
241
242 int total_errs = 0;
243
244 void utf8_emit(void *ctx, long output)
245 {
246 wchar_t **p = (wchar_t **)ctx;
247 *(*p)++ = output;
248 }
249
250 void utf8_read_test(int line, char *input, int inlen, ...)
251 {
252 va_list ap;
253 wchar_t *p, str[512];
254 int i;
255 charset_state state;
256 unsigned long l;
257
258 state.s0 = 0;
259 p = str;
260
261 for (i = 0; i < inlen; i++)
262 read_utf8(NULL, input[i] & 0xFF, &state, utf8_emit, &p);
263
264 va_start(ap, inlen);
265 l = 0;
266 for (i = 0; i < p - str; i++) {
267 l = va_arg(ap, long int);
268 if (l == -1) {
269 printf("%d: correct string shorter than output\n", line);
270 total_errs++;
271 break;
272 }
273 if (l != str[i]) {
274 printf("%d: char %d came out as %08x, should be %08x\n",
275 line, i, str[i], l);
276 total_errs++;
277 }
278 }
279 if (l != -1) {
280 l = va_arg(ap, long int);
281 if (l != -1) {
282 printf("%d: correct string longer than output\n", line);
283 total_errs++;
284 }
285 }
286 va_end(ap);
287 }
288
289 void utf8_write_test(int line, const long *input, int inlen, ...)
290 {
291 va_list ap;
292 wchar_t *p, str[512];
293 int i;
294 charset_state state;
295 unsigned long l;
296
297 state.s0 = 0;
298 p = str;
299
300 for (i = 0; i < inlen; i++) {
301 if (!write_utf8(NULL, input[i], &state, utf8_emit, &p))
302 utf8_emit(&p, ERROR);
303 }
304
305 va_start(ap, inlen);
306 l = 0;
307 for (i = 0; i < p - str; i++) {
308 l = va_arg(ap, long int);
309 if (l == -1) {
310 printf("%d: correct string shorter than output\n", line);
311 total_errs++;
312 break;
313 }
314 if (l != str[i]) {
315 printf("%d: char %d came out as %08x, should be %08x\n",
316 line, i, str[i], l);
317 total_errs++;
318 }
319 }
320 if (l != -1) {
321 l = va_arg(ap, long int);
322 if (l != -1) {
323 printf("%d: correct string longer than output\n", line);
324 total_errs++;
325 }
326 }
327 va_end(ap);
328 }
329
330 /* Macro to concoct the first three parameters of utf8_read_test. */
331 #define TESTSTR(x) __LINE__, x, lenof(x)
332
333 int main(void)
334 {
335 printf("read tests beginning\n");
336 utf8_read_test(TESTSTR("\xCE\xBA\xE1\xBD\xB9\xCF\x83\xCE\xBC\xCE\xB5"),
337 0x000003BA, /* GREEK SMALL LETTER KAPPA */
338 0x00001F79, /* GREEK SMALL LETTER OMICRON WITH OXIA */
339 0x000003C3, /* GREEK SMALL LETTER SIGMA */
340 0x000003BC, /* GREEK SMALL LETTER MU */
341 0x000003B5, /* GREEK SMALL LETTER EPSILON */
342 0, -1);
343 utf8_read_test(TESTSTR("\x00"),
344 0x00000000, /* <control> */
345 0, -1);
346 utf8_read_test(TESTSTR("\xC2\x80"),
347 0x00000080, /* <control> */
348 0, -1);
349 utf8_read_test(TESTSTR("\xE0\xA0\x80"),
350 0x00000800, /* <no name available> */
351 0, -1);
352 utf8_read_test(TESTSTR("\xF0\x90\x80\x80"),
353 0x00010000, /* <no name available> */
354 0, -1);
355 utf8_read_test(TESTSTR("\xF8\x88\x80\x80\x80"),
356 0x00200000, /* <no name available> */
357 0, -1);
358 utf8_read_test(TESTSTR("\xFC\x84\x80\x80\x80\x80"),
359 0x04000000, /* <no name available> */
360 0, -1);
361 utf8_read_test(TESTSTR("\x7F"),
362 0x0000007F, /* <control> */
363 0, -1);
364 utf8_read_test(TESTSTR("\xDF\xBF"),
365 0x000007FF, /* <no name available> */
366 0, -1);
367 utf8_read_test(TESTSTR("\xEF\xBF\xBD"),
368 0x0000FFFD, /* REPLACEMENT CHARACTER */
369 0, -1);
370 utf8_read_test(TESTSTR("\xEF\xBF\xBF"),
371 ERROR, /* <no name available> (invalid char) */
372 0, -1);
373 utf8_read_test(TESTSTR("\xF7\xBF\xBF\xBF"),
374 0x001FFFFF, /* <no name available> */
375 0, -1);
376 utf8_read_test(TESTSTR("\xFB\xBF\xBF\xBF\xBF"),
377 0x03FFFFFF, /* <no name available> */
378 0, -1);
379 utf8_read_test(TESTSTR("\xFD\xBF\xBF\xBF\xBF\xBF"),
380 0x7FFFFFFF, /* <no name available> */
381 0, -1);
382 utf8_read_test(TESTSTR("\xED\x9F\xBF"),
383 0x0000D7FF, /* <no name available> */
384 0, -1);
385 utf8_read_test(TESTSTR("\xEE\x80\x80"),
386 0x0000E000, /* <Private Use, First> */
387 0, -1);
388 utf8_read_test(TESTSTR("\xEF\xBF\xBD"),
389 0x0000FFFD, /* REPLACEMENT CHARACTER */
390 0, -1);
391 utf8_read_test(TESTSTR("\xF4\x8F\xBF\xBF"),
392 0x0010FFFF, /* <no name available> */
393 0, -1);
394 utf8_read_test(TESTSTR("\xF4\x90\x80\x80"),
395 0x00110000, /* <no name available> */
396 0, -1);
397 utf8_read_test(TESTSTR("\x80"),
398 ERROR, /* (unexpected continuation byte) */
399 0, -1);
400 utf8_read_test(TESTSTR("\xBF"),
401 ERROR, /* (unexpected continuation byte) */
402 0, -1);
403 utf8_read_test(TESTSTR("\x80\xBF"),
404 ERROR, /* (unexpected continuation byte) */
405 ERROR, /* (unexpected continuation byte) */
406 0, -1);
407 utf8_read_test(TESTSTR("\x80\xBF\x80"),
408 ERROR, /* (unexpected continuation byte) */
409 ERROR, /* (unexpected continuation byte) */
410 ERROR, /* (unexpected continuation byte) */
411 0, -1);
412 utf8_read_test(TESTSTR("\x80\xBF\x80\xBF"),
413 ERROR, /* (unexpected continuation byte) */
414 ERROR, /* (unexpected continuation byte) */
415 ERROR, /* (unexpected continuation byte) */
416 ERROR, /* (unexpected continuation byte) */
417 0, -1);
418 utf8_read_test(TESTSTR("\x80\xBF\x80\xBF\x80"),
419 ERROR, /* (unexpected continuation byte) */
420 ERROR, /* (unexpected continuation byte) */
421 ERROR, /* (unexpected continuation byte) */
422 ERROR, /* (unexpected continuation byte) */
423 ERROR, /* (unexpected continuation byte) */
424 0, -1);
425 utf8_read_test(TESTSTR("\x80\xBF\x80\xBF\x80\xBF"),
426 ERROR, /* (unexpected continuation byte) */
427 ERROR, /* (unexpected continuation byte) */
428 ERROR, /* (unexpected continuation byte) */
429 ERROR, /* (unexpected continuation byte) */
430 ERROR, /* (unexpected continuation byte) */
431 ERROR, /* (unexpected continuation byte) */
432 0, -1);
433 utf8_read_test(TESTSTR("\x80\xBF\x80\xBF\x80\xBF\x80"),
434 ERROR, /* (unexpected continuation byte) */
435 ERROR, /* (unexpected continuation byte) */
436 ERROR, /* (unexpected continuation byte) */
437 ERROR, /* (unexpected continuation byte) */
438 ERROR, /* (unexpected continuation byte) */
439 ERROR, /* (unexpected continuation byte) */
440 ERROR, /* (unexpected continuation byte) */
441 0, -1);
442 utf8_read_test(TESTSTR("\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8A\x8B\x8C\x8D\x8E\x8F\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9A\x9B\x9C\x9D\x9E\x9F\xA0\xA1\xA2\xA3\xA4\xA5\xA6\xA7\xA8\xA9\xAA\xAB\xAC\xAD\xAE\xAF\xB0\xB1\xB2\xB3\xB4\xB5\xB6\xB7\xB8\xB9\xBA\xBB\xBC\xBD\xBE\xBF"),
443 ERROR, /* (unexpected continuation byte) */
444 ERROR, /* (unexpected continuation byte) */
445 ERROR, /* (unexpected continuation byte) */
446 ERROR, /* (unexpected continuation byte) */
447 ERROR, /* (unexpected continuation byte) */
448 ERROR, /* (unexpected continuation byte) */
449 ERROR, /* (unexpected continuation byte) */
450 ERROR, /* (unexpected continuation byte) */
451 ERROR, /* (unexpected continuation byte) */
452 ERROR, /* (unexpected continuation byte) */
453 ERROR, /* (unexpected continuation byte) */
454 ERROR, /* (unexpected continuation byte) */
455 ERROR, /* (unexpected continuation byte) */
456 ERROR, /* (unexpected continuation byte) */
457 ERROR, /* (unexpected continuation byte) */
458 ERROR, /* (unexpected continuation byte) */
459 ERROR, /* (unexpected continuation byte) */
460 ERROR, /* (unexpected continuation byte) */
461 ERROR, /* (unexpected continuation byte) */
462 ERROR, /* (unexpected continuation byte) */
463 ERROR, /* (unexpected continuation byte) */
464 ERROR, /* (unexpected continuation byte) */
465 ERROR, /* (unexpected continuation byte) */
466 ERROR, /* (unexpected continuation byte) */
467 ERROR, /* (unexpected continuation byte) */
468 ERROR, /* (unexpected continuation byte) */
469 ERROR, /* (unexpected continuation byte) */
470 ERROR, /* (unexpected continuation byte) */
471 ERROR, /* (unexpected continuation byte) */
472 ERROR, /* (unexpected continuation byte) */
473 ERROR, /* (unexpected continuation byte) */
474 ERROR, /* (unexpected continuation byte) */
475 ERROR, /* (unexpected continuation byte) */
476 ERROR, /* (unexpected continuation byte) */
477 ERROR, /* (unexpected continuation byte) */
478 ERROR, /* (unexpected continuation byte) */
479 ERROR, /* (unexpected continuation byte) */
480 ERROR, /* (unexpected continuation byte) */
481 ERROR, /* (unexpected continuation byte) */
482 ERROR, /* (unexpected continuation byte) */
483 ERROR, /* (unexpected continuation byte) */
484 ERROR, /* (unexpected continuation byte) */
485 ERROR, /* (unexpected continuation byte) */
486 ERROR, /* (unexpected continuation byte) */
487 ERROR, /* (unexpected continuation byte) */
488 ERROR, /* (unexpected continuation byte) */
489 ERROR, /* (unexpected continuation byte) */
490 ERROR, /* (unexpected continuation byte) */
491 ERROR, /* (unexpected continuation byte) */
492 ERROR, /* (unexpected continuation byte) */
493 ERROR, /* (unexpected continuation byte) */
494 ERROR, /* (unexpected continuation byte) */
495 ERROR, /* (unexpected continuation byte) */
496 ERROR, /* (unexpected continuation byte) */
497 ERROR, /* (unexpected continuation byte) */
498 ERROR, /* (unexpected continuation byte) */
499 ERROR, /* (unexpected continuation byte) */
500 ERROR, /* (unexpected continuation byte) */
501 ERROR, /* (unexpected continuation byte) */
502 ERROR, /* (unexpected continuation byte) */
503 ERROR, /* (unexpected continuation byte) */
504 ERROR, /* (unexpected continuation byte) */
505 ERROR, /* (unexpected continuation byte) */
506 ERROR, /* (unexpected continuation byte) */
507 0, -1);
508 utf8_read_test(TESTSTR("\xC0\x20\xC1\x20\xC2\x20\xC3\x20\xC4\x20\xC5\x20\xC6\x20\xC7\x20"),
509 ERROR, /* (incomplete sequence) */
510 0x00000020, /* SPACE */
511 ERROR, /* (incomplete sequence) */
512 0x00000020, /* SPACE */
513 ERROR, /* (incomplete sequence) */
514 0x00000020, /* SPACE */
515 ERROR, /* (incomplete sequence) */
516 0x00000020, /* SPACE */
517 ERROR, /* (incomplete sequence) */
518 0x00000020, /* SPACE */
519 ERROR, /* (incomplete sequence) */
520 0x00000020, /* SPACE */
521 ERROR, /* (incomplete sequence) */
522 0x00000020, /* SPACE */
523 ERROR, /* (incomplete sequence) */
524 0x00000020, /* SPACE */
525 0, -1);
526 utf8_read_test(TESTSTR("\xE0\x20\xE1\x20\xE2\x20\xE3\x20\xE4\x20\xE5\x20\xE6\x20\xE7\x20\xE8\x20\xE9\x20\xEA\x20\xEB\x20\xEC\x20\xED\x20\xEE\x20\xEF\x20"),
527 ERROR, /* (incomplete sequence) */
528 0x00000020, /* SPACE */
529 ERROR, /* (incomplete sequence) */
530 0x00000020, /* SPACE */
531 ERROR, /* (incomplete sequence) */
532 0x00000020, /* SPACE */
533 ERROR, /* (incomplete sequence) */
534 0x00000020, /* SPACE */
535 ERROR, /* (incomplete sequence) */
536 0x00000020, /* SPACE */
537 ERROR, /* (incomplete sequence) */
538 0x00000020, /* SPACE */
539 ERROR, /* (incomplete sequence) */
540 0x00000020, /* SPACE */
541 ERROR, /* (incomplete sequence) */
542 0x00000020, /* SPACE */
543 ERROR, /* (incomplete sequence) */
544 0x00000020, /* SPACE */
545 ERROR, /* (incomplete sequence) */
546 0x00000020, /* SPACE */
547 ERROR, /* (incomplete sequence) */
548 0x00000020, /* SPACE */
549 ERROR, /* (incomplete sequence) */
550 0x00000020, /* SPACE */
551 ERROR, /* (incomplete sequence) */
552 0x00000020, /* SPACE */
553 ERROR, /* (incomplete sequence) */
554 0x00000020, /* SPACE */
555 ERROR, /* (incomplete sequence) */
556 0x00000020, /* SPACE */
557 ERROR, /* (incomplete sequence) */
558 0x00000020, /* SPACE */
559 0, -1);
560 utf8_read_test(TESTSTR("\xF0\x20\xF1\x20\xF2\x20\xF3\x20\xF4\x20\xF5\x20\xF6\x20\xF7\x20"),
561 ERROR, /* (incomplete sequence) */
562 0x00000020, /* SPACE */
563 ERROR, /* (incomplete sequence) */
564 0x00000020, /* SPACE */
565 ERROR, /* (incomplete sequence) */
566 0x00000020, /* SPACE */
567 ERROR, /* (incomplete sequence) */
568 0x00000020, /* SPACE */
569 ERROR, /* (incomplete sequence) */
570 0x00000020, /* SPACE */
571 ERROR, /* (incomplete sequence) */
572 0x00000020, /* SPACE */
573 ERROR, /* (incomplete sequence) */
574 0x00000020, /* SPACE */
575 ERROR, /* (incomplete sequence) */
576 0x00000020, /* SPACE */
577 0, -1);
578 utf8_read_test(TESTSTR("\xF8\x20\xF9\x20\xFA\x20\xFB\x20"),
579 ERROR, /* (incomplete sequence) */
580 0x00000020, /* SPACE */
581 ERROR, /* (incomplete sequence) */
582 0x00000020, /* SPACE */
583 ERROR, /* (incomplete sequence) */
584 0x00000020, /* SPACE */
585 ERROR, /* (incomplete sequence) */
586 0x00000020, /* SPACE */
587 0, -1);
588 utf8_read_test(TESTSTR("\xFC\x20\xFD\x20"),
589 ERROR, /* (incomplete sequence) */
590 0x00000020, /* SPACE */
591 ERROR, /* (incomplete sequence) */
592 0x00000020, /* SPACE */
593 0, -1);
594 utf8_read_test(TESTSTR("\xC0"),
595 ERROR, /* (incomplete sequence) */
596 0, -1);
597 utf8_read_test(TESTSTR("\xE0\x80"),
598 ERROR, /* (incomplete sequence) */
599 0, -1);
600 utf8_read_test(TESTSTR("\xF0\x80\x80"),
601 ERROR, /* (incomplete sequence) */
602 0, -1);
603 utf8_read_test(TESTSTR("\xF8\x80\x80\x80"),
604 ERROR, /* (incomplete sequence) */
605 0, -1);
606 utf8_read_test(TESTSTR("\xFC\x80\x80\x80\x80"),
607 ERROR, /* (incomplete sequence) */
608 0, -1);
609 utf8_read_test(TESTSTR("\xDF"),
610 ERROR, /* (incomplete sequence) */
611 0, -1);
612 utf8_read_test(TESTSTR("\xEF\xBF"),
613 ERROR, /* (incomplete sequence) */
614 0, -1);
615 utf8_read_test(TESTSTR("\xF7\xBF\xBF"),
616 ERROR, /* (incomplete sequence) */
617 0, -1);
618 utf8_read_test(TESTSTR("\xFB\xBF\xBF\xBF"),
619 ERROR, /* (incomplete sequence) */
620 0, -1);
621 utf8_read_test(TESTSTR("\xFD\xBF\xBF\xBF\xBF"),
622 ERROR, /* (incomplete sequence) */
623 0, -1);
624 utf8_read_test(TESTSTR("\xC0\xE0\x80\xF0\x80\x80\xF8\x80\x80\x80\xFC\x80\x80\x80\x80\xDF\xEF\xBF\xF7\xBF\xBF\xFB\xBF\xBF\xBF\xFD\xBF\xBF\xBF\xBF"),
625 ERROR, /* (incomplete sequence) */
626 ERROR, /* (incomplete sequence) */
627 ERROR, /* (incomplete sequence) */
628 ERROR, /* (incomplete sequence) */
629 ERROR, /* (incomplete sequence) */
630 ERROR, /* (incomplete sequence) */
631 ERROR, /* (incomplete sequence) */
632 ERROR, /* (incomplete sequence) */
633 ERROR, /* (incomplete sequence) */
634 ERROR, /* (incomplete sequence) */
635 0, -1);
636 utf8_read_test(TESTSTR("\xFE"),
637 ERROR, /* (invalid UTF-8 byte) */
638 0, -1);
639 utf8_read_test(TESTSTR("\xFF"),
640 ERROR, /* (invalid UTF-8 byte) */
641 0, -1);
642 utf8_read_test(TESTSTR("\xFE\xFE\xFF\xFF"),
643 ERROR, /* (invalid UTF-8 byte) */
644 ERROR, /* (invalid UTF-8 byte) */
645 ERROR, /* (invalid UTF-8 byte) */
646 ERROR, /* (invalid UTF-8 byte) */
647 0, -1);
648 utf8_read_test(TESTSTR("\xC0\xAF"),
649 ERROR, /* SOLIDUS (overlong form of 2F) */
650 0, -1);
651 utf8_read_test(TESTSTR("\xE0\x80\xAF"),
652 ERROR, /* SOLIDUS (overlong form of 2F) */
653 0, -1);
654 utf8_read_test(TESTSTR("\xF0\x80\x80\xAF"),
655 ERROR, /* SOLIDUS (overlong form of 2F) */
656 0, -1);
657 utf8_read_test(TESTSTR("\xF8\x80\x80\x80\xAF"),
658 ERROR, /* SOLIDUS (overlong form of 2F) */
659 0, -1);
660 utf8_read_test(TESTSTR("\xFC\x80\x80\x80\x80\xAF"),
661 ERROR, /* SOLIDUS (overlong form of 2F) */
662 0, -1);
663 utf8_read_test(TESTSTR("\xC1\xBF"),
664 ERROR, /* <control> (overlong form of 7F) */
665 0, -1);
666 utf8_read_test(TESTSTR("\xE0\x9F\xBF"),
667 ERROR, /* <no name available> (overlong form of DF BF) */
668 0, -1);
669 utf8_read_test(TESTSTR("\xF0\x8F\xBF\xBF"),
670 ERROR, /* <no name available> (overlong form of EF BF BF) (invalid char) */
671 0, -1);
672 utf8_read_test(TESTSTR("\xF8\x87\xBF\xBF\xBF"),
673 ERROR, /* <no name available> (overlong form of F7 BF BF BF) */
674 0, -1);
675 utf8_read_test(TESTSTR("\xFC\x83\xBF\xBF\xBF\xBF"),
676 ERROR, /* <no name available> (overlong form of FB BF BF BF BF) */
677 0, -1);
678 utf8_read_test(TESTSTR("\xC0\x80"),
679 ERROR, /* <control> (overlong form of 00) */
680 0, -1);
681 utf8_read_test(TESTSTR("\xE0\x80\x80"),
682 ERROR, /* <control> (overlong form of 00) */
683 0, -1);
684 utf8_read_test(TESTSTR("\xF0\x80\x80\x80"),
685 ERROR, /* <control> (overlong form of 00) */
686 0, -1);
687 utf8_read_test(TESTSTR("\xF8\x80\x80\x80\x80"),
688 ERROR, /* <control> (overlong form of 00) */
689 0, -1);
690 utf8_read_test(TESTSTR("\xFC\x80\x80\x80\x80\x80"),
691 ERROR, /* <control> (overlong form of 00) */
692 0, -1);
693 utf8_read_test(TESTSTR("\xED\xA0\x80"),
694 ERROR, /* <Non Private Use High Surrogate, First> (surrogate) */
695 0, -1);
696 utf8_read_test(TESTSTR("\xED\xAD\xBF"),
697 ERROR, /* <Non Private Use High Surrogate, Last> (surrogate) */
698 0, -1);
699 utf8_read_test(TESTSTR("\xED\xAE\x80"),
700 ERROR, /* <Private Use High Surrogate, First> (surrogate) */
701 0, -1);
702 utf8_read_test(TESTSTR("\xED\xAF\xBF"),
703 ERROR, /* <Private Use High Surrogate, Last> (surrogate) */
704 0, -1);
705 utf8_read_test(TESTSTR("\xED\xB0\x80"),
706 ERROR, /* <Low Surrogate, First> (surrogate) */
707 0, -1);
708 utf8_read_test(TESTSTR("\xED\xBE\x80"),
709 ERROR, /* <no name available> (surrogate) */
710 0, -1);
711 utf8_read_test(TESTSTR("\xED\xBF\xBF"),
712 ERROR, /* <Low Surrogate, Last> (surrogate) */
713 0, -1);
714 utf8_read_test(TESTSTR("\xED\xA0\x80\xED\xB0\x80"),
715 ERROR, /* <Non Private Use High Surrogate, First> (surrogate) */
716 ERROR, /* <Low Surrogate, First> (surrogate) */
717 0, -1);
718 utf8_read_test(TESTSTR("\xED\xA0\x80\xED\xBF\xBF"),
719 ERROR, /* <Non Private Use High Surrogate, First> (surrogate) */
720 ERROR, /* <Low Surrogate, Last> (surrogate) */
721 0, -1);
722 utf8_read_test(TESTSTR("\xED\xAD\xBF\xED\xB0\x80"),
723 ERROR, /* <Non Private Use High Surrogate, Last> (surrogate) */
724 ERROR, /* <Low Surrogate, First> (surrogate) */
725 0, -1);
726 utf8_read_test(TESTSTR("\xED\xAD\xBF\xED\xBF\xBF"),
727 ERROR, /* <Non Private Use High Surrogate, Last> (surrogate) */
728 ERROR, /* <Low Surrogate, Last> (surrogate) */
729 0, -1);
730 utf8_read_test(TESTSTR("\xED\xAE\x80\xED\xB0\x80"),
731 ERROR, /* <Private Use High Surrogate, First> (surrogate) */
732 ERROR, /* <Low Surrogate, First> (surrogate) */
733 0, -1);
734 utf8_read_test(TESTSTR("\xED\xAE\x80\xED\xBF\xBF"),
735 ERROR, /* <Private Use High Surrogate, First> (surrogate) */
736 ERROR, /* <Low Surrogate, Last> (surrogate) */
737 0, -1);
738 utf8_read_test(TESTSTR("\xED\xAF\xBF\xED\xB0\x80"),
739 ERROR, /* <Private Use High Surrogate, Last> (surrogate) */
740 ERROR, /* <Low Surrogate, First> (surrogate) */
741 0, -1);
742 utf8_read_test(TESTSTR("\xED\xAF\xBF\xED\xBF\xBF"),
743 ERROR, /* <Private Use High Surrogate, Last> (surrogate) */
744 ERROR, /* <Low Surrogate, Last> (surrogate) */
745 0, -1);
746 utf8_read_test(TESTSTR("\xEF\xBF\xBE"),
747 ERROR, /* <no name available> (invalid char) */
748 0, -1);
749 utf8_read_test(TESTSTR("\xEF\xBF\xBF"),
750 ERROR, /* <no name available> (invalid char) */
751 0, -1);
752 printf("read tests completed\n");
753 printf("write tests beginning\n");
754 {
755 const static long str[] =
756 {0x03BAL, 0x1F79L, 0x03C3L, 0x03BCL, 0x03B5L, 0};
757 utf8_write_test(TESTSTR(str),
758 0xCE, 0xBA,
759 0xE1, 0xBD, 0xB9,
760 0xCF, 0x83,
761 0xCE, 0xBC,
762 0xCE, 0xB5,
763 0, -1);
764 }
765 {
766 const static long str[] = {0x0000L, 0};
767 utf8_write_test(TESTSTR(str),
768 0x00,
769 0, -1);
770 }
771 {
772 const static long str[] = {0x0080L, 0};
773 utf8_write_test(TESTSTR(str),
774 0xC2, 0x80,
775 0, -1);
776 }
777 {
778 const static long str[] = {0x0800L, 0};
779 utf8_write_test(TESTSTR(str),
780 0xE0, 0xA0, 0x80,
781 0, -1);
782 }
783 {
784 const static long str[] = {0x00010000L, 0};
785 utf8_write_test(TESTSTR(str),
786 0xF0, 0x90, 0x80, 0x80,
787 0, -1);
788 }
789 {
790 const static long str[] = {0x00200000L, 0};
791 utf8_write_test(TESTSTR(str),
792 0xF8, 0x88, 0x80, 0x80, 0x80,
793 0, -1);
794 }
795 {
796 const static long str[] = {0x04000000L, 0};
797 utf8_write_test(TESTSTR(str),
798 0xFC, 0x84, 0x80, 0x80, 0x80, 0x80,
799 0, -1);
800 }
801 {
802 const static long str[] = {0x007FL, 0};
803 utf8_write_test(TESTSTR(str),
804 0x7F,
805 0, -1);
806 }
807 {
808 const static long str[] = {0x07FFL, 0};
809 utf8_write_test(TESTSTR(str),
810 0xDF, 0xBF,
811 0, -1);
812 }
813 {
814 const static long str[] = {0xFFFDL, 0};
815 utf8_write_test(TESTSTR(str),
816 0xEF, 0xBF, 0xBD,
817 0, -1);
818 }
819 {
820 const static long str[] = {0xFFFFL, 0};
821 utf8_write_test(TESTSTR(str),
822 ERROR,
823 0, -1);
824 }
825 {
826 const static long str[] = {0x001FFFFFL, 0};
827 utf8_write_test(TESTSTR(str),
828 0xF7, 0xBF, 0xBF, 0xBF,
829 0, -1);
830 }
831 {
832 const static long str[] = {0x03FFFFFFL, 0};
833 utf8_write_test(TESTSTR(str),
834 0xFB, 0xBF, 0xBF, 0xBF, 0xBF,
835 0, -1);
836 }
837 {
838 const static long str[] = {0x7FFFFFFFL, 0};
839 utf8_write_test(TESTSTR(str),
840 0xFD, 0xBF, 0xBF, 0xBF, 0xBF, 0xBF,
841 0, -1);
842 }
843 {
844 const static long str[] = {0xD7FFL, 0};
845 utf8_write_test(TESTSTR(str),
846 0xED, 0x9F, 0xBF,
847 0, -1);
848 }
849 {
850 const static long str[] = {0xD800L, 0};
851 utf8_write_test(TESTSTR(str),
852 ERROR,
853 0, -1);
854 }
855 {
856 const static long str[] = {0xD800L, 0xDC00L, 0};
857 utf8_write_test(TESTSTR(str),
858 ERROR,
859 ERROR,
860 0, -1);
861 }
862 {
863 const static long str[] = {0xDFFFL, 0};
864 utf8_write_test(TESTSTR(str),
865 ERROR,
866 0, -1);
867 }
868 {
869 const static long str[] = {0xE000L, 0};
870 utf8_write_test(TESTSTR(str),
871 0xEE, 0x80, 0x80,
872 0, -1);
873 }
874 printf("write tests completed\n");
875
876 printf("total: %d errors\n", total_errs);
877 return (total_errs != 0);
878 }
879 #endif /* TESTMODE */
880
881 const charset_spec charset_CS_UTF8 = {
882 CS_UTF8, read_utf8, write_utf8, NULL
883 };
884
885 #else /* ENUM_CHARSETS */
886
887 ENUM_CHARSET(CS_UTF8)
888
889 #endif /* ENUM_CHARSETS */
Simon Tatham's `charset' library; import of svn://svn.tartarus.org/sgt/charset