1 /* This file is part of the YAZ toolkit.
2 * Copyright (C) 1995-2010 Index Data
3 * See the file LICENSE for details.
14 #include <yaz/yaz-util.h>
19 static int compare_buffers(char *msg, int no,
20 int expect_len, const char *expect_buf,
21 int got_len, const char *got_buf)
23 if (expect_len == got_len
24 && !memcmp(expect_buf, got_buf, expect_len))
27 if (0) /* use 1 see how the buffers differ (for debug purposes) */
30 printf("tsticonv test=%s i=%d failed\n", msg, no);
31 printf("off got exp\n");
32 for (i = 0; i<got_len || i<expect_len; i++)
38 sprintf(got_char, "%02X", got_buf[i]);
40 sprintf(got_char, "? ");
43 sprintf(expect_char, "%02X", expect_buf[i]);
45 sprintf(expect_char, "? ");
47 printf("%02d %s %s %c\n",
48 i, got_char, expect_char, got_buf[i] == expect_buf[i] ?
56 static int tst_convert_l(yaz_iconv_t cd, size_t in_len, const char *in_buf,
57 size_t expect_len, const char *expect_buf)
60 char *inbuf= (char*) in_buf;
61 size_t inbytesleft = in_len > 0 ? in_len : strlen(in_buf);
63 char *outbuf = outbuf0;
67 size_t outbytesleft = outbuf0 + sizeof(outbuf0) - outbuf;
68 if (outbytesleft > 12)
70 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
71 if (r == (size_t) (-1))
73 int e = yaz_iconv_error(cd);
74 if (e != YAZ_ICONV_E2BIG)
79 yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
84 return compare_buffers("tsticonv 22", 0,
85 expect_len, expect_buf,
86 outbuf - outbuf0, outbuf0);
89 static int tst_convert_x(yaz_iconv_t cd, const char *buf, const char *cmpbuf,
93 WRBUF b = wrbuf_alloc();
95 size_t inbytesleft = strlen(buf);
96 const char *inp = buf;
98 for (rounds = 0; inbytesleft && rounds < (int) sizeof(outbuf); rounds++)
100 size_t outbytesleft = sizeof(outbuf);
102 size_t r = yaz_iconv(cd, (char**) &inp, &inbytesleft,
103 &outp, &outbytesleft);
104 wrbuf_write(b, outbuf, outp - outbuf);
105 if (r == (size_t) (-1))
107 int e = yaz_iconv_error(cd);
108 if (e != YAZ_ICONV_E2BIG)
110 if (expect_error != -1)
111 if (e != expect_error)
118 size_t outbytesleft = sizeof(outbuf);
120 r = yaz_iconv(cd, 0, 0, &outp, &outbytesleft);
121 wrbuf_write(b, outbuf, outp - outbuf);
122 if (expect_error != -1)
128 if (wrbuf_len(b) == strlen(cmpbuf)
129 && !memcmp(cmpbuf, wrbuf_buf(b), wrbuf_len(b)))
133 WRBUF w = wrbuf_alloc();
137 wrbuf_puts_escaped(w, buf);
138 yaz_log(YLOG_LOG, "input %s", wrbuf_cstr(w));
141 wrbuf_write_escaped(w, wrbuf_buf(b), wrbuf_len(b));
142 yaz_log(YLOG_LOG, "got %s", wrbuf_cstr(w));
145 wrbuf_puts_escaped(w, cmpbuf);
146 yaz_log(YLOG_LOG, "exp %s", wrbuf_cstr(w));
155 static int tst_convert(yaz_iconv_t cd, const char *buf, const char *cmpbuf)
157 return tst_convert_x(cd, buf, cmpbuf, 0);
160 /* some test strings in ISO-8859-1 format */
161 static const char *iso_8859_1_a[] = {
170 static void tst_marc8_to_ucs4b(void)
172 yaz_iconv_t cd = yaz_iconv_open("UCS4", "MARC8");
177 YAZ_CHECK(tst_convert_l(
180 "\033$1" "\x21\x2B\x3B" /* FF1F */ "\033(B" "o",
182 "\x00\x00\xFF\x1F" "\x00\x00\x00o"));
183 YAZ_CHECK(tst_convert_l(
186 "\033$1" "\x6F\x77\x29" /* AE0E */
187 "\x6F\x52\x7C" /* c0F4 */ "\033(B",
189 "\x00\x00\xAE\x0E" "\x00\x00\xC0\xF4"));
190 YAZ_CHECK(tst_convert_l(
194 "\x21\x50\x6E" /* UCS 7CFB */
195 "\x21\x51\x31" /* UCS 7D71 */
196 "\x21\x3A\x67" /* UCS 5B89 */
197 "\x21\x33\x22" /* UCS 5168 */
198 "\x21\x33\x53" /* UCS 5206 */
199 "\x21\x44\x2B" /* UCS 6790 */
207 "\x00\x00\x67\x90"));
209 YAZ_CHECK(tst_convert_l(
212 "\xB0\xB2", /* AYN and oSLASH */
214 "\x00\x00\x02\xBB" "\x00\x00\x00\xF8"));
215 YAZ_CHECK(tst_convert_l(
218 "\xF6\x61", /* a underscore */
220 "\x00\x00\x00\x61" "\x00\x00\x03\x32"));
222 YAZ_CHECK(tst_convert_l(
225 "\x61\xC2", /* a, phonorecord mark */
227 "\x00\x00\x00\x61" "\x00\x00\x21\x17"));
230 YAZ_CHECK(tst_convert_l(
233 "el" "\xe8" "am\xe8" "an", /* elaman where a is a" */
242 "\x00\x00\x00" "n"));
244 YAZ_CHECK(tst_convert_l(
249 "\x00\x00\x00\x41" "\x00\x00\x03\x04" "\x00\x00\x03\x08"));
251 YAZ_CHECK(tst_convert_l(
256 "\x00\x00\x00\x74" "\x00\x00\x03\x61" "\x00\x00\x00\x73"));
258 YAZ_CHECK(tst_convert_l(
263 "\x00\x00\x00\x74" "\x00\x00\x03\x60" "\x00\x00\x00\x73"));
268 static void tst_ucs4b_to_utf8(void)
270 yaz_iconv_t cd = yaz_iconv_open("UTF8", "UCS4");
274 YAZ_CHECK(tst_convert_l(
277 "\x00\x00\xFF\x1F\x00\x00\x00o",
279 "\xEF\xBC\x9F\x6F"));
281 YAZ_CHECK(tst_convert_l(
284 "\x00\x00\xAE\x0E\x00\x00\xC0\xF4",
286 "\xEA\xB8\x8E\xEC\x83\xB4"));
290 static void dconvert(int mandatory, const char *tmpcode)
295 for (i = 0; iso_8859_1_a[i]; i++)
298 char *inbuf = (char*) iso_8859_1_a[i];
299 size_t inbytesleft = strlen(inbuf);
302 char *outbuf = outbuf0;
303 size_t outbytesleft = sizeof(outbuf0);
305 cd = yaz_iconv_open(tmpcode, "ISO-8859-1");
306 YAZ_CHECK(cd || !mandatory);
309 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
310 YAZ_CHECK(r != (size_t) (-1));
312 r = yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
313 YAZ_CHECK(r != (size_t) (-1));
315 if (r == (size_t) (-1))
318 cd = yaz_iconv_open("ISO-8859-1", tmpcode);
319 YAZ_CHECK(cd || !mandatory);
323 inbytesleft = sizeof(outbuf0) - outbytesleft;
326 outbytesleft = sizeof(outbuf1);
327 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
328 YAZ_CHECK(r != (size_t) (-1));
330 r = yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
331 if (r == (size_t)(-1))
333 fprintf(stderr, "failed\n");
335 YAZ_CHECK(r != (size_t) (-1));
337 if (r != (size_t)(-1))
339 ret = compare_buffers("dconvert", i,
340 strlen(iso_8859_1_a[i]), iso_8859_1_a[i],
341 sizeof(outbuf1) - outbytesleft, outbuf1);
348 int utf8_check(unsigned c)
357 size_t inbytesleft = 4;
358 char *outbuf = utf8buf;
359 size_t outbytesleft = sizeof(utf8buf);
361 yaz_iconv_t cd = yaz_iconv_open("UTF-8", "UCS4LE");
364 for (i = 0; i<4; i++)
367 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
370 if (r == (size_t)(-1))
373 cd = yaz_iconv_open("UCS4LE", "UTF-8");
376 inbytesleft = sizeof(utf8buf) - outbytesleft;
382 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
383 if (r == (size_t)(-1))
388 if (memcmp(src, dst, 4))
394 static void tst_marc8_to_utf8(void)
396 yaz_iconv_t cd = yaz_iconv_open("UTF-8", "MARC8");
402 YAZ_CHECK(tst_convert(cd, "Cours de math",
404 /* COMBINING ACUTE ACCENT */
405 YAZ_CHECK(tst_convert(cd, "Cours de mathâe",
406 "Cours de mathe\xcc\x81"));
408 YAZ_CHECK(tst_convert(cd, "\xea" "a", "a\xcc\x8a"));
409 YAZ_CHECK(tst_convert(cd, "a" "\xea" "\x1e", "a" "\x1e\xcc\x8a"));
410 YAZ_CHECK(tst_convert(cd, "a" "\xea" "p", "a" "p\xcc\x8a"));
412 YAZ_CHECK(tst_convert_x(cd, "a\xea", "a", YAZ_ICONV_EINVAL));
413 YAZ_CHECK(tst_convert(cd, "p", "\xcc\x8a")); /* note: missing p */
414 yaz_iconv(cd, 0, 0, 0, 0); /* incomplete. so we have to reset */
417 YAZ_CHECK(tst_convert(cd, ESC "(N" ESC ")Qp" ESC "(B", "\xd0\x9f"));
419 YAZ_CHECK(tst_convert_x(cd, ESC , "", YAZ_ICONV_EINVAL));
420 YAZ_CHECK(tst_convert_x(cd, ESC "(", "", YAZ_ICONV_EINVAL));
421 YAZ_CHECK(tst_convert_x(cd, ESC "(B", "", 0));
423 YAZ_CHECK(tst_convert(cd, ESC "(B" "\x31", "1")); /* ASCII in G0 */
424 YAZ_CHECK(tst_convert(cd, ESC ")B" "\xB1", "1")); /* ASCII in G1 */
429 static void tst_marc8s_to_utf8(void)
431 yaz_iconv_t cd = yaz_iconv_open("UTF-8", "MARC8s");
437 YAZ_CHECK(tst_convert(cd, "Cours de math",
439 /* E9: LATIN SMALL LETTER E WITH ACUTE */
440 YAZ_CHECK(tst_convert(cd, "Cours de mathâe",
441 "Cours de math\xc3\xa9"));
447 static void tst_marc8_to_latin1(void)
449 yaz_iconv_t cd = yaz_iconv_open("ISO-8859-1", "MARC8");
455 YAZ_CHECK(tst_convert(cd, "ax", "ax"));
457 /* latin capital letter o with stroke */
458 YAZ_CHECK(tst_convert(cd, "\xa2", "\xd8"));
460 /* with latin small letter ae */
461 YAZ_CHECK(tst_convert(cd, "eneb\xb5r", "eneb\346r"));
463 YAZ_CHECK(tst_convert(cd, "\xea" "a\xa2", "\xe5" "\xd8"));
465 YAZ_CHECK(tst_convert(cd, "\xea" "a\xa2" "b", "\xe5" "\xd8" "b"));
467 YAZ_CHECK(tst_convert(cd, "\xea" "a" "\xea" "a", "\xe5" "\xe5"));
469 YAZ_CHECK(tst_convert(cd, "Cours de math",
471 YAZ_CHECK(tst_convert(cd, "Cours de mathâe",
473 YAZ_CHECK(tst_convert(cd, "12345678âe",
475 YAZ_CHECK(tst_convert(cd, "123456789âe",
477 YAZ_CHECK(tst_convert(cd, "1234567890âe",
479 YAZ_CHECK(tst_convert(cd, "12345678901âe",
481 YAZ_CHECK(tst_convert(cd, "Cours de mathâem",
483 YAZ_CHECK(tst_convert(cd, "Cours de mathâematiques",
484 "Cours de mathématiques"));
489 static void tst_utf8_to_marc8(const char *marc8_type)
491 yaz_iconv_t cd = yaz_iconv_open(marc8_type, "UTF-8");
497 YAZ_CHECK(tst_convert(cd, "Cours ", "Cours "));
499 /** Pure ASCII. 11 characters (sizeof(outbuf)-1) */
500 YAZ_CHECK(tst_convert(cd, "Cours de mat", "Cours de mat"));
502 /** Pure ASCII. 12 characters (sizeof(outbuf)) */
503 YAZ_CHECK(tst_convert(cd, "Cours de math", "Cours de math"));
505 /** Pure ASCII. 13 characters (sizeof(outbuf)+1) */
506 YAZ_CHECK(tst_convert(cd, "Cours de math.", "Cours de math."));
508 /** UPPERCASE SCANDINAVIAN O */
509 YAZ_CHECK(tst_convert(cd, "S\xc3\x98", "S\xa2"));
512 YAZ_CHECK(tst_convert(cd, "A" "\xCC\x8A", "\xEA" "A"));
514 /** A MACRON + UMLAUT, DIAERESIS */
515 YAZ_CHECK(tst_convert(cd, "A" "\xCC\x84" "\xCC\x88",
518 /* Ligature spanning two characters */
519 YAZ_CHECK(tst_convert(cd,
520 "\x74" "\xCD\xA1" "\x73", /* UTF-8 */
521 "\xEB\x74\xEC\x73")); /* MARC-8 */
523 /* Double title spanning two characters */
524 YAZ_CHECK(tst_convert(cd,
525 "\x74" "\xCD\xA0" "\x73", /* UTF-8 */
526 "\xFA\x74\xFB\x73")); /* MARC-8 */
528 /** Ideographic question mark (Unicode FF1F) */
529 YAZ_CHECK(tst_convert(cd,
530 "\xEF\xBC\x9F" "o", /* UTF-8 */
531 "\033$1" "\x21\x2B\x3B" "\033(B" "o" ));
534 /** Ideographic space per ANSI Z39.64 */
535 YAZ_CHECK(tst_convert(cd,
536 "\xe3\x80\x80" "o", /* UTF-8 */
537 "\033$1" "\x21\x23\x21" "\033(B" "o" ));
539 /** Superscript 0 . bug #642 */
540 YAZ_CHECK(tst_convert(cd,
541 "(\xe2\x81\xb0)", /* UTF-8 */
546 YAZ_CHECK(tst_convert(cd,
547 /* offset 0x530 in UTF-8 rec marccol4.u8.marc */
548 "\xE3\x83\xB3" "\xE3\x82\xBF"
549 "\xCC\x84" "\xCC\x84" "\xE3\x83\xBC" /* UTF-8 */,
550 "\x1B\x24\x31" "\x69\x25\x73"
551 "\x1B\x28\x42" "\xE5\xE5" "\x1B\x24\x31"
553 "\x69\x21\x3C" "\x1B\x28\x42"));
557 YAZ_CHECK(tst_convert(cd,
558 "\xCE\x94\xCE\xB5\xCF\x84"
559 "\xCE\xBF\xCF\x81\xCE\xB1"
560 "\xCE\xBA\xCE\xB7\xCF\x82\x2C",
562 "\x1B\x28\x53\x45\x66\x78\x72\x75"
568 char *inbuf0 = "\xe2\x81\xb0";
569 char *inbuf = inbuf0;
570 size_t inbytesleft = strlen(inbuf);
572 char *outbuf = outbuf0;
573 size_t outbytesleft = sizeof(outbuf0)-1;
578 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
579 YAZ_CHECK(r != (size_t) (-1));
582 *outbuf = '\0'; /* so we know when to stop printing */
583 for (i = 0; outbuf0[i]; i++)
585 int ch = outbuf0[i] & 0xff;
586 yaz_log(YLOG_LOG, "ch%d %02X %c", i, ch, ch >= ' ' ? ch : '?');
590 r = yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
591 YAZ_CHECK(r != (size_t) (-1));
592 *outbuf = '\0'; /* for strcmp test below and printing */
594 for (i = 0; outbuf0[i]; i++)
596 int ch = outbuf0[i] & 0xff;
597 yaz_log(YLOG_LOG, "ch%d %02X %c", i, ch, ch >= ' ' ? ch : '?');
600 YAZ_CHECK(strcmp("\033p0\x1bs", outbuf0) == 0);
602 yaz_iconv(cd, 0, 0, 0, 0);
606 static void tst_advance_to_utf8(void)
608 yaz_iconv_t cd = yaz_iconv_open("utf-8", "advancegreek");
614 YAZ_CHECK(tst_convert(cd, "Cours ", "Cours "));
618 static void tst_utf8_to_advance(void)
620 yaz_iconv_t cd = yaz_iconv_open("advancegreek", "utf-8");
626 YAZ_CHECK(tst_convert(cd, "Cours ", "Cours "));
630 static void tst_latin1_to_marc8(void)
632 yaz_iconv_t cd = yaz_iconv_open("MARC8", "ISO-8859-1");
638 YAZ_CHECK(tst_convert(cd, "Cours ", "Cours "));
640 /** Pure ASCII. 11 characters (sizeof(outbuf)-1) */
641 YAZ_CHECK(tst_convert(cd, "Cours de mat", "Cours de mat"));
643 /** Pure ASCII. 12 characters (sizeof(outbuf)) */
644 YAZ_CHECK(tst_convert(cd, "Cours de math", "Cours de math"));
646 /** Pure ASCII. 13 characters (sizeof(outbuf)) */
647 YAZ_CHECK(tst_convert(cd, "Cours de math.", "Cours de math."));
649 /** D8: UPPERCASE SCANDINAVIAN O */
650 YAZ_CHECK(tst_convert(cd, "S\xd8", "S\xa2"));
652 /** E9: LATIN SMALL LETTER E WITH ACUTE */
653 YAZ_CHECK(tst_convert(cd, "Cours de math\xe9", "Cours de mathâe"));
654 YAZ_CHECK(tst_convert(cd, "Cours de math", "Cours de math"
656 YAZ_CHECK(tst_convert(cd, "Cours de mathé", "Cours de mathâe" ));
657 YAZ_CHECK(tst_convert(cd, "12345678é","12345678âe"));
658 YAZ_CHECK(tst_convert(cd, "123456789é", "123456789âe"));
659 YAZ_CHECK(tst_convert(cd, "1234567890é","1234567890âe"));
660 YAZ_CHECK(tst_convert(cd, "12345678901é", "12345678901âe"));
661 YAZ_CHECK(tst_convert(cd, "Cours de mathém", "Cours de mathâem"));
662 YAZ_CHECK(tst_convert(cd, "Cours de mathématiques",
663 "Cours de mathâematiques"));
667 static void tst_utf8_codes(void)
669 YAZ_CHECK(utf8_check(3));
670 YAZ_CHECK(utf8_check(127));
671 YAZ_CHECK(utf8_check(128));
672 YAZ_CHECK(utf8_check(255));
673 YAZ_CHECK(utf8_check(256));
674 YAZ_CHECK(utf8_check(900));
675 YAZ_CHECK(utf8_check(1000));
676 YAZ_CHECK(utf8_check(10000));
677 YAZ_CHECK(utf8_check(100000));
678 YAZ_CHECK(utf8_check(1000000));
679 YAZ_CHECK(utf8_check(10000000));
680 YAZ_CHECK(utf8_check(100000000));
683 static void tst_danmarc_to_latin1(void)
685 yaz_iconv_t cd = yaz_iconv_open("iso-8859-1", "danmarc");
691 YAZ_CHECK(tst_convert(cd, "ax", "ax"));
693 YAZ_CHECK(tst_convert(cd, "a@@b", "a@b"));
694 YAZ_CHECK(tst_convert(cd, "a@@@@b", "a@@b"));
695 YAZ_CHECK(tst_convert(cd, "@000ab", "\nb"));
697 YAZ_CHECK(tst_convert(cd, "@\xe5", "aa"));
698 YAZ_CHECK(tst_convert(cd, "@\xc5.", "Aa."));
704 int main (int argc, char **argv)
706 YAZ_CHECK_INIT(argc, argv);
712 tst_marc8s_to_utf8();
714 tst_marc8_to_latin1();
716 tst_advance_to_utf8();
717 tst_utf8_to_advance();
719 tst_utf8_to_marc8("marc8");
720 tst_utf8_to_marc8("marc8lossy");
721 tst_utf8_to_marc8("marc8lossless");
723 tst_danmarc_to_latin1();
725 tst_latin1_to_marc8();
727 tst_marc8_to_ucs4b();
730 dconvert(1, "UTF-8");
731 dconvert(1, "ISO-8859-1");
733 dconvert(1, "UCS4LE");
734 dconvert(0, "CP865");
741 * c-file-style: "Stroustrup"
742 * indent-tabs-mode: nil
744 * vim: shiftwidth=4 tabstop=8 expandtab