2 * Copyright (C) 1995-2007, Index Data ApS
3 * See the file LICENSE for details.
5 * $Id: tsticonv.c,v 1.35 2008-03-12 08:53:28 adam Exp $
17 #include <yaz/yaz-util.h>
22 static int compare_buffers(char *msg, int no,
23 int expect_len, const char *expect_buf,
24 int got_len, const char *got_buf)
26 if (expect_len == got_len
27 && !memcmp(expect_buf, got_buf, expect_len))
30 if (0) /* use 1 see how the buffers differ (for debug purposes) */
33 printf("tsticonv test=%s i=%d failed\n", msg, no);
34 printf("off got exp\n");
35 for (i = 0; i<got_len || i<expect_len; i++)
41 sprintf(got_char, "%02X", got_buf[i]);
43 sprintf(got_char, "? ");
46 sprintf(expect_char, "%02X", expect_buf[i]);
48 sprintf(expect_char, "? ");
50 printf("%02d %s %s %c\n",
51 i, got_char, expect_char, got_buf[i] == expect_buf[i] ?
59 static int tst_convert_l(yaz_iconv_t cd, size_t in_len, const char *in_buf,
60 size_t expect_len, const char *expect_buf)
63 char *inbuf= (char*) in_buf;
64 size_t inbytesleft = in_len > 0 ? in_len : strlen(in_buf);
66 char *outbuf = outbuf0;
70 size_t outbytesleft = outbuf0 + sizeof(outbuf0) - outbuf;
71 if (outbytesleft > 12)
73 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
74 if (r == (size_t) (-1))
76 int e = yaz_iconv_error(cd);
77 if (e != YAZ_ICONV_E2BIG)
82 yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
87 return compare_buffers("tsticonv 22", 0,
88 expect_len, expect_buf,
89 outbuf - outbuf0, outbuf0);
92 static int tst_convert_x(yaz_iconv_t cd, const char *buf, const char *cmpbuf,
96 WRBUF b = wrbuf_alloc();
98 size_t inbytesleft = strlen(buf);
99 const char *inp = buf;
101 for (rounds = 0; inbytesleft && rounds < sizeof(outbuf); rounds++)
103 size_t outbytesleft = sizeof(outbuf);
105 size_t r = yaz_iconv(cd, (char**) &inp, &inbytesleft,
106 &outp, &outbytesleft);
107 wrbuf_write(b, outbuf, outp - outbuf);
108 if (r == (size_t) (-1))
110 int e = yaz_iconv_error(cd);
111 if (e != YAZ_ICONV_E2BIG)
113 if (expect_error != -1)
114 if (e != expect_error)
121 size_t outbytesleft = sizeof(outbuf);
123 r = yaz_iconv(cd, 0, 0, &outp, &outbytesleft);
124 wrbuf_write(b, outbuf, outp - outbuf);
125 if (expect_error != -1)
131 if (wrbuf_len(b) == strlen(cmpbuf)
132 && !memcmp(cmpbuf, wrbuf_buf(b), wrbuf_len(b)))
136 WRBUF w = wrbuf_alloc();
140 wrbuf_puts_escaped(w, buf);
141 yaz_log(YLOG_LOG, "input %s", wrbuf_cstr(w));
144 wrbuf_write_escaped(w, wrbuf_buf(b), wrbuf_len(b));
145 yaz_log(YLOG_LOG, "got %s", wrbuf_cstr(w));
148 wrbuf_puts_escaped(w, cmpbuf);
149 yaz_log(YLOG_LOG, "exp %s", wrbuf_cstr(w));
158 static int tst_convert(yaz_iconv_t cd, const char *buf, const char *cmpbuf)
160 return tst_convert_x(cd, buf, cmpbuf, 0);
163 /* some test strings in ISO-8859-1 format */
164 static const char *iso_8859_1_a[] = {
173 static void tst_marc8_to_ucs4b(void)
175 yaz_iconv_t cd = yaz_iconv_open("UCS4", "MARC8");
180 YAZ_CHECK(tst_convert_l(
183 "\033$1" "\x21\x2B\x3B" /* FF1F */ "\033(B" "o",
185 "\x00\x00\xFF\x1F" "\x00\x00\x00o"));
186 YAZ_CHECK(tst_convert_l(
189 "\033$1" "\x6F\x77\x29" /* AE0E */
190 "\x6F\x52\x7C" /* c0F4 */ "\033(B",
192 "\x00\x00\xAE\x0E" "\x00\x00\xC0\xF4"));
193 YAZ_CHECK(tst_convert_l(
197 "\x21\x50\x6E" /* UCS 7CFB */
198 "\x21\x51\x31" /* UCS 7D71 */
199 "\x21\x3A\x67" /* UCS 5B89 */
200 "\x21\x33\x22" /* UCS 5168 */
201 "\x21\x33\x53" /* UCS 5206 */
202 "\x21\x44\x2B" /* UCS 6790 */
210 "\x00\x00\x67\x90"));
212 YAZ_CHECK(tst_convert_l(
215 "\xB0\xB2", /* AYN and oSLASH */
217 "\x00\x00\x02\xBB" "\x00\x00\x00\xF8"));
218 YAZ_CHECK(tst_convert_l(
221 "\xF6\x61", /* a underscore */
223 "\x00\x00\x00\x61" "\x00\x00\x03\x32"));
225 YAZ_CHECK(tst_convert_l(
228 "\x61\xC2", /* a, phonorecord mark */
230 "\x00\x00\x00\x61" "\x00\x00\x21\x17"));
233 YAZ_CHECK(tst_convert_l(
236 "el" "\xe8" "am\xe8" "an", /* elaman where a is a" */
245 "\x00\x00\x00" "n"));
247 YAZ_CHECK(tst_convert_l(
252 "\x00\x00\x00\x41" "\x00\x00\x03\x04" "\x00\x00\x03\x08"));
254 YAZ_CHECK(tst_convert_l(
259 "\x00\x00\x00\x74" "\x00\x00\x03\x61" "\x00\x00\x00\x73"));
261 YAZ_CHECK(tst_convert_l(
266 "\x00\x00\x00\x74" "\x00\x00\x03\x60" "\x00\x00\x00\x73"));
271 static void tst_ucs4b_to_utf8(void)
273 yaz_iconv_t cd = yaz_iconv_open("UTF8", "UCS4");
277 YAZ_CHECK(tst_convert_l(
280 "\x00\x00\xFF\x1F\x00\x00\x00o",
282 "\xEF\xBC\x9F\x6F"));
284 YAZ_CHECK(tst_convert_l(
287 "\x00\x00\xAE\x0E\x00\x00\xC0\xF4",
289 "\xEA\xB8\x8E\xEC\x83\xB4"));
293 static void dconvert(int mandatory, const char *tmpcode)
298 for (i = 0; iso_8859_1_a[i]; i++)
301 char *inbuf = (char*) iso_8859_1_a[i];
302 size_t inbytesleft = strlen(inbuf);
305 char *outbuf = outbuf0;
306 size_t outbytesleft = sizeof(outbuf0);
308 cd = yaz_iconv_open(tmpcode, "ISO-8859-1");
309 YAZ_CHECK(cd || !mandatory);
312 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
313 YAZ_CHECK(r != (size_t) (-1));
315 r = yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
316 YAZ_CHECK(r != (size_t) (-1));
318 if (r == (size_t) (-1))
321 cd = yaz_iconv_open("ISO-8859-1", tmpcode);
322 YAZ_CHECK(cd || !mandatory);
326 inbytesleft = sizeof(outbuf0) - outbytesleft;
329 outbytesleft = sizeof(outbuf1);
330 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
331 YAZ_CHECK(r != (size_t) (-1));
333 r = yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
334 if (r == (size_t)(-1))
336 fprintf(stderr, "failed\n");
338 YAZ_CHECK(r != (size_t) (-1));
340 if (r != (size_t)(-1))
342 ret = compare_buffers("dconvert", i,
343 strlen(iso_8859_1_a[i]), iso_8859_1_a[i],
344 sizeof(outbuf1) - outbytesleft, outbuf1);
351 int utf8_check(unsigned c)
360 size_t inbytesleft = 4;
361 char *outbuf = utf8buf;
362 size_t outbytesleft = sizeof(utf8buf);
364 yaz_iconv_t cd = yaz_iconv_open("UTF-8", "UCS4LE");
367 for (i = 0; i<4; i++)
370 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
373 if (r == (size_t)(-1))
376 cd = yaz_iconv_open("UCS4LE", "UTF-8");
379 inbytesleft = sizeof(utf8buf) - outbytesleft;
385 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
386 if (r == (size_t)(-1))
391 if (memcmp(src, dst, 4))
397 static void tst_marc8_to_utf8(void)
399 yaz_iconv_t cd = yaz_iconv_open("UTF-8", "MARC8");
405 YAZ_CHECK(tst_convert(cd, "Cours de math",
407 /* COMBINING ACUTE ACCENT */
408 YAZ_CHECK(tst_convert(cd, "Cours de mathâe",
409 "Cours de mathe\xcc\x81"));
411 YAZ_CHECK(tst_convert(cd, "\xea" "a", "a\xcc\x8a"));
412 YAZ_CHECK(tst_convert(cd, "a" "\xea" "\x1e", "a" "\x1e\xcc\x8a"));
413 YAZ_CHECK(tst_convert(cd, "a" "\xea" "p", "a" "p\xcc\x8a"));
415 YAZ_CHECK(tst_convert_x(cd, "a\xea", "a", YAZ_ICONV_EINVAL));
416 YAZ_CHECK(tst_convert(cd, "p", "\xcc\x8a")); /* note: missing p */
417 yaz_iconv(cd, 0, 0, 0, 0); /* incomplete. so we have to reset */
420 YAZ_CHECK(tst_convert(cd, ESC "(N" ESC ")Qp" ESC "(B", "\xd0\x9f"));
422 YAZ_CHECK(tst_convert_x(cd, ESC , "", YAZ_ICONV_EINVAL));
423 YAZ_CHECK(tst_convert_x(cd, ESC "(", "", YAZ_ICONV_EINVAL));
424 YAZ_CHECK(tst_convert_x(cd, ESC "(B", "", 0));
426 YAZ_CHECK(tst_convert(cd, ESC "(B" "\x31", "1")); /* ASCII in G0 */
427 YAZ_CHECK(tst_convert(cd, ESC ")B" "\xB1", "1")); /* ASCII in G1 */
432 static void tst_marc8s_to_utf8(void)
434 yaz_iconv_t cd = yaz_iconv_open("UTF-8", "MARC8s");
440 YAZ_CHECK(tst_convert(cd, "Cours de math",
442 /* E9: LATIN SMALL LETTER E WITH ACUTE */
443 YAZ_CHECK(tst_convert(cd, "Cours de mathâe",
444 "Cours de math\xc3\xa9"));
450 static void tst_marc8_to_latin1(void)
452 yaz_iconv_t cd = yaz_iconv_open("ISO-8859-1", "MARC8");
458 YAZ_CHECK(tst_convert(cd, "ax", "ax"));
460 /* latin capital letter o with stroke */
461 YAZ_CHECK(tst_convert(cd, "\xa2", "\xd8"));
463 /* with latin small letter ae */
464 YAZ_CHECK(tst_convert(cd, "eneb\xb5r", "eneb\346r"));
466 YAZ_CHECK(tst_convert(cd, "\xea" "a\xa2", "\xe5" "\xd8"));
468 YAZ_CHECK(tst_convert(cd, "\xea" "a\xa2" "b", "\xe5" "\xd8" "b"));
470 YAZ_CHECK(tst_convert(cd, "\xea" "a" "\xea" "a", "\xe5" "\xe5"));
472 YAZ_CHECK(tst_convert(cd, "Cours de math",
474 YAZ_CHECK(tst_convert(cd, "Cours de mathâe",
476 YAZ_CHECK(tst_convert(cd, "12345678âe",
478 YAZ_CHECK(tst_convert(cd, "123456789âe",
480 YAZ_CHECK(tst_convert(cd, "1234567890âe",
482 YAZ_CHECK(tst_convert(cd, "12345678901âe",
484 YAZ_CHECK(tst_convert(cd, "Cours de mathâem",
486 YAZ_CHECK(tst_convert(cd, "Cours de mathâematiques",
487 "Cours de mathématiques"));
492 static void tst_utf8_to_marc8(void)
494 yaz_iconv_t cd = yaz_iconv_open("MARC8", "UTF-8");
500 YAZ_CHECK(tst_convert(cd, "Cours ", "Cours "));
502 /** Pure ASCII. 11 characters (sizeof(outbuf)-1) */
503 YAZ_CHECK(tst_convert(cd, "Cours de mat", "Cours de mat"));
505 /** Pure ASCII. 12 characters (sizeof(outbuf)) */
506 YAZ_CHECK(tst_convert(cd, "Cours de math", "Cours de math"));
508 /** Pure ASCII. 13 characters (sizeof(outbuf)+1) */
509 YAZ_CHECK(tst_convert(cd, "Cours de math.", "Cours de math."));
511 /** UPPERCASE SCANDINAVIAN O */
512 YAZ_CHECK(tst_convert(cd, "S\xc3\x98", "S\xa2"));
515 YAZ_CHECK(tst_convert(cd, "A" "\xCC\x8A", "\xEA" "A"));
517 /** A MACRON + UMLAUT, DIAERESIS */
518 YAZ_CHECK(tst_convert(cd, "A" "\xCC\x84" "\xCC\x88",
521 /* Ligature spanning two characters */
522 YAZ_CHECK(tst_convert(cd,
523 "\x74" "\xCD\xA1" "\x73", /* UTF-8 */
524 "\xEB\x74\xEC\x73")); /* MARC-8 */
526 /* Double title spanning two characters */
527 YAZ_CHECK(tst_convert(cd,
528 "\x74" "\xCD\xA0" "\x73", /* UTF-8 */
529 "\xFA\x74\xFB\x73")); /* MARC-8 */
531 /** Ideographic question mark (Unicode FF1F) */
532 YAZ_CHECK(tst_convert(cd,
533 "\xEF\xBC\x9F" "o", /* UTF-8 */
534 "\033$1" "\x21\x2B\x3B" "\033(B" "o" ));
537 /** Ideographic space per ANSI Z39.64 */
538 YAZ_CHECK(tst_convert(cd,
539 "\xe3\x80\x80" "o", /* UTF-8 */
540 "\033$1" "\x21\x23\x21" "\033(B" "o" ));
542 /** Superscript 0 . bug #642 */
543 YAZ_CHECK(tst_convert(cd,
544 "(\xe2\x81\xb0)", /* UTF-8 */
549 YAZ_CHECK(tst_convert(cd,
550 /* offset 0x530 in UTF-8 rec marccol4.u8.marc */
551 "\xE3\x83\xB3" "\xE3\x82\xBF"
552 "\xCC\x84" "\xCC\x84" "\xE3\x83\xBC" /* UTF-8 */,
553 "\x1B\x24\x31" "\x69\x25\x73"
554 "\x1B\x28\x42" "\xE5\xE5" "\x1B\x24\x31"
556 "\x69\x21\x3C" "\x1B\x28\x42"));
560 YAZ_CHECK(tst_convert(cd,
561 "\xCE\x94\xCE\xB5\xCF\x84"
562 "\xCE\xBF\xCF\x81\xCE\xB1"
563 "\xCE\xBA\xCE\xB7\xCF\x82\x2C",
565 "\x1B\x28\x53\x45\x66\x78\x72\x75"
571 char *inbuf0 = "\xe2\x81\xb0";
572 char *inbuf = inbuf0;
573 size_t inbytesleft = strlen(inbuf);
575 char *outbuf = outbuf0;
576 size_t outbytesleft = sizeof(outbuf0)-1;
581 r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
582 YAZ_CHECK(r != (size_t) (-1));
585 *outbuf = '\0'; /* so we know when to stop printing */
586 for (i = 0; outbuf0[i]; i++)
588 int ch = outbuf0[i] & 0xff;
589 yaz_log(YLOG_LOG, "ch%d %02X %c", i, ch, ch >= ' ' ? ch : '?');
593 r = yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
594 YAZ_CHECK(r != (size_t) (-1));
595 *outbuf = '\0'; /* for strcmp test below and printing */
597 for (i = 0; outbuf0[i]; i++)
599 int ch = outbuf0[i] & 0xff;
600 yaz_log(YLOG_LOG, "ch%d %02X %c", i, ch, ch >= ' ' ? ch : '?');
603 YAZ_CHECK(strcmp("\033p0\x1bs", outbuf0) == 0);
608 static void tst_advance_to_utf8(void)
610 yaz_iconv_t cd = yaz_iconv_open("utf-8", "advancegreek");
616 YAZ_CHECK(tst_convert(cd, "Cours ", "Cours "));
620 static void tst_utf8_to_advance(void)
622 yaz_iconv_t cd = yaz_iconv_open("advancegreek", "utf-8");
628 YAZ_CHECK(tst_convert(cd, "Cours ", "Cours "));
632 static void tst_latin1_to_marc8(void)
634 yaz_iconv_t cd = yaz_iconv_open("MARC8", "ISO-8859-1");
640 YAZ_CHECK(tst_convert(cd, "Cours ", "Cours "));
642 /** Pure ASCII. 11 characters (sizeof(outbuf)-1) */
643 YAZ_CHECK(tst_convert(cd, "Cours de mat", "Cours de mat"));
645 /** Pure ASCII. 12 characters (sizeof(outbuf)) */
646 YAZ_CHECK(tst_convert(cd, "Cours de math", "Cours de math"));
648 /** Pure ASCII. 13 characters (sizeof(outbuf)) */
649 YAZ_CHECK(tst_convert(cd, "Cours de math.", "Cours de math."));
651 /** D8: UPPERCASE SCANDINAVIAN O */
652 YAZ_CHECK(tst_convert(cd, "S\xd8", "S\xa2"));
654 /** E9: LATIN SMALL LETTER E WITH ACUTE */
655 YAZ_CHECK(tst_convert(cd, "Cours de math\xe9", "Cours de mathâe"));
656 YAZ_CHECK(tst_convert(cd, "Cours de math", "Cours de math"
658 YAZ_CHECK(tst_convert(cd, "Cours de mathé", "Cours de mathâe" ));
659 YAZ_CHECK(tst_convert(cd, "12345678é","12345678âe"));
660 YAZ_CHECK(tst_convert(cd, "123456789é", "123456789âe"));
661 YAZ_CHECK(tst_convert(cd, "1234567890é","1234567890âe"));
662 YAZ_CHECK(tst_convert(cd, "12345678901é", "12345678901âe"));
663 YAZ_CHECK(tst_convert(cd, "Cours de mathém", "Cours de mathâem"));
664 YAZ_CHECK(tst_convert(cd, "Cours de mathématiques",
665 "Cours de mathâematiques"));
669 static void tst_utf8_codes(void)
671 YAZ_CHECK(utf8_check(3));
672 YAZ_CHECK(utf8_check(127));
673 YAZ_CHECK(utf8_check(128));
674 YAZ_CHECK(utf8_check(255));
675 YAZ_CHECK(utf8_check(256));
676 YAZ_CHECK(utf8_check(900));
677 YAZ_CHECK(utf8_check(1000));
678 YAZ_CHECK(utf8_check(10000));
679 YAZ_CHECK(utf8_check(100000));
680 YAZ_CHECK(utf8_check(1000000));
681 YAZ_CHECK(utf8_check(10000000));
682 YAZ_CHECK(utf8_check(100000000));
685 int main (int argc, char **argv)
687 YAZ_CHECK_INIT(argc, argv);
693 tst_marc8s_to_utf8();
695 tst_marc8_to_latin1();
697 tst_advance_to_utf8();
698 tst_utf8_to_advance();
702 tst_latin1_to_marc8();
704 tst_marc8_to_ucs4b();
707 dconvert(1, "UTF-8");
708 dconvert(1, "ISO-8859-1");
710 dconvert(1, "UCS4LE");
711 dconvert(0, "CP865");
718 * indent-tabs-mode: nil
720 * vim: shiftwidth=4 tabstop=8 expandtab