* Copyright (C) 1995-2007, Index Data ApS
* See the file LICENSE for details.
*
- * $Id: tsticonv.c,v 1.24 2007-01-03 08:42:16 adam Exp $
+ * $Id: tsticonv.c,v 1.35 2008-03-12 08:53:28 adam Exp $
*/
#if HAVE_CONFIG_H
#include <yaz/yaz-util.h>
#include <yaz/test.h>
+#define ESC "\x1b"
+
static int compare_buffers(char *msg, int no,
int expect_len, const char *expect_buf,
int got_len, const char *got_buf)
return 0;
}
else
+ {
+ yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
break;
+ }
}
+
return compare_buffers("tsticonv 22", 0,
expect_len, expect_buf,
outbuf - outbuf0, outbuf0);
}
-static int tst_convert(yaz_iconv_t cd, const char *buf, const char *cmpbuf)
+static int tst_convert_x(yaz_iconv_t cd, const char *buf, const char *cmpbuf,
+ int expect_error)
{
- int ret = 0;
+ int ret = 1;
WRBUF b = wrbuf_alloc();
- char outbuf[12];
+ char outbuf[16];
size_t inbytesleft = strlen(buf);
const char *inp = buf;
- while (inbytesleft)
+ int rounds = 0;
+ for (rounds = 0; inbytesleft && rounds < sizeof(outbuf); rounds++)
{
size_t outbytesleft = sizeof(outbuf);
char *outp = outbuf;
size_t r = yaz_iconv(cd, (char**) &inp, &inbytesleft,
&outp, &outbytesleft);
+ wrbuf_write(b, outbuf, outp - outbuf);
if (r == (size_t) (-1))
{
int e = yaz_iconv_error(cd);
if (e != YAZ_ICONV_E2BIG)
+ {
+ if (expect_error != -1)
+ if (e != expect_error)
+ ret = 0;
break;
+ }
+ }
+ else
+ {
+ size_t outbytesleft = sizeof(outbuf);
+ char *outp = outbuf;
+ r = yaz_iconv(cd, 0, 0, &outp, &outbytesleft);
+ wrbuf_write(b, outbuf, outp - outbuf);
+ if (expect_error != -1)
+ if (expect_error)
+ ret = 0;
+ break;
}
- wrbuf_write(b, outbuf, outp - outbuf);
}
if (wrbuf_len(b) == strlen(cmpbuf)
&& !memcmp(cmpbuf, wrbuf_buf(b), wrbuf_len(b)))
- ret = 1;
+ ;
else
- yaz_log(YLOG_LOG, "GOT (%.*s)", wrbuf_len(b), wrbuf_buf(b));
- wrbuf_free(b, 1);
+ {
+ WRBUF w = wrbuf_alloc();
+
+ ret = 0;
+ wrbuf_rewind(w);
+ wrbuf_puts_escaped(w, buf);
+ yaz_log(YLOG_LOG, "input %s", wrbuf_cstr(w));
+
+ wrbuf_rewind(w);
+ wrbuf_write_escaped(w, wrbuf_buf(b), wrbuf_len(b));
+ yaz_log(YLOG_LOG, "got %s", wrbuf_cstr(w));
+
+ wrbuf_rewind(w);
+ wrbuf_puts_escaped(w, cmpbuf);
+ yaz_log(YLOG_LOG, "exp %s", wrbuf_cstr(w));
+
+ wrbuf_destroy(w);
+ }
+
+ wrbuf_destroy(b);
return ret;
}
+static int tst_convert(yaz_iconv_t cd, const char *buf, const char *cmpbuf)
+{
+ return tst_convert_x(cd, buf, cmpbuf, 0);
+}
/* some test strings in ISO-8859-1 format */
static const char *iso_8859_1_a[] = {
return;
r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
YAZ_CHECK(r != (size_t) (-1));
+
+ r = yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
+ YAZ_CHECK(r != (size_t) (-1));
yaz_iconv_close(cd);
if (r == (size_t) (-1))
return;
outbytesleft = sizeof(outbuf1);
r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
YAZ_CHECK(r != (size_t) (-1));
+
+ r = yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
+ if (r == (size_t)(-1))
+ {
+ fprintf(stderr, "failed\n");
+ }
+ YAZ_CHECK(r != (size_t) (-1));
+
if (r != (size_t)(-1))
{
ret = compare_buffers("dconvert", i,
strlen(iso_8859_1_a[i]), iso_8859_1_a[i],
- sizeof(outbuf1) - outbytesleft, outbuf1);
+ sizeof(outbuf1) - outbytesleft, outbuf1);
YAZ_CHECK(ret);
}
yaz_iconv_close(cd);
/* COMBINING ACUTE ACCENT */
YAZ_CHECK(tst_convert(cd, "Cours de mathâe",
"Cours de mathe\xcc\x81"));
+
+ YAZ_CHECK(tst_convert(cd, "\xea" "a", "a\xcc\x8a"));
+ YAZ_CHECK(tst_convert(cd, "a" "\xea" "\x1e", "a" "\x1e\xcc\x8a"));
+ YAZ_CHECK(tst_convert(cd, "a" "\xea" "p", "a" "p\xcc\x8a"));
+
+ YAZ_CHECK(tst_convert_x(cd, "a\xea", "a", YAZ_ICONV_EINVAL));
+ YAZ_CHECK(tst_convert(cd, "p", "\xcc\x8a")); /* note: missing p */
+ yaz_iconv(cd, 0, 0, 0, 0); /* incomplete. so we have to reset */
+
+ /* bug #2115 */
+ YAZ_CHECK(tst_convert(cd, ESC "(N" ESC ")Qp" ESC "(B", "\xd0\x9f"));
+
+ YAZ_CHECK(tst_convert_x(cd, ESC , "", YAZ_ICONV_EINVAL));
+ YAZ_CHECK(tst_convert_x(cd, ESC "(", "", YAZ_ICONV_EINVAL));
+ YAZ_CHECK(tst_convert_x(cd, ESC "(B", "", 0));
+
+ YAZ_CHECK(tst_convert(cd, ESC "(B" "\x31", "1")); /* ASCII in G0 */
+ YAZ_CHECK(tst_convert(cd, ESC ")B" "\xB1", "1")); /* ASCII in G1 */
+
yaz_iconv_close(cd);
}
yaz_iconv_close(cd);
}
-static void tst_utf8_to_marc8(void)
+static void tst_utf8_to_marc8(const char *marc8_type)
{
- yaz_iconv_t cd = yaz_iconv_open("MARC8", "UTF-8");
+ yaz_iconv_t cd = yaz_iconv_open(marc8_type, "UTF-8");
YAZ_CHECK(cd);
if (!cd)
/** Pure ASCII. 12 characters (sizeof(outbuf)) */
YAZ_CHECK(tst_convert(cd, "Cours de math", "Cours de math"));
- /** Pure ASCII. 13 characters (sizeof(outbuf)) */
+ /** Pure ASCII. 13 characters (sizeof(outbuf)+1) */
YAZ_CHECK(tst_convert(cd, "Cours de math.", "Cours de math."));
/** UPPERCASE SCANDINAVIAN O */
"\033$1" "\x21\x2B\x3B" "\033(B" "o" ));
+ /** Ideographic space per ANSI Z39.64 */
+ YAZ_CHECK(tst_convert(cd,
+ "\xe3\x80\x80" "o", /* UTF-8 */
+ "\033$1" "\x21\x23\x21" "\033(B" "o" ));
+
/** Superscript 0 . bug #642 */
YAZ_CHECK(tst_convert(cd,
"(\xe2\x81\xb0)", /* UTF-8 */
"(\033p0\x1bs)"));
+
+ /** bug #1778 */
+ YAZ_CHECK(tst_convert(cd,
+ /* offset 0x530 in UTF-8 rec marccol4.u8.marc */
+ "\xE3\x83\xB3" "\xE3\x82\xBF"
+ "\xCC\x84" "\xCC\x84" "\xE3\x83\xBC" /* UTF-8 */,
+ "\x1B\x24\x31" "\x69\x25\x73"
+ "\x1B\x28\x42" "\xE5\xE5" "\x1B\x24\x31"
+ "\x69\x25\x3F"
+ "\x69\x21\x3C" "\x1B\x28\x42"));
+
+
+ /** bug #2120 */
+ YAZ_CHECK(tst_convert(cd,
+ "\xCE\x94\xCE\xB5\xCF\x84"
+ "\xCE\xBF\xCF\x81\xCE\xB1"
+ "\xCE\xBA\xCE\xB7\xCF\x82\x2C",
+
+ "\x1B\x28\x53\x45\x66\x78\x72\x75"
+ "\x61\x6D\x6A\x77"
+ "\x1B\x28\x42\x2C"
+ ));
+ {
+ char *inbuf0 = "\xe2\x81\xb0";
+ char *inbuf = inbuf0;
+ size_t inbytesleft = strlen(inbuf);
+ char outbuf0[64];
+ char *outbuf = outbuf0;
+ size_t outbytesleft = sizeof(outbuf0)-1;
+ size_t r;
+#if 0
+ int i;
+#endif
+ r = yaz_iconv(cd, &inbuf, &inbytesleft, &outbuf, &outbytesleft);
+ YAZ_CHECK(r != (size_t) (-1));
+
+#if 0
+ *outbuf = '\0'; /* so we know when to stop printing */
+ for (i = 0; outbuf0[i]; i++)
+ {
+ int ch = outbuf0[i] & 0xff;
+ yaz_log(YLOG_LOG, "ch%d %02X %c", i, ch, ch >= ' ' ? ch : '?');
+ }
+#endif
+
+ r = yaz_iconv(cd, 0, 0, &outbuf, &outbytesleft);
+ YAZ_CHECK(r != (size_t) (-1));
+ *outbuf = '\0'; /* for strcmp test below and printing */
+#if 0
+ for (i = 0; outbuf0[i]; i++)
+ {
+ int ch = outbuf0[i] & 0xff;
+ yaz_log(YLOG_LOG, "ch%d %02X %c", i, ch, ch >= ' ' ? ch : '?');
+ }
+#endif
+ YAZ_CHECK(strcmp("\033p0\x1bs", outbuf0) == 0);
+ }
yaz_iconv_close(cd);
}
+static void tst_advance_to_utf8(void)
+{
+ yaz_iconv_t cd = yaz_iconv_open("utf-8", "advancegreek");
+
+ YAZ_CHECK(cd);
+ if (!cd)
+ return;
+
+ YAZ_CHECK(tst_convert(cd, "Cours ", "Cours "));
+ yaz_iconv_close(cd);
+}
+
+static void tst_utf8_to_advance(void)
+{
+ yaz_iconv_t cd = yaz_iconv_open("advancegreek", "utf-8");
+
+ YAZ_CHECK(cd);
+ if (!cd)
+ return;
+
+ YAZ_CHECK(tst_convert(cd, "Cours ", "Cours "));
+ yaz_iconv_close(cd);
+}
static void tst_latin1_to_marc8(void)
{
tst_marc8_to_latin1();
- tst_utf8_to_marc8();
+ tst_advance_to_utf8();
+ tst_utf8_to_advance();
+
+ tst_utf8_to_marc8("marc8");
+ tst_utf8_to_marc8("marc8lossy");
+ tst_utf8_to_marc8("marc8lossless");
tst_latin1_to_marc8();