-/*
- * Copyright (C) 2005-2007, Index Data ApS
+/* This file is part of the YAZ toolkit.
+ * Copyright (C) 1995-2010 Index Data
* See the file LICENSE for details.
- *
- * $Id: tst_icu_I18N.c,v 1.15 2007-11-12 11:11:16 adam Exp $
- *
*/
/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
#endif
#include <yaz/test.h>
+#include <yaz/log.h>
+#include <yaz/wrbuf.h>
#if YAZ_HAVE_ICU
#include <yaz/icu_I18N.h>
+#if YAZ_POSIX_THREADS
+#include <pthread.h>
+#endif
+
#include <string.h>
#include <stdlib.h>
-/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
-
-
#define MAX_KEY_SIZE 256
struct icu_termmap
{
};
-
-int icu_termmap_cmp(const void *vp1, const void *vp2)
+static int icu_termmap_cmp(const void *vp1, const void *vp2)
{
struct icu_termmap *itmp1 = *(struct icu_termmap **) vp1;
struct icu_termmap *itmp2 = *(struct icu_termmap **) vp2;
cmp = strcmp((const char *)itmp1->sort_key,
(const char *)itmp2->sort_key);
return cmp;
-};
-
-
+}
-int test_icu_casemap(const char * locale, char action,
- const char * src8cstr, const char * chk8cstr)
+static int test_icu_casemap(const char * locale, char action,
+ const char * src8cstr, const char * chk8cstr)
{
int success = 0;
UErrorCode status = U_ZERO_ERROR;
/* converting to UTF8 */
icu_utf16_to_utf8(dest8, dest16, &status);
-
-
/* determine success */
if (dest8->utf8
&& (dest8->utf8_len == strlen(chk8cstr))
success = 0;
/* report failures */
- if (!success){
- printf("\nERROR\n");
- printf("original string: '%s' (%d)\n", src8cstr, src8cstr_len);
- printf("icu_casemap '%s:%c' '%s' (%d)\n",
- locale, action, dest8->utf8, dest8->utf8_len);
- printf("expected string: '%s' (%d)\n", chk8cstr, chk8cstr_len);
+ if (!success)
+ {
+ yaz_log(YLOG_WARN, "test_icu_casemap failed");
+ yaz_log(YLOG_LOG, "Original string: '%s' (%d)",
+ src8cstr, src8cstr_len);
+ yaz_log(YLOG_LOG, "icu_casemap '%s:%c' '%s' (%d)",
+ locale, action, dest8->utf8, dest8->utf8_len);
+ yaz_log(YLOG_LOG, "expected string: '%s' (%d)",
+ chk8cstr, chk8cstr_len);
}
/* clean the buffers */
icu_buf_utf8_destroy(dest8);
icu_buf_utf16_destroy(src16);
icu_buf_utf16_destroy(dest16);
-
-
+
return success;
}
-
-
-/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
-
-void test_icu_I18N_casemap(int argc, char **argv)
+static void check_icu_casemap(void)
{
-
/* Locale 'en' */
/* successful tests */
"A ReD fOx hunTS sQUirriLs",
"A Red Fox Hunts Squirrils"));
-
/* Locale 'da' */
/* success expected */
}
-
-/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
-
-int test_icu_sortmap(const char * locale, int src_list_len,
- const char ** src_list, const char ** chk_list)
+static int test_icu_sortmap(const char * locale, int src_list_len,
+ const char ** src_list, const char ** chk_list)
{
int success = 1;
UCollator *coll = ucol_open(locale, &status);
icu_check_status(status);
- if(U_FAILURE(status))
+ if (U_FAILURE(status))
return 0;
/* assigning display terms and sort keys using buf 8 and buf16 */
- for( i = 0; i < src_list_len; i++)
- {
+ for (i = 0; i < src_list_len; i++)
+ {
- list[i] = (struct icu_termmap *) malloc(sizeof(struct icu_termmap));
+ list[i] = (struct icu_termmap *) malloc(sizeof(struct icu_termmap));
- /* copy display term */
- strcpy(list[i]->disp_term, src_list[i]);
+ /* copy display term */
+ strcpy(list[i]->disp_term, src_list[i]);
- /* transforming to UTF16 */
- icu_utf16_from_utf8_cstr(buf16, list[i]->disp_term, &status);
- icu_check_status(status);
+ /* transforming to UTF16 */
+ icu_utf16_from_utf8_cstr(buf16, list[i]->disp_term, &status);
+ icu_check_status(status);
- /* computing sortkeys */
- icu_sortkey8_from_utf16(coll, buf8, buf16, &status);
- icu_check_status(status);
+ /* computing sortkeys */
+ icu_sortkey8_from_utf16(coll, buf8, buf16, &status);
+ icu_check_status(status);
- /* assigning sortkeys */
- memcpy(list[i]->sort_key, buf8->utf8, buf8->utf8_len);
- }
-
+ /* assigning sortkeys */
+ memcpy(list[i]->sort_key, buf8->utf8, buf8->utf8_len);
+ }
/* do the sorting */
- qsort(list, src_list_len,
- sizeof(struct icu_termmap *), icu_termmap_cmp);
+ qsort(list, src_list_len, sizeof(struct icu_termmap *), icu_termmap_cmp);
/* checking correct sorting */
- for (i = 0; i < src_list_len; i++){
+ for (i = 0; i < src_list_len; i++)
+ {
if (0 != strcmp(list[i]->disp_term, chk_list[i])){
success = 0;
}
}
- if(!success){
- printf("\nERROR\n");
- printf("Input str: '%s' : ", locale);
+ if (!success)
+ {
+ yaz_log(YLOG_LOG, "ERROR");
+ yaz_log(YLOG_LOG, "Input str:'%s':", locale);
for (i = 0; i < src_list_len; i++) {
- printf(" '%s'", list[i]->disp_term);
+ yaz_log(YLOG_LOG, " '%s'", list[i]->disp_term);
}
- printf("\n");
- printf("ICU sort: '%s' : ", locale);
+ yaz_log(YLOG_LOG, "ICU sort: '%s':", locale);
for (i = 0; i < src_list_len; i++) {
- printf(" '%s'", list[i]->disp_term);
+ yaz_log(YLOG_LOG, " '%s'", list[i]->disp_term);
}
- printf("\n");
- printf("Expected: '%s' : ", locale);
+ yaz_log(YLOG_LOG, "Expected: '%s':", locale);
for (i = 0; i < src_list_len; i++) {
- printf(" '%s'", chk_list[i]);
+ yaz_log(YLOG_LOG, " '%s'", chk_list[i]);
}
- printf("\n");
}
-
-
- for( i = 0; i < src_list_len; i++)
+ for (i = 0; i < src_list_len; i++)
free(list[i]);
-
-
+
ucol_close(coll);
icu_buf_utf8_destroy(buf8);
return success;
}
-
-/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
-
-void test_icu_I18N_sortmap(int argc, char **argv)
+static void check_icu_sortmap(void)
{
-
/* successful tests */
size_t en_1_len = 6;
const char * en_1_src[6] = {"z", "K", "a", "A", "Z", "k"};
YAZ_CHECK(test_icu_sortmap("de_AT", de_1_len, de_1_src, de_1_cck));
YAZ_CHECK(test_icu_sortmap("de_DE", de_1_len, de_1_src, de_1_cck));
}
-
}
-
-/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
-
-
-
-
-int test_icu_normalizer(const char * rules8cstr,
- const char * src8cstr,
- const char * chk8cstr)
+static int test_icu_normalizer(const char * rules8cstr,
+ const char * src8cstr,
+ const char * chk8cstr)
{
int success = 0;
struct icu_buf_utf16 * src16 = icu_buf_utf16_create(0);
struct icu_buf_utf16 * dest16 = icu_buf_utf16_create(0);
struct icu_buf_utf8 * dest8 = icu_buf_utf8_create(0);
- struct icu_normalizer * normalizer
- = icu_normalizer_create(rules8cstr, 'f', &status);
+ struct icu_transform * transform
+ = icu_transform_create(rules8cstr, 'f', 0, &status);
icu_check_status(status);
icu_utf16_from_utf8_cstr(src16, src8cstr, &status);
icu_check_status(status);
- icu_normalizer_normalize(normalizer, dest16, src16, &status);
+ icu_transform_trans(transform, dest16, src16, &status);
icu_check_status(status);
icu_utf16_to_utf8(dest8, dest16, &status);
icu_check_status(status);
- if(!strcmp((const char *) dest8->utf8,
+ if (!strcmp((const char *) dest8->utf8,
(const char *) chk8cstr))
success = 1;
- else {
+ else
+ {
success = 0;
- printf("Normalization\n");
- printf("Rules: '%s'\n", rules8cstr);
- printf("Input: '%s'\n", src8cstr);
- printf("Normalized: '%s'\n", dest8->utf8);
- printf("Expected: '%s'\n", chk8cstr);
+ yaz_log(YLOG_LOG, "Normalization");
+ yaz_log(YLOG_LOG, " Rules: '%s'", rules8cstr);
+ yaz_log(YLOG_LOG, " Input: '%s'", src8cstr);
+ yaz_log(YLOG_LOG, " Normalized: '%s'", dest8->utf8);
+ yaz_log(YLOG_LOG, " Expected: '%s'", chk8cstr);
}
-
- icu_normalizer_destroy(normalizer);
+ icu_transform_destroy(transform);
icu_buf_utf16_destroy(src16);
icu_buf_utf16_destroy(dest16);
icu_buf_utf8_destroy(dest8);
return success;
-};
-
-
-/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
+}
-void test_icu_I18N_normalizer(int argc, char **argv)
+static void check_icu_normalizer(void)
{
-
YAZ_CHECK(test_icu_normalizer("[:Punctuation:] Any-Remove",
"Don't shoot!",
"Dont shoot"));
" word4you? ",
"word4you"));
-
YAZ_CHECK(test_icu_normalizer("NFD; [:Nonspacing Mark:] Remove; NFC",
"à côté de l'alcôve ovoïde",
"a cote de l'alcove ovoide"));
-
}
-
-/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
-
-int test_icu_tokenizer(const char * locale, char action,
- const char * src8cstr, int count)
+static int test_icu_tokenizer(const char * locale, char action,
+ const char * src8cstr, int count)
{
int success = 1;
/* attach text buffer to tokenizer */
icu_tokenizer_attach(tokenizer, src16, &status);
icu_check_status(status);
- YAZ_CHECK(tokenizer->bi);
/* perform work on tokens */
- while(icu_tokenizer_next_token(tokenizer, tkn16, &status)){
+ while (icu_tokenizer_next_token(tokenizer, tkn16, &status))
+ {
icu_check_status(status);
/* converting to UTF8 */
icu_utf16_to_utf8(tkn8, tkn16, &status);
}
- if (count != icu_tokenizer_token_count(tokenizer)){
+ if (count != icu_tokenizer_token_count(tokenizer))
+ {
success = 0;
- printf("\nTokenizer '%s:%c' Error: \n", locale, action);
- printf("Input: '%s'\n", src8cstr);
- printf("Tokens: %d", icu_tokenizer_token_count(tokenizer));
- printf(", expected: %d\n", count);
+ yaz_log(YLOG_LOG, "Tokenizer '%s:%c' Error:", locale, action);
+ yaz_log(YLOG_LOG, " Input: '%s'", src8cstr);
+ yaz_log(YLOG_LOG, " Tokens: %d", icu_tokenizer_token_count(tokenizer));
+ yaz_log(YLOG_LOG, " Expected: %d", count);
}
icu_tokenizer_destroy(tokenizer);
return success;
}
-
-/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
-
-void test_icu_I18N_tokenizer(int argc, char **argv)
+static void check_icu_tokenizer(void)
{
-
-
const char * en_str
= "O Romeo, Romeo! wherefore art thou Romeo?";
YAZ_CHECK(test_icu_tokenizer("en", 'w', en_str, 16));
YAZ_CHECK(test_icu_tokenizer("en", 'c', en_str, 41));
-
-
{
const char * da_str
= "Blåbærtærte. Denne kage stammer fra Finland. "
YAZ_CHECK(test_icu_tokenizer("da", 'w', da_str, 37));
YAZ_CHECK(test_icu_tokenizer("da", 'c', da_str, 110));
}
-
}
-
-void test_icu_I18N_chain(int argc, char **argv)
+static void check_icu_chain(void)
{
const char * en_str
= "O Romeo, Romeo! wherefore art thou\t Romeo?";
UErrorCode status = U_ZERO_ERROR;
struct icu_chain * chain = 0;
-
const char * xml_str = "<icu locale=\"en\">"
"<transform rule=\"[:Control:] Any-Remove\"/>"
"<tokenize rule=\"l\"/>"
YAZ_CHECK(icu_chain_assign_cstr(chain, en_str, &status));
- while (icu_chain_next_token(chain, &status)){
- ;
- /* printf("%d '%s' '%s'\n",
- icu_chain_token_number(chain),
- icu_chain_token_norm(chain),
- icu_chain_token_display(chain)); */
+ while (icu_chain_next_token(chain, &status))
+ {
+ yaz_log(YLOG_LOG, "%d '%s' '%s'",
+ icu_chain_token_number(chain),
+ icu_chain_token_norm(chain),
+ icu_chain_token_display(chain));
}
YAZ_CHECK_EQ(icu_chain_token_number(chain), 7);
YAZ_CHECK(icu_chain_assign_cstr(chain, "what is this?", &status));
- while (icu_chain_next_token(chain, &status)){
- ;
- /* printf("%d '%s' '%s'\n",
- icu_chain_token_number(chain),
- icu_chain_token_norm(chain),
- icu_chain_token_display(chain)); */
+ while (icu_chain_next_token(chain, &status))
+ {
+ yaz_log(YLOG_LOG, "%d '%s' '%s'",
+ icu_chain_token_number(chain),
+ icu_chain_token_norm(chain),
+ icu_chain_token_display(chain));
}
}
-void test_bug_1140(void)
+static void check_bug_1140(void)
{
UErrorCode status = U_ZERO_ERROR;
struct icu_chain * chain = 0;
chain, "O Romeo, Romeo! wherefore art thou\t Romeo?",
&status));
- while (icu_chain_next_token(chain, &status)){
+ while (icu_chain_next_token(chain, &status))
+ {
;
/* printf("%d '%s' '%s'\n",
- icu_chain_token_number(chain),
- icu_chain_token_norm(chain),
- icu_chain_token_display(chain)); */
-
+ icu_chain_token_number(chain),
+ icu_chain_token_norm(chain),
+ icu_chain_token_display(chain)); */
}
YAZ_CHECK(icu_chain_assign_cstr(chain, "what is this?", &status));
- while (icu_chain_next_token(chain, &status)){
- ;
- /* printf("%d '%s' '%s'\n",
- icu_chain_token_number(chain),
- icu_chain_token_norm(chain),
- icu_chain_token_display(chain)); */
+ while (icu_chain_next_token(chain, &status))
+ {
+ ;
+ /* printf("%d '%s' '%s'\n",
+ icu_chain_token_number(chain),
+ icu_chain_token_norm(chain),
+ icu_chain_token_display(chain)); */
}
/* we expect 'what' 'is' 'this', i.e. 3 tokens */
}
-
-void test_chain_empty_token(void)
+static void check_chain_empty_token(void)
{
UErrorCode status = U_ZERO_ERROR;
struct icu_chain * chain = 0;
chain, "a string with 15 tokenss and 8 displays",
&status));
- while (icu_chain_next_token(chain, &status)){
+ while (icu_chain_next_token(chain, &status))
+ {
;
/* printf("%d '%s' '%s'\n",
- icu_chain_token_number(chain),
- icu_chain_token_norm(chain),
- icu_chain_token_display(chain)); */
+ icu_chain_token_number(chain),
+ icu_chain_token_norm(chain),
+ icu_chain_token_display(chain)); */
}
YAZ_CHECK_EQ(icu_chain_token_number(chain), 15);
icu_chain_destroy(chain);
}
-void test_chain_empty_chain(void)
+static void check_chain_empty_chain(void)
{
UErrorCode status = U_ZERO_ERROR;
struct icu_chain * chain = 0;
chain, src8,
&status));
- while (icu_chain_next_token(chain, &status)){
+ while (icu_chain_next_token(chain, &status))
+ {
;
/* printf("%d '%s' '%s'\n",
- icu_chain_token_number(chain),
- icu_chain_token_norm(chain),
- icu_chain_token_display(chain)); */
+ icu_chain_token_number(chain),
+ icu_chain_token_norm(chain),
+ icu_chain_token_display(chain)); */
}
YAZ_CHECK_EQ(icu_chain_token_number(chain), 1);
dest8 = (char *) icu_chain_token_norm(chain);
YAZ_CHECK_EQ(strcmp(src8, dest8), 0);
+ icu_chain_destroy(chain);
+}
+static void check_icu_iter1(void)
+{
+ UErrorCode status = U_ZERO_ERROR;
+ struct icu_chain * chain = 0;
+ xmlNode *xml_node;
+ yaz_icu_iter_t iter;
+
+ const char * xml_str = "<icu locale=\"en\">"
+ "<tokenize rule=\"w\"/>"
+ "<transform rule=\"[[:WhiteSpace:][:Punctuation:]] Remove\"/>"
+ "</icu>";
+
+ xmlDoc *doc = xmlParseMemory(xml_str, strlen(xml_str));
+ YAZ_CHECK(doc);
+ if (!doc)
+ return;
+ xml_node = xmlDocGetRootElement(doc);
+ YAZ_CHECK(xml_node);
+ if (!xml_node)
+ return ;
+
+ chain = icu_chain_xml_config(xml_node, 0, &status);
+
+ xmlFreeDoc(doc);
+ YAZ_CHECK(chain);
+
+ iter = icu_iter_create(chain);
+ icu_iter_first(iter, "a string with 15 tokens and 8 displays");
+ YAZ_CHECK(iter);
+ if (!iter)
+ return;
+ while (icu_iter_next(iter))
+ {
+ yaz_log(YLOG_LOG, "[%s]", icu_iter_get_norm(iter));
+ }
+ icu_iter_destroy(iter);
icu_chain_destroy(chain);
}
-#endif /* YAZ_HAVE_ICU */
+static int test_iter(struct icu_chain *chain, const char *input,
+ const char *expected)
+{
+ yaz_icu_iter_t iter = icu_iter_create(chain);
+ WRBUF result, second;
+ int success = 1;
-/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
+ if (!iter)
+ {
+ yaz_log(YLOG_WARN, "test_iter: input=%s !iter", input);
+ return 0;
+ }
-int main(int argc, char **argv)
+ if (icu_iter_next(iter))
+ {
+ yaz_log(YLOG_WARN, "test_iter: expecting 0 before icu_iter_first");
+ return 0;
+ }
+
+ result = wrbuf_alloc();
+ icu_iter_first(iter, input);
+ while (icu_iter_next(iter))
+ {
+ wrbuf_puts(result, "[");
+ wrbuf_puts(result, icu_iter_get_norm(iter));
+ wrbuf_puts(result, "]");
+ }
+
+ second = wrbuf_alloc();
+ icu_iter_first(iter, input);
+ while (icu_iter_next(iter))
+ {
+ wrbuf_puts(second, "[");
+ wrbuf_puts(second, icu_iter_get_norm(iter));
+ wrbuf_puts(second, "]");
+ }
+
+ icu_iter_destroy(iter);
+
+ if (strcmp(expected, wrbuf_cstr(result)))
+ {
+ yaz_log(YLOG_WARN, "test_iter: input=%s expected=%s got=%s",
+ input, expected, wrbuf_cstr(result));
+ success = 0;
+ }
+
+ if (strcmp(expected, wrbuf_cstr(second)))
+ {
+ yaz_log(YLOG_WARN, "test_iter: input=%s expected=%s got=%s (2nd)",
+ input, expected, wrbuf_cstr(second));
+ success = 0;
+ }
+
+ wrbuf_destroy(result);
+ wrbuf_destroy(second);
+ return success;
+}
+
+static void *iter_thread(void *p)
{
+ struct icu_chain *chain = (struct icu_chain *) p;
+ int i;
+
+ for (i = 0; i < 10000; i++)
+ {
+ YAZ_CHECK(test_iter(chain, "Adobe Acrobat Reader, 1991-1999.",
+ "[adobe][acrobat][reader][1991][][1999][]"));
+ }
+ return 0;
+}
+
+static void check_iter_threads(struct icu_chain *chain)
+{
+#if YAZ_POSIX_THREADS
+#define NO_THREADS 1
+
+ pthread_t t[NO_THREADS];
+ int i;
+
+ for (i = 0; i < NO_THREADS; i++)
+ pthread_create(t + i, 0, iter_thread, chain);
+
+ for (i = 0; i < NO_THREADS; i++)
+ pthread_join(t[i], 0);
+#endif
+}
+static void check_icu_iter2(void)
+{
+ UErrorCode status = U_ZERO_ERROR;
+ struct icu_chain * chain = 0;
+ xmlNode *xml_node;
+
+ const char * xml_str = "<icu locale=\"en\">"
+ "<transform rule=\"[:Control:] Any-Remove\"/>"
+ "<tokenize rule=\"l\"/>"
+ "<tokenize rule=\"w\"/>"
+ "<transform rule=\"[[:WhiteSpace:][:Punctuation:]] Remove\"/>"
+ "<display/>"
+ "<casemap rule=\"l\"/>"
+ "</icu>";
+
+ xmlDoc *doc = xmlParseMemory(xml_str, strlen(xml_str));
+ YAZ_CHECK(doc);
+ if (!doc)
+ return;
+ xml_node = xmlDocGetRootElement(doc);
+ YAZ_CHECK(xml_node);
+ if (!xml_node)
+ return ;
+
+ chain = icu_chain_xml_config(xml_node, 0, &status);
+
+ xmlFreeDoc(doc);
+ YAZ_CHECK(chain);
+
+ YAZ_CHECK(test_iter(chain, "Adobe Acrobat Reader, 1991-1999.",
+ "[adobe][acrobat][reader][1991][][1999][]"));
+
+ check_iter_threads(chain);
+
+ icu_chain_destroy(chain);
+}
+#endif /* YAZ_HAVE_ICU */
+
+int main(int argc, char **argv)
+{
YAZ_CHECK_INIT(argc, argv);
YAZ_CHECK_LOG();
#if YAZ_HAVE_ICU
- test_icu_I18N_casemap(argc, argv);
- test_icu_I18N_sortmap(argc, argv);
- test_icu_I18N_normalizer(argc, argv);
- test_icu_I18N_tokenizer(argc, argv);
- test_icu_I18N_chain(argc, argv);
- test_chain_empty_token();
- test_chain_empty_chain();
- test_bug_1140();
+ check_icu_casemap();
+ check_icu_sortmap();
+ check_icu_normalizer();
+ check_icu_tokenizer();
+ check_icu_chain();
+ check_chain_empty_token();
+ check_chain_empty_chain();
+ check_icu_iter1();
+ check_icu_iter2();
+
+ check_bug_1140();
#else /* YAZ_HAVE_ICU */
- printf("ICU unit tests omitted.\n"
- "Please install libicu36-dev and icu-doc or similar\n");
+ yaz_log(YLOG_LOG, "ICU unit tests omitted");
YAZ_CHECK(0 == 0);
#endif /* YAZ_HAVE_ICU */
YAZ_CHECK_TERM;
}
-
-/* DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8 */
-
-
-
/*
* Local variables:
* c-basic-offset: 4
+ * c-file-style: "Stroustrup"
* indent-tabs-mode: nil
* End:
* vim: shiftwidth=4 tabstop=8 expandtab
*/
+
projects / yaz-moved-to-github.git / blobdiff