-/* $Id: test_icu_I18N.c,v 1.7 2007-05-02 14:01:36 marc Exp $
+/* $Id: test_icu_I18N.c,v 1.25 2007-05-24 10:35:21 adam Exp $
Copyright (c) 2006-2007, Index Data.
-This file is part of Pazpar2.
+ This file is part of Pazpar2.
-Pazpar2 is free software; you can redistribute it and/or modify it under
-the terms of the GNU General Public License as published by the Free
-Software Foundation; either version 2, or (at your option) any later
-version.
+ Pazpar2 is free software; you can redistribute it and/or modify it under
+ the terms of the GNU General Public License as published by the Free
+ Software Foundation; either version 2, or (at your option) any later
+ version.
-Pazpar2 is distributed in the hope that it will be useful, but WITHOUT ANY
-WARRANTY; without even the implied warranty of MERCHANTABILITY or
-FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
-for more details.
+ Pazpar2 is distributed in the hope that it will be useful, but WITHOUT ANY
+ WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ for more details.
-You should have received a copy of the GNU General Public License
-along with Pazpar2; see the file LICENSE. If not, write to the
-Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA
-02111-1307, USA.
- */
+ You should have received a copy of the GNU General Public License
+ along with Pazpar2; see the file LICENSE. If not, write to the
+ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA
+ 02111-1307, USA.
+*/
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
#ifdef HAVE_ICU
#include "icu_I18N.h"
+
#include <string.h>
#include <stdlib.h>
-
+//#include <unicode/ustring.h>
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
-int test_icu_casemap(const char * locale, char action,
- const char * src8, const char * check8)
+
+#define MAX_KEY_SIZE 256
+struct icu_termmap
{
- NMEM nmem = nmem_create();
- size_t buf_cap = 128;
- char buf[buf_cap];
- const char * dest8 = 0;
- size_t dest8_len = 0;
- //size_t src8_len = strlen(src8);
- int sucess = 0;
+ uint8_t sort_key[MAX_KEY_SIZE]; // standard C string '\0' terminated
+ char disp_term[MAX_KEY_SIZE]; // standard C utf-8 string
+};
- //printf("original string: '%s' (%d)\n", src8, (int) src8_len);
- //these shall succeed
- dest8 = icu_casemap(nmem, buf, buf_cap, &dest8_len,
- src8, locale, action);
+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;
+
+ int cmp = 0;
+
+ cmp = strcmp((const char *)itmp1->sort_key,
+ (const char *)itmp2->sort_key);
+ return cmp;
+};
- //printf("icu_casemap '%s:%c' '%s' (%d)\n",
- // locale, action, dest8, (int) dest8_len);
- if (dest8
- && (dest8_len == strlen(check8))
- && !strcmp(check8, dest8))
- sucess = dest8_len;
- nmem_destroy(nmem);
- return sucess;
+int test_icu_casemap(const char * locale, char action,
+ const char * src8cstr, const char * chk8cstr)
+{
+ int success = 0;
+ UErrorCode status = U_ZERO_ERROR;
+
+ struct icu_buf_utf8 * src8 = icu_buf_utf8_create(0);
+ struct icu_buf_utf8 * dest8 = icu_buf_utf8_create(0);
+ struct icu_buf_utf16 * src16 = icu_buf_utf16_create(0);
+ struct icu_buf_utf16 * dest16 = icu_buf_utf16_create(0);
+
+
+ int src8cstr_len = strlen(src8cstr);
+ int chk8cstr_len = strlen(chk8cstr);
+
+ // converting to UTF16
+ icu_utf16_from_utf8_cstr(src16, src8cstr, &status);
+
+ // perform case mapping
+ icu_utf16_casemap(dest16, src16, locale, action, &status);
+
+ // converting to UTF8
+ icu_utf16_to_utf8(dest8, dest16, &status);
+
+
+
+ // determine success
+ if (dest8->utf8
+ && (dest8->utf8_len == strlen(chk8cstr))
+ && !strcmp(chk8cstr, (const char *) dest8->utf8))
+ success = 1;
+ else
+ 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);
+ }
+
+ // clean the buffers
+ icu_buf_utf8_destroy(src8);
+ 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)
"A ReD fOx hunTS sQUirriLs",
"a red fox hunts squirrils"));
- // this one fails and needs more investigation ..
- YAZ_CHECK(0 == test_icu_casemap("en", 't',
+ YAZ_CHECK(test_icu_casemap("en", 't',
"A ReD fOx hunTS sQUirriLs",
"A Red Fox Hunts Squirrils"));
-
+
// Locale 'da'
// sucess expected
YAZ_CHECK(test_icu_casemap("de", 'l',
- "zWÖlf ärgerliche Würste rollen ÜBer die StRAße",
- "zwölf ärgerliche würste rollen über die straße"));
+ "zWÖlf ärgerliche Würste rollen ÜBer die StRAße",
+ "zwölf ärgerliche würste rollen über die straße"));
YAZ_CHECK(test_icu_casemap("de", 'u',
- "zWÖlf ärgerliche Würste rollen ÜBer die StRAße",
- "ZWÖLF ÄRGERLICHE WÜRSTE ROLLEN ÜBER DIE STRASSE"));
+ "zWÖlf ärgerliche Würste rollen ÜBer die StRAße",
+ "ZWÖLF ÄRGERLICHE WÜRSTE ROLLEN ÜBER DIE STRASSE"));
YAZ_CHECK(test_icu_casemap("de", 'f',
- "zWÖlf ärgerliche Würste rollen ÜBer die StRAße",
- "zwölf ärgerliche würste rollen über die strasse"));
+ "zWÖlf ärgerliche Würste rollen ÜBer die StRAße",
+ "zwölf ärgerliche würste rollen über die strasse"));
YAZ_CHECK(test_icu_casemap("de", 't',
- "zWÖlf ärgerliche Würste rollen ÜBer die StRAße",
- "Zwölf Ärgerliche Würste Rollen Über Die Straße"));
+ "zWÖlf ärgerliche Würste rollen ÜBer die StRAße",
+ "Zwölf Ärgerliche Würste Rollen Über Die Straße"));
}
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
-void test_icu_I18N_casemap_failures(int argc, char **argv)
+int test_icu_sortmap(const char * locale, int src_list_len,
+ const char ** src_list, const char ** chk_list)
{
+ int success = 1;
- size_t buf_cap = 128;
- char buf[buf_cap];
- size_t dest8_len = 0;
- NMEM nmem = nmem_create();
- char * dest8 = 0;
+ UErrorCode status = U_ZERO_ERROR;
- const char * src8 = "A ReD fOx hunTS sQUirriLs";
- //size_t src8_len = strlen(src8);
-
- //printf("original string: '%s' (%d)\n", src8, (int) src8_len);
-
- // some calling error needs investigation
- dest8 = icu_casemap(nmem, buf, buf_cap, &dest8_len,
- src8, "en", 't');
- YAZ_CHECK(0 == dest8_len);
- //printf("icu_casemap 'en:t' '%s' (%d)\n", dest8, (int) dest8_len);
-
-
- // attention: does not fail even if no locale 'xy_zz' defined
- // it seems to default to english locale
- dest8 = icu_casemap(nmem, buf, buf_cap, &dest8_len,
- src8, "zz_abc", 'l');
- YAZ_CHECK(dest8_len);
- //printf("icu_casemap 'zz:l' '%s' (%d)\n", dest8, (int) dest8_len);
-
-
- // shall fail - no buf buffer defined
- dest8 = icu_casemap(nmem, 0, buf_cap, &dest8_len,
- src8, "en", 'l');
- YAZ_CHECK(0 == dest8_len);
- //printf("icu_casemap 'en:l' '%s' (%d)\n", dest8, (int) dest8_len);
-
- // shall fail - no buf_cap defined
- dest8 = icu_casemap(nmem, buf, 0, &dest8_len,
- src8, "en", 'l');
- YAZ_CHECK(0 == dest8_len);
- //printf("icu_casemap 'en:l' '%s' (%d)\n", dest8, (int) dest8_len);
-
- // shall fail - no action 'x' defined
- dest8 = icu_casemap(nmem, buf, buf_cap, &dest8_len,
- src8, "en", 'x');
- YAZ_CHECK(0 == dest8_len);
- //printf("icu_casemap 'en:x' '%s' (%d)\n", dest8, (int) dest8_len);
-
- nmem_destroy(nmem);
-}
+ struct icu_buf_utf8 * buf8 = icu_buf_utf8_create(0);
+ struct icu_buf_utf16 * buf16 = icu_buf_utf16_create(0);
-// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
+ int i;
+
+ struct icu_termmap * list[src_list_len];
+
+ UCollator *coll = ucol_open(locale, &status);
+ icu_check_status(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++)
+ {
+
+ list[i] = (struct icu_termmap *) malloc(sizeof(struct icu_termmap));
+
+ // 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);
+
+ // 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);
+ //strncpy(list[i]->sort_key, buf8->utf8, buf8->utf8_len);
+ //strcpy((char *) list[i]->sort_key, (const char *) buf8->utf8);
+ }
-int test_icu_sortmap(const char * locale, size_t list_len,
- const char ** src8_list, const char ** check8_list)
-{
- int sucess = 1;
-
- size_t i = 0;
-
-
- NMEM nmem = nmem_create();
- size_t buf_cap = 128;
- char buf[buf_cap];
- struct icu_termmap ** dest8_list
- = nmem_malloc(nmem, sizeof(struct icu_termmap *) * list_len);
- //size_t dest8_len = 0;
- //size_t src8_len = strlen(src8);
-
- // initializing icu_termmap
- for (i = 0; i < list_len; i++){
- dest8_list[i] = icu_termmap_create(nmem);
- dest8_list[i]->norm_term = nmem_strdup(nmem, src8_list[i]);
- dest8_list[i]->disp_term = nmem_strdup(nmem, src8_list[i]);
- //dest8_list[i]->sort_key = nmem_strdup(nmem, src8_list[i]);
- //dest8_list[i]->sort_len = strlen(src8_list[i]);
- dest8_list[i]->sort_key
- = icu_sortmap(nmem, buf, buf_cap, 0, src8_list[i], locale);
- // = icu_sortmap(nmem, buf, buf_cap, &(dest8_list[i]->sort_len),
- // src8_list[i], locale);
- }
// do the sorting
- qsort(dest8_list, list_len,
+ qsort(list, src_list_len,
sizeof(struct icu_termmap *), icu_termmap_cmp);
// checking correct sorting
- for (i = 0; i < list_len; i++){
- if (0 != strcmp(dest8_list[i]->disp_term, check8_list[i])){
- sucess = 0;
+ for (i = 0; i < src_list_len; i++){
+ if (0 != strcmp(list[i]->disp_term, chk_list[i])){
+ success = 0;
}
}
- if (1 || !sucess){
- printf("\n");
- printf("Input '%s':", locale);
- for (i = 0; i < list_len; i++)
- printf(" '%s'", src8_list[i]);
- printf("\n");
- printf("ICU sort '%s':", locale);
- for (i = 0; i < list_len; i++)
- printf(" '%s'", dest8_list[i]->disp_term);
- if (sucess)
- printf(" OK");
- else
- printf(" ERROR ??");
+ if(!success){
+ printf("\nERROR\n");
+ printf("Input str: '%s' : ", locale);
+ for (i = 0; i < src_list_len; i++) {
+ printf(" '%s'", list[i]->disp_term);
+ }
printf("\n");
- printf("Expected '%s':", locale);
- for (i = 0; i < list_len; i++)
- printf(" '%s'", check8_list[i]);
- printf("\n");
+ printf("ICU sort: '%s' : ", locale);
+ for (i = 0; i < src_list_len; i++) {
+ printf(" '%s'", list[i]->disp_term);
+ //printf("(%d|%d)", list[i]->sort_key[0],list[i]->sort_key[1]);
+ }
+ printf("\n");
+ printf("Expected: '%s' : ", locale);
+ for (i = 0; i < src_list_len; i++) {
+ printf(" '%s'", chk_list[i]);
+ }
+ printf("\n");
}
+
+
+
+ for( i = 0; i < src_list_len; i++)
+ free(list[i]);
+
- nmem_destroy(nmem);
+ ucol_close(coll);
- return sucess;
+ icu_buf_utf8_destroy(buf8);
+ icu_buf_utf16_destroy(buf16);
+
+ return success;
}
// sucessful tests
size_t en_1_len = 6;
const char * en_1_src[6] = {"z", "K", "a", "A", "Z", "k"};
- const char * en_1_cck[6] = {"a", "A", "K", "k", "z", "Z"};
+ const char * en_1_cck[6] = {"a", "A", "k", "K", "z", "Z"};
YAZ_CHECK(test_icu_sortmap("en", en_1_len, en_1_src, en_1_cck));
- YAZ_CHECK(0 == test_icu_sortmap("en_AU", en_1_len, en_1_src, en_1_cck));
- YAZ_CHECK(0 == test_icu_sortmap("en_CA", en_1_len, en_1_src, en_1_cck));
- YAZ_CHECK(0 == test_icu_sortmap("en_GB", en_1_len, en_1_src, en_1_cck));
- YAZ_CHECK(0 == test_icu_sortmap("en_US", en_1_len, en_1_src, en_1_cck));
+ YAZ_CHECK(test_icu_sortmap("en_AU", en_1_len, en_1_src, en_1_cck));
+ YAZ_CHECK(test_icu_sortmap("en_CA", en_1_len, en_1_src, en_1_cck));
+ YAZ_CHECK(test_icu_sortmap("en_GB", en_1_len, en_1_src, en_1_cck));
+ YAZ_CHECK(test_icu_sortmap("en_US", en_1_len, en_1_src, en_1_cck));
- // sucessful tests - this one fails and should not!!!
+ // sucessful tests
size_t da_1_len = 6;
const char * da_1_src[6] = {"z", "å", "o", "æ", "a", "ø"};
const char * da_1_cck[6] = {"a", "o", "z", "æ", "ø", "å"};
- YAZ_CHECK(0 == test_icu_sortmap("da", da_1_len, da_1_src, da_1_cck));
- YAZ_CHECK(0 == test_icu_sortmap("da_DK", da_1_len, da_1_src, da_1_cck));
+ YAZ_CHECK(test_icu_sortmap("da", da_1_len, da_1_src, da_1_cck));
+ YAZ_CHECK(test_icu_sortmap("da_DK", da_1_len, da_1_src, da_1_cck));
// sucessful tests
size_t de_1_len = 9;
const char * de_1_src[9] = {"u", "ä", "o", "t", "s", "ß", "ü", "ö", "a"};
- const char * de_1_cck[9] = {"ä", "a", "o", "ö", "s", "ß", "t", "u", "ü"};
+ const char * de_1_cck[9] = {"a","ä", "o", "ö", "s", "ß", "t", "u", "ü"};
YAZ_CHECK(test_icu_sortmap("de", de_1_len, de_1_src, de_1_cck));
- YAZ_CHECK(0 == test_icu_sortmap("de_AT", de_1_len, de_1_src, de_1_cck));
- YAZ_CHECK(0 == test_icu_sortmap("de_DE", de_1_len, de_1_src, de_1_cck));
+ 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)
+{
+ int success = 0;
+
+ UErrorCode status = U_ZERO_ERROR;
+
+ 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);
+ icu_check_status(status);
+
+ icu_utf16_from_utf8_cstr(src16, src8cstr, &status);
+ icu_check_status(status);
+
+ icu_normalizer_normalize(normalizer, dest16, src16, &status);
+ icu_check_status(status);
+
+ icu_utf16_to_utf8(dest8, dest16, &status);
+ icu_check_status(status);
+
+
+ if(!strcmp((const char *) dest8->utf8,
+ (const char *) chk8cstr))
+ success = 1;
+ 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);
+ }
+
+
+ icu_normalizer_destroy(normalizer);
+ 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)
+{
+
+ YAZ_CHECK(test_icu_normalizer("[:Punctuation:] Any-Remove",
+ "Don't shoot!",
+ "Dont shoot"));
+
+ YAZ_CHECK(test_icu_normalizer("[:Control:] Any-Remove",
+ "Don't\n shoot!",
+ "Don't shoot!"));
+
+ YAZ_CHECK(test_icu_normalizer("[:Decimal_Number:] Any-Remove",
+ "This is 4 you!",
+ "This is you!"));
+
+ YAZ_CHECK(test_icu_normalizer("Lower; [:^Letter:] Remove",
+ "Don't shoot!",
+ "dontshoot"));
+
+ YAZ_CHECK(test_icu_normalizer("[:^Number:] Remove",
+ "Monday 15th of April",
+ "15"));
+
+ YAZ_CHECK(test_icu_normalizer("Lower;"
+ "[[:WhiteSpace:][:Punctuation:]] Remove",
+ " 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)
+{
+ int success = 1;
+
+ UErrorCode status = U_ZERO_ERROR;
+ struct icu_buf_utf16 * src16 = icu_buf_utf16_create(0);
+ struct icu_buf_utf16 * tkn16 = icu_buf_utf16_create(0);
+ struct icu_buf_utf8 * tkn8 = icu_buf_utf8_create(0);
+
+ //printf("Input: '%s'\n", src8cstr);
+
+ // transforming to UTF16
+ icu_utf16_from_utf8_cstr(src16, src8cstr, &status);
+ icu_check_status(status);
+
+ // set up tokenizer
+ struct icu_tokenizer * tokenizer
+ = icu_tokenizer_create(locale, action, &status);
+ icu_check_status(status);
+ YAZ_CHECK(tokenizer);
+
+ // attach text buffer to tokenizer
+ icu_tokenizer_attach(tokenizer, src16, &status);
+ icu_check_status(status);
+ YAZ_CHECK(tokenizer->bi);
+
+ // perform work on tokens
+ //printf("Tokens: ");
+ while(icu_tokenizer_next_token(tokenizer, tkn16, &status)){
+ icu_check_status(status);
+
+ // converting to UTF8
+ icu_utf16_to_utf8(tkn8, tkn16, &status);
+
+ //printf("token %d %d %d %d '%s'\n",
+ //
+ // icu_tokenizer_token_start(tokenizer),
+ // icu_tokenizer_token_end(tokenizer),
+ // icu_tokenizer_token_length(tokenizer),
+ // tkn8->utf8);
+ }
+
+ 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);
+ }
+
+ icu_tokenizer_destroy(tokenizer);
+ icu_buf_utf16_destroy(src16);
+ icu_buf_utf16_destroy(tkn16);
+ icu_buf_utf8_destroy(tkn8);
+
+ return success;
+}
+
+
+// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
+
+void test_icu_I18N_tokenizer(int argc, char **argv)
+{
+
+
+ const char * en_str
+ = "O Romeo, Romeo! wherefore art thou Romeo?";
+ YAZ_CHECK(test_icu_tokenizer("en", 's', en_str, 2));
+ YAZ_CHECK(test_icu_tokenizer("en", 'l', en_str, 7));
+ 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. "
+ "Den er med blåbær, men alle sommerens forskellige bær kan bruges.";
+
+ YAZ_CHECK(test_icu_tokenizer("da", 's', da_str, 3));
+ YAZ_CHECK(test_icu_tokenizer("dar", 'l', da_str, 17));
+ YAZ_CHECK(test_icu_tokenizer("da", 'w', da_str, 37));
+ YAZ_CHECK(test_icu_tokenizer("da", 'c', da_str, 110));
+
}
-#endif
+void test_icu_I18N_chain(int argc, char **argv)
+{
+ const char * en_str
+ = "O Romeo, Romeo! wherefore art thou\t Romeo?";
+
+ printf("ICU chain:\ninput: '%s'\n", en_str);
+
+ UErrorCode status = U_ZERO_ERROR;
+ //struct icu_chain_step * step = 0;
+ struct icu_chain * chain = 0;
+
+
+ const char * xml_str = "<icu_chain id=\"en:word\" locale=\"en\">"
+ "<normalize rule=\"[:Control:] Any-Remove\"/>"
+ "<tokenize rule=\"l\"/>"
+ "<normalize rule=\"[[:WhiteSpace:][:Punctuation:]] Remove\"/>"
+ "<display/>"
+ "<casemap rule=\"l\"/>"
+ "<normal/>"
+ "<sort/>"
+ "</icu_chain>";
+
+
+ xmlDoc *doc = xmlParseMemory(xml_str, strlen(xml_str));
+ xmlNode *xml_node = xmlDocGetRootElement(doc);
+ YAZ_CHECK(xml_node);
+
+
+ chain = icu_chain_xml_config(xml_node, &status);
+
+#if 0
+ chain = icu_chain_create((uint8_t *) "en:word", (uint8_t *) "en");
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_normalize,
+ (const uint8_t *) "[:Control:] Any-Remove",
+ &status);
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_tokenize,
+ (const uint8_t *) "s",
+ &status);
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_tokenize,
+ (const uint8_t *) "l",
+ &status);
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_normalize,
+ (const uint8_t *)
+ "[[:WhiteSpace:][:Punctuation:]] Any-Remove",
+ &status);
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_display,
+ (const uint8_t *)"",
+ &status);
+/* step = icu_chain_insert_step(chain, ICU_chain_step_type_normalize, */
+/* (const uint8_t *) "Lower", */
+/* &status); */
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_casemap,
+ (const uint8_t *) "l",
+ &status);
+ step = icu_chain_insert_step(chain, ICU_chain_step_type_norm,
+ (const uint8_t *)"",
+ &status);
+/* step = icu_chain_insert_step(chain, ICU_chain_step_type_sort, */
+/* (const uint8_t *)"", */
+/* &status); */
+
+#endif
+
+ xmlFreeDoc(doc);
+ YAZ_CHECK(chain);
+
+ YAZ_CHECK(icu_chain_assign_cstr(chain, en_str, &status));
+
+ while (icu_chain_next_token(chain, &status)){
+ printf("%d '%s' '%s'\n",
+ icu_chain_get_token_count(chain),
+ icu_chain_get_norm(chain),
+ icu_chain_get_display(chain));
+ }
+
+ YAZ_CHECK_EQ(icu_chain_get_token_count(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_get_token_count(chain),
+ icu_chain_get_norm(chain),
+ icu_chain_get_display(chain));
+ }
+
+
+ YAZ_CHECK_EQ(icu_chain_get_token_count(chain), 3);
+
+ icu_chain_destroy(chain);
+}
+
+
+void test_bug_1140(void)
+{
+ const char * en_str
+ = "O Romeo, Romeo! wherefore art thou\t Romeo?";
+
+ printf("ICU chain:\ninput: '%s'\n", en_str);
+
+ UErrorCode status = U_ZERO_ERROR;
+ //struct icu_chain_step * step = 0;
+ struct icu_chain * chain = 0;
+
+ const char * xml_str = "<icu_chain id=\"en:word\" locale=\"en\">"
+
+ /* if the first rule is normalize instead. Then it works */
+#if 0
+ "<normalize rule=\"[:Control:] Any-Remove\"/>"
+#endif
+ "<tokenize rule=\"l\"/>"
+ "<normalize rule=\"[[:WhiteSpace:][:Punctuation:]] Remove\"/>"
+ "<display/>"
+ "<casemap rule=\"l\"/>"
+ "<normal/>"
+ "<sort/>"
+ "</icu_chain>";
+
+
+ xmlDoc *doc = xmlParseMemory(xml_str, strlen(xml_str));
+ xmlNode *xml_node = xmlDocGetRootElement(doc);
+ YAZ_CHECK(xml_node);
+
+ chain = icu_chain_xml_config(xml_node, &status);
+
+ xmlFreeDoc(doc);
+ YAZ_CHECK(chain);
+
+ YAZ_CHECK(icu_chain_assign_cstr(
+ chain, "O Romeo, Romeo! wherefore art thou\t Romeo?",
+ &status));
+
+ while (icu_chain_next_token(chain, &status))
+ ;
+
+ YAZ_CHECK_EQ(icu_chain_get_token_count(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_get_token_count(chain),
+ icu_chain_get_norm(chain),
+ icu_chain_get_display(chain));
+ }
+
+ /* we expect 'what' 'is' 'this', i.e. 3 tokens */
+ YAZ_CHECK_EQ(icu_chain_get_token_count(chain), 3);
+
+ icu_chain_destroy(chain);
+}
+
+#endif // HAVE_ICU
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
#ifdef HAVE_ICU
- test_icu_I18N_casemap_failures(argc, argv);
+ //test_icu_I18N_casemap_failures(argc, argv);
test_icu_I18N_casemap(argc, argv);
test_icu_I18N_sortmap(argc, argv);
-
-#else
+ test_icu_I18N_normalizer(argc, argv);
+ test_icu_I18N_tokenizer(argc, argv);
+ test_icu_I18N_chain(argc, argv);
+#if 0
+ /* currently fails */
+ test_bug_1140();
+#endif
+
+#else // HAVE_ICU
printf("ICU unit tests omitted.\n"
"Please install libicu36-dev and icu-doc or similar\n");
YAZ_CHECK(0 == 0);
-#endif
+#endif // HAVE_ICU
YAZ_CHECK_TERM;
}
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
+
+
/*
* Local variables:
* c-basic-offset: 4