-/* $Id: icu_I18N.c,v 1.6 2007-05-07 12:18:34 marc Exp $
+/* $Id: icu_I18N.c,v 1.7 2007-05-07 12:52:04 marc 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.
+*/
#if HAVE_CONFIG_H
#include "cconfig.h"
{
//if(U_FAILURE(status))
if(!U_SUCCESS(status))
- yaz_log(YLOG_WARN,
- "ICU: %d %s\n", status, u_errorName(status));
- return status;
+ yaz_log(YLOG_WARN,
+ "ICU: %d %s\n", status, u_errorName(status));
+ return status;
}
struct icu_buf_utf16 * icu_buf_utf16_create(size_t capacity)
{
- struct icu_buf_utf16 * buf16
- = (struct icu_buf_utf16 *) malloc(sizeof(struct icu_buf_utf16));
-
- buf16->utf16 = 0;
- buf16->utf16_len = 0;
- buf16->utf16_cap = 0;
-
- if (capacity > 0){
- buf16->utf16 = (UChar *) malloc(sizeof(UChar) * capacity);
- buf16->utf16[0] = (UChar) 0;
- buf16->utf16_cap = capacity;
- }
- return buf16;
+ struct icu_buf_utf16 * buf16
+ = (struct icu_buf_utf16 *) malloc(sizeof(struct icu_buf_utf16));
+
+ buf16->utf16 = 0;
+ buf16->utf16_len = 0;
+ buf16->utf16_cap = 0;
+
+ if (capacity > 0){
+ buf16->utf16 = (UChar *) malloc(sizeof(UChar) * capacity);
+ buf16->utf16[0] = (UChar) 0;
+ buf16->utf16_cap = capacity;
+ }
+ return buf16;
};
struct icu_buf_utf16 * icu_buf_utf16_resize(struct icu_buf_utf16 * buf16,
size_t capacity)
{
- if (buf16){
- if (capacity > 0){
- if (0 == buf16->utf16)
- buf16->utf16 = (UChar *) malloc(sizeof(UChar) * capacity);
- else
- buf16->utf16
- = (UChar *) realloc(buf16->utf16, sizeof(UChar) * capacity);
- buf16->utf16[0] = (UChar) 0;
- buf16->utf16_len = 0;
- buf16->utf16_cap = capacity;
- }
- else {
- if (buf16->utf16)
- free(buf16->utf16);
- buf16->utf16 = 0;
- buf16->utf16_len = 0;
- buf16->utf16_cap = 0;
+ if (buf16){
+ if (capacity > 0){
+ if (0 == buf16->utf16)
+ buf16->utf16 = (UChar *) malloc(sizeof(UChar) * capacity);
+ else
+ buf16->utf16
+ = (UChar *) realloc(buf16->utf16, sizeof(UChar) * capacity);
+ buf16->utf16[0] = (UChar) 0;
+ buf16->utf16_len = 0;
+ buf16->utf16_cap = capacity;
+ }
+ else {
+ if (buf16->utf16)
+ free(buf16->utf16);
+ buf16->utf16 = 0;
+ buf16->utf16_len = 0;
+ buf16->utf16_cap = 0;
+ }
}
- }
- return buf16;
+ return buf16;
};
void icu_buf_utf16_destroy(struct icu_buf_utf16 * buf16)
{
- if (buf16){
- if (buf16->utf16)
- free(buf16->utf16);
- free(buf16);
- }
+ if (buf16){
+ if (buf16->utf16)
+ free(buf16->utf16);
+ free(buf16);
+ }
};
struct icu_buf_utf8 * icu_buf_utf8_create(size_t capacity)
{
- struct icu_buf_utf8 * buf8
- = (struct icu_buf_utf8 *) malloc(sizeof(struct icu_buf_utf8));
-
- buf8->utf8 = 0;
- buf8->utf8_len = 0;
- buf8->utf8_cap = 0;
-
- if (capacity > 0){
- buf8->utf8 = (uint8_t *) malloc(sizeof(uint8_t) * capacity);
- buf8->utf8[0] = (uint8_t) 0;
- buf8->utf8_cap = capacity;
- }
- return buf8;
+ struct icu_buf_utf8 * buf8
+ = (struct icu_buf_utf8 *) malloc(sizeof(struct icu_buf_utf8));
+
+ buf8->utf8 = 0;
+ buf8->utf8_len = 0;
+ buf8->utf8_cap = 0;
+
+ if (capacity > 0){
+ buf8->utf8 = (uint8_t *) malloc(sizeof(uint8_t) * capacity);
+ buf8->utf8[0] = (uint8_t) 0;
+ buf8->utf8_cap = capacity;
+ }
+ return buf8;
};
struct icu_buf_utf8 * icu_buf_utf8_resize(struct icu_buf_utf8 * buf8,
size_t capacity)
{
- if (buf8){
- if (capacity > 0){
- if (0 == buf8->utf8)
- buf8->utf8 = (uint8_t *) malloc(sizeof(uint8_t) * capacity);
- else
- buf8->utf8
- = (uint8_t *) realloc(buf8->utf8, sizeof(uint8_t) * capacity);
- buf8->utf8[0] = (uint8_t) 0;
- buf8->utf8_len = 0;
- buf8->utf8_cap = capacity;
- }
- else {
- if (buf8->utf8)
- free(buf8->utf8);
- buf8->utf8 = 0;
- buf8->utf8_len = 0;
- buf8->utf8_cap = 0;
+ if (buf8){
+ if (capacity > 0){
+ if (0 == buf8->utf8)
+ buf8->utf8 = (uint8_t *) malloc(sizeof(uint8_t) * capacity);
+ else
+ buf8->utf8
+ = (uint8_t *) realloc(buf8->utf8, sizeof(uint8_t) * capacity);
+ buf8->utf8[0] = (uint8_t) 0;
+ buf8->utf8_len = 0;
+ buf8->utf8_cap = capacity;
+ }
+ else {
+ if (buf8->utf8)
+ free(buf8->utf8);
+ buf8->utf8 = 0;
+ buf8->utf8_len = 0;
+ buf8->utf8_cap = 0;
+ }
}
- }
- return buf8;
+ return buf8;
};
void icu_buf_utf8_destroy(struct icu_buf_utf8 * buf8)
{
- if (buf8){
- if (buf8->utf8)
- free(buf8->utf8);
- free(buf8);
- }
+ if (buf8){
+ if (buf8->utf8)
+ free(buf8->utf8);
+ free(buf8);
+ }
};
struct icu_buf_utf8 * src8,
UErrorCode * status)
{
- int32_t utf16_len = 0;
-
- u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
- &utf16_len,
- (const char *) src8->utf8, src8->utf8_len, status);
+ int32_t utf16_len = 0;
- // check for buffer overflow, resize and retry
- if (*status == U_BUFFER_OVERFLOW_ERROR
- //|| dest16->utf16_len > dest16->utf16_cap
- ){
- icu_buf_utf16_resize(dest16, utf16_len * 2);
- *status = U_ZERO_ERROR;
u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
&utf16_len,
(const char *) src8->utf8, src8->utf8_len, status);
- }
+
+ // check for buffer overflow, resize and retry
+ if (*status == U_BUFFER_OVERFLOW_ERROR
+ //|| dest16->utf16_len > dest16->utf16_cap
+ ){
+ icu_buf_utf16_resize(dest16, utf16_len * 2);
+ *status = U_ZERO_ERROR;
+ u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
+ &utf16_len,
+ (const char *) src8->utf8, src8->utf8_len, status);
+ }
//if (*status != U_BUFFER_OVERFLOW_ERROR
- if (U_SUCCESS(*status)
- && utf16_len < dest16->utf16_cap)
- dest16->utf16_len = utf16_len;
- else {
- dest16->utf16[0] = (UChar) 0;
- dest16->utf16_len = 0;
- }
+ if (U_SUCCESS(*status)
+ && utf16_len < dest16->utf16_cap)
+ dest16->utf16_len = utf16_len;
+ else {
+ dest16->utf16[0] = (UChar) 0;
+ dest16->utf16_len = 0;
+ }
- return *status;
+ return *status;
};
const char * src8cstr,
UErrorCode * status)
{
- size_t src8cstr_len = 0;
- int32_t utf16_len = 0;
+ size_t src8cstr_len = 0;
+ int32_t utf16_len = 0;
- src8cstr_len = strlen(src8cstr);
-
- u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
- &utf16_len,
- src8cstr, src8cstr_len, status);
+ src8cstr_len = strlen(src8cstr);
- // check for buffer overflow, resize and retry
- if (*status == U_BUFFER_OVERFLOW_ERROR
- //|| dest16->utf16_len > dest16->utf16_cap
- ){
- icu_buf_utf16_resize(dest16, utf16_len * 2);
- *status = U_ZERO_ERROR;
u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
&utf16_len,
src8cstr, src8cstr_len, status);
- }
-
- // if (*status != U_BUFFER_OVERFLOW_ERROR
- if (U_SUCCESS(*status)
- && utf16_len < dest16->utf16_cap)
- dest16->utf16_len = utf16_len;
- else {
- dest16->utf16[0] = (UChar) 0;
- dest16->utf16_len = 0;
- }
- return *status;
+ // check for buffer overflow, resize and retry
+ if (*status == U_BUFFER_OVERFLOW_ERROR
+ //|| dest16->utf16_len > dest16->utf16_cap
+ ){
+ icu_buf_utf16_resize(dest16, utf16_len * 2);
+ *status = U_ZERO_ERROR;
+ u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
+ &utf16_len,
+ src8cstr, src8cstr_len, status);
+ }
+
+ // if (*status != U_BUFFER_OVERFLOW_ERROR
+ if (U_SUCCESS(*status)
+ && utf16_len < dest16->utf16_cap)
+ dest16->utf16_len = utf16_len;
+ else {
+ dest16->utf16[0] = (UChar) 0;
+ dest16->utf16_len = 0;
+ }
+
+ return *status;
};
UErrorCode icu_utf16_to_utf8(struct icu_buf_utf8 * dest8,
- struct icu_buf_utf16 * src16,
- UErrorCode * status)
+ struct icu_buf_utf16 * src16,
+ UErrorCode * status)
{
- int32_t utf8_len = 0;
+ int32_t utf8_len = 0;
- u_strToUTF8((char *) dest8->utf8, dest8->utf8_cap,
- &utf8_len,
- src16->utf16, src16->utf16_len, status);
-
- // check for buffer overflow, resize and retry
- if (*status == U_BUFFER_OVERFLOW_ERROR
- //|| dest8->utf8_len > dest8->utf8_cap
- ){
- icu_buf_utf8_resize(dest8, utf8_len * 2);
- *status = U_ZERO_ERROR;
u_strToUTF8((char *) dest8->utf8, dest8->utf8_cap,
&utf8_len,
src16->utf16, src16->utf16_len, status);
-
- }
-
- //if (*status != U_BUFFER_OVERFLOW_ERROR
- if (U_SUCCESS(*status)
- && utf8_len < dest8->utf8_cap)
- dest8->utf8_len = utf8_len;
- else {
- dest8->utf8[0] = (uint8_t) 0;
- dest8->utf8_len = 0;
- }
- return *status;
-};
-
-
+ // check for buffer overflow, resize and retry
+ if (*status == U_BUFFER_OVERFLOW_ERROR
+ //|| dest8->utf8_len > dest8->utf8_cap
+ ){
+ icu_buf_utf8_resize(dest8, utf8_len * 2);
+ *status = U_ZERO_ERROR;
+ u_strToUTF8((char *) dest8->utf8, dest8->utf8_cap,
+ &utf8_len,
+ src16->utf16, src16->utf16_len, status);
+ }
-UErrorCode icu_sortkey8_from_utf16(UCollator *coll,
- struct icu_buf_utf8 * dest8,
- struct icu_buf_utf16 * src16,
- UErrorCode * status)
-{
+ //if (*status != U_BUFFER_OVERFLOW_ERROR
+ if (U_SUCCESS(*status)
+ && utf8_len < dest8->utf8_cap)
+ dest8->utf8_len = utf8_len;
+ else {
+ dest8->utf8[0] = (uint8_t) 0;
+ dest8->utf8_len = 0;
+ }
- int32_t sortkey_len = 0;
-
- sortkey_len = ucol_getSortKey(coll, src16->utf16, src16->utf16_len,
- dest8->utf8, dest8->utf8_cap);
-
- // check for buffer overflow, resize and retry
- if (sortkey_len > dest8->utf8_cap) {
- icu_buf_utf8_resize(dest8, sortkey_len * 2);
- sortkey_len = ucol_getSortKey(coll, src16->utf16, src16->utf16_len,
- dest8->utf8, dest8->utf8_cap);
- }
-
- if (U_SUCCESS(*status)
- && sortkey_len > 0)
- dest8->utf8_len = sortkey_len;
- else {
- dest8->utf8[0] = (UChar) 0;
- dest8->utf8_len = 0;
- }
-
- return *status;
+ return *status;
};
-
-
-/// CRAP FOLLOWING HERE ...
-
-#if 0
-
-// forward declarations for helper functions
-
-int icu_check_status (UErrorCode status);
-
-UChar* icu_utf16_from_utf8(UChar *utf16,
- int32_t utf16_cap,
- int32_t *utf16_len,
- const char *utf8);
-
-UChar* icu_utf16_from_utf8n(UChar *utf16,
- int32_t utf16_cap,
- int32_t *utf16_len,
- const char *utf8,
- size_t utf8_len);
-
-
-char* icu_utf16_to_utf8(char *utf8,
- size_t utf8_cap,
- size_t *utf8_len,
- const UChar *utf16,
- int32_t utf16_len);
-
-
-int32_t icu_utf16_casemap(UChar *dest16, int32_t dest16_cap,
- const UChar *src16, int32_t src16_len,
- const char *locale, char action);
-
-
-// source code of all functions
-
-int icu_check_status (UErrorCode status)
+int icu_utf16_casemap(struct icu_buf_utf16 * dest16,
+ struct icu_buf_utf16 * src16,
+ const char *locale, char action,
+ UErrorCode *status)
{
- //if(U_FAILURE(status))
- if(!U_SUCCESS(status))
- yaz_log(YLOG_WARN,
- "ICU: %d %s\n", status, u_errorName(status));
- return status;
-}
-
-
-UChar* icu_utf16_from_utf8(UChar *utf16,
- int32_t utf16_cap,
- int32_t *utf16_len,
- const char *utf8)
-{
- size_t utf8_len = strlen(utf8);
- return icu_utf16_from_utf8n(utf16, utf16_cap, utf16_len,
- utf8, utf8_len);
-}
-
-
-UChar* icu_utf16_from_utf8n(UChar *utf16,
- int32_t utf16_cap,
- int32_t *utf16_len,
- const char *utf8,
- size_t utf8_len)
-{
- UErrorCode status = U_ZERO_ERROR;
- u_strFromUTF8(utf16, utf16_cap, utf16_len, utf8, (int32_t) utf8_len,
- &status);
- if (U_ZERO_ERROR != icu_check_status(status))
- return 0;
- else
- return utf16;
-}
-
-
-char* icu_utf16_to_utf8(char *utf8,
- size_t utf8_cap,
- size_t *utf8_len,
- const UChar *utf16,
- int32_t utf16_len)
-{
- UErrorCode status = U_ZERO_ERROR;
- u_strToUTF8(utf8, (int32_t) utf8_cap, (int32_t *)utf8_len,
- utf16, utf16_len, &status);
- if (U_ZERO_ERROR != icu_check_status(status))
- return 0;
- else
- return utf8;
-}
-
-
-int32_t icu_utf16_casemap(UChar *dest16, int32_t dest16_cap,
- const UChar *src16, int32_t src16_len,
- const char *locale, char action)
-{
- UErrorCode status = U_ZERO_ERROR;
int32_t dest16_len = 0;
switch(action) {
case 'l':
- dest16_len = u_strToLower(dest16, dest16_cap, src16, src16_len,
- locale, &status);
+ dest16_len = u_strToLower(dest16->utf16, dest16->utf16_cap,
+ src16->utf16, src16->utf16_len,
+ locale, status);
break;
case 'u':
- dest16_len = u_strToUpper(dest16, dest16_cap, src16, src16_len,
- locale, &status);
+ dest16_len = u_strToUpper(dest16->utf16, dest16->utf16_cap,
+ src16->utf16, src16->utf16_len,
+ locale, status);
break;
case 't':
- dest16_len = u_strToTitle(dest16, dest16_cap, src16, src16_len,
- 0, locale, &status);
+ dest16_len = u_strToTitle(dest16->utf16, dest16->utf16_cap,
+ src16->utf16, src16->utf16_len,
+ 0, locale, status);
break;
case 'f':
- dest16_len = u_strFoldCase(dest16, dest16_cap, src16, src16_len,
- U_FOLD_CASE_DEFAULT, &status);
+ dest16_len = u_strFoldCase(dest16->utf16, dest16->utf16_cap,
+ src16->utf16, src16->utf16_len,
+ U_FOLD_CASE_DEFAULT, status);
break;
default:
- return 0;
+ return U_UNSUPPORTED_ERROR;
break;
}
- if (U_ZERO_ERROR != icu_check_status(status))
- return 0;
- else
- return dest16_len;
-}
-
-
-char * icu_casemap(NMEM nmem, char *buf, size_t buf_cap,
- size_t *dest8_len, const char *src8,
- const char *locale, char action)
-{
- size_t src8_len = strlen(src8);
- int32_t buf_len = 0;
- char * dest8 = 0;
-
- if (dest8_len)
- *dest8_len = 0;
+ // check for buffer overflow, resize and retry
+ if (*status == U_BUFFER_OVERFLOW_ERROR
+ //|| dest16_len > dest16->utf16_cap
+ ){
+ icu_buf_utf16_resize(dest16, dest16_len * 2);
+ *status = U_ZERO_ERROR;
- if (!buf || !(buf_cap > 0) || !src8_len)
- return 0;
-
- // converting buf to utf16
- buf = (char *)icu_utf16_from_utf8n((UChar *) buf,
- (int32_t) buf_cap, &buf_len,
- src8, src8_len);
- // case mapping
- buf_len = (size_t) icu_utf16_casemap((UChar *)buf, (int32_t) buf_cap,
- (const UChar *)buf, (int32_t) buf_len,
- locale, action);
-
- // converting buf to utf8
- buf = icu_utf16_to_utf8(buf, buf_cap, (size_t *) &buf_len,
- (const UChar *) buf, (int32_t) buf_len);
-
+ switch(action) {
+ case 'l':
+ dest16_len = u_strToLower(dest16->utf16, dest16->utf16_cap,
+ src16->utf16, src16->utf16_len,
+ locale, status);
+ break;
+ case 'u':
+ dest16_len = u_strToUpper(dest16->utf16, dest16->utf16_cap,
+ src16->utf16, src16->utf16_len,
+ locale, status);
+ break;
+ case 't':
+ dest16_len = u_strToTitle(dest16->utf16, dest16->utf16_cap,
+ src16->utf16, src16->utf16_len,
+ 0, locale, status);
+ break;
+ case 'f':
+ dest16_len = u_strFoldCase(dest16->utf16, dest16->utf16_cap,
+ src16->utf16, src16->utf16_len,
+ U_FOLD_CASE_DEFAULT, status);
+ break;
+
+ default:
+ return U_UNSUPPORTED_ERROR;
+ break;
+ }
+ }
- // copying out to nmem
- buf[buf_len] = '\0';
-
- if(dest8_len)
- *dest8_len = buf_len;
-
- dest8 = nmem_strdup(nmem, buf);
- return dest8;
-}
-
-
-struct icu_termmap * icu_termmap_create(NMEM nmem)
-{
- struct icu_termmap *itmp = nmem_malloc(nmem, sizeof(*itmp));
- itmp->sort_key = 0;
- itmp->norm_term = 0;
- itmp->disp_term = 0;
- return itmp;
+ if (U_SUCCESS(*status)
+ && dest16_len < dest16->utf16_cap)
+ dest16->utf16_len = dest16_len;
+ else {
+ dest16->utf16[0] = (UChar) 0;
+ dest16->utf16_len = 0;
+ }
+
+ return *status;
};
-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;
-
- return strcmp(itmp1->sort_key, itmp2->sort_key);
-}
-
-char * icu_sortmap(NMEM nmem, char *buf, size_t buf_cap,
- size_t *dest8_len, const char *src8,
- const char *locale)
-{
- size_t src8_len = strlen(src8);
- int32_t buf_len = 0;
- char * dest8 = 0;
-
- if (dest8_len)
- *dest8_len = 0;
-
- if (!buf || !(buf_cap > 0) || !src8_len)
- return 0;
-
- // converting buf to utf16
- buf = (char *)icu_utf16_from_utf8n((UChar *) buf,
- (int32_t) buf_cap, &buf_len,
- src8, src8_len);
-
- // sort mapping
- //buf_len = (size_t) icu_utf16_casemap((UChar *)buf, (int32_t) buf_cap,
- // (const UChar *)buf, (int32_t) buf_len,
- // locale, action);
-
-
- {
- UErrorCode status = U_ZERO_ERROR;
+UErrorCode icu_sortkey8_from_utf16(UCollator *coll,
+ struct icu_buf_utf8 * dest8,
+ struct icu_buf_utf16 * src16,
+ UErrorCode * status)
+{
+
+ int32_t sortkey_len = 0;
- UCollator * coll = ucol_open (locale, &status);
- if (U_ZERO_ERROR != icu_check_status(status))
- buf_len = 0;
+ sortkey_len = ucol_getSortKey(coll, src16->utf16, src16->utf16_len,
+ dest8->utf8, dest8->utf8_cap);
- ucol_getSortKey(coll, (const UChar *) buf, (int32_t) buf_len,
- (uint8_t *) buf, (int32_t) buf_cap);
-
- ucol_close(coll);
+ // check for buffer overflow, resize and retry
+ if (sortkey_len > dest8->utf8_cap) {
+ icu_buf_utf8_resize(dest8, sortkey_len * 2);
+ sortkey_len = ucol_getSortKey(coll, src16->utf16, src16->utf16_len,
+ dest8->utf8, dest8->utf8_cap);
}
-
- // copying out to nmem
- buf[buf_len] = '\0';
-
- if(dest8_len)
- *dest8_len = buf_len;
+ if (U_SUCCESS(*status)
+ && sortkey_len > 0)
+ dest8->utf8_len = sortkey_len;
+ else {
+ dest8->utf8[0] = (UChar) 0;
+ dest8->utf8_len = 0;
+ }
- dest8 = nmem_strdup(nmem, buf);
- return dest8;
-}
+ return *status;
+};
-#endif
-/* $Id: icu_I18N.h,v 1.6 2007-05-07 12:18:34 marc Exp $
+/* $Id: icu_I18N.h,v 1.7 2007-05-07 12:52:04 marc 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.
+*/
#ifndef ICU_I18NL_H
#define ICU_I18NL_H
struct icu_buf_utf16 * src16,
UErrorCode * status);
+int icu_utf16_casemap(struct icu_buf_utf16 * dest16,
+ struct icu_buf_utf16 * src16,
+ const char *locale, char action,
+ UErrorCode *status);
UErrorCode icu_sortkey8_from_utf16(UCollator *coll,
struct icu_buf_utf8 * dest8,
UErrorCode * status);
-
-
-
-
-
-
-
-// CRAP to Follow here ...
-
-#if 0
-struct icu_termmap
-{
- char * sort_key; // standard C string '\0' terminated
- char * norm_term; // standard C utf-8 string
- char * disp_term; // standard C utf-8 string
-};
-
-struct icu_termmap * icu_termmap_create(NMEM nmem);
-
-int icu_termmap_cmp(const void *vp1, const void *vp2);
-
-char * icu_casemap(NMEM nmem, char *buf, size_t buf_cap,
- size_t *dest8_len, const char *src8,
- const char *locale, char action);
-
-char * icu_sortmap(NMEM nmem, char *buf, size_t buf_cap,
- size_t *dest8_len, const char *src8,
- const char *locale);
-
-#endif // 0
-
-
#endif // HAVE_ICU
#endif // ICU_I18NL_H
-/* $Id: test_icu_I18N.c,v 1.9 2007-05-07 12:18:34 marc Exp $
+/* $Id: test_icu_I18N.c,v 1.10 2007-05-07 12:52:04 marc 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
#include <string.h>
#include <stdlib.h>
-#include <unicode/ustring.h>
+//#include <unicode/ustring.h>
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
#define MAX_KEY_SIZE 256
-
struct icu_termmap
{
uint8_t sort_key[MAX_KEY_SIZE]; // standard C string '\0' terminated
-int icu_utf16_casemap(struct icu_buf_utf16 * dest16,
- struct icu_buf_utf16 * src16,
- const char *locale, char action,
- UErrorCode *status)
-{
- int32_t dest16_len = 0;
-
- switch(action) {
- case 'l':
- dest16_len = u_strToLower(dest16->utf16, dest16->utf16_cap,
- src16->utf16, src16->utf16_len,
- locale, status);
- break;
- case 'u':
- dest16_len = u_strToUpper(dest16->utf16, dest16->utf16_cap,
- src16->utf16, src16->utf16_len,
- locale, status);
- break;
- case 't':
- dest16_len = u_strToTitle(dest16->utf16, dest16->utf16_cap,
- src16->utf16, src16->utf16_len,
- 0, locale, status);
- break;
- case 'f':
- dest16_len = u_strFoldCase(dest16->utf16, dest16->utf16_cap,
- src16->utf16, src16->utf16_len,
- U_FOLD_CASE_DEFAULT, status);
- break;
-
- default:
- return U_UNSUPPORTED_ERROR;
- break;
- }
-
- // check for buffer overflow, resize and retry
- if (*status == U_BUFFER_OVERFLOW_ERROR
- //|| dest16_len > dest16->utf16_cap
- ){
- icu_buf_utf16_resize(dest16, dest16_len * 2);
- *status = U_ZERO_ERROR;
-
-
- switch(action) {
- case 'l':
- dest16_len = u_strToLower(dest16->utf16, dest16->utf16_cap,
- src16->utf16, src16->utf16_len,
- locale, status);
- break;
- case 'u':
- dest16_len = u_strToUpper(dest16->utf16, dest16->utf16_cap,
- src16->utf16, src16->utf16_len,
- locale, status);
- break;
- case 't':
- dest16_len = u_strToTitle(dest16->utf16, dest16->utf16_cap,
- src16->utf16, src16->utf16_len,
- 0, locale, status);
- break;
- case 'f':
- dest16_len = u_strFoldCase(dest16->utf16, dest16->utf16_cap,
- src16->utf16, src16->utf16_len,
- U_FOLD_CASE_DEFAULT, status);
- break;
-
- default:
- return U_UNSUPPORTED_ERROR;
- break;
- }
- }
-
- if (U_SUCCESS(*status)
- && dest16_len < dest16->utf16_cap)
- dest16->utf16_len = dest16_len;
- else {
- dest16->utf16[0] = (UChar) 0;
- dest16->utf16_len = 0;
- }
-
- return *status;
-};
-
-
int test_icu_casemap(const char * locale, char action,
const char * src8cstr, const char * chk8cstr)
-#if 0
-
-int test_icu_casemap(const char * locale, char action,
- const char * src8, const char * check8)
-{
- 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;
-
- //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);
-
-
- //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;
-}
-
-#endif
-
-
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
void test_icu_I18N_casemap(int argc, char **argv)
// 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"));
}
-#if 0
-
// 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;
-
- 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);
+ UErrorCode status = U_ZERO_ERROR;
- // 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);
+ struct icu_buf_utf8 * buf8 = icu_buf_utf8_create(0);
+ struct icu_buf_utf16 * buf16 = icu_buf_utf16_create(0);
- // 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);
+ int i;
- nmem_destroy(nmem);
-}
+ struct icu_termmap * list[src_list_len];
+ UCollator *coll = ucol_open(locale, &status);
+ icu_check_status(status);
+ if(!U_SUCCESS(status))
+ return 0;
-#endif
+ // assigning display terms and sort keys using buf 8 and buf16
+ for( i = 0; i < src_list_len; i++)
+ {
-// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
+ list[i] = (struct icu_termmap *) malloc(sizeof(struct icu_termmap));
-#if 0
-int test_icu_sortmap(const char * locale, size_t list_len,
- const char ** src8_list, const char ** check8_list)
-{
- int sucess = 1;
+ // copy display term
+ strcpy(list[i]->disp_term, src_list[i]);
- size_t i = 0;
+ // 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);
+ }
- 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");
- }
-
- nmem_destroy(nmem);
-
- return sucess;
-}
-
-#else
-
-int test_icu_sortmap(const char * locale, int src_list_len,
- const char ** src_list, const char ** chk_list)
-{
- int success = 1;
-
- UErrorCode status = U_ZERO_ERROR;
-
- struct icu_buf_utf8 * buf8 = icu_buf_utf8_create(0);
- struct icu_buf_utf16 * buf16 = icu_buf_utf16_create(0);
-
- int i;
-
- struct icu_termmap * list[src_list_len];
-
- UCollator *coll = ucol_open(locale, &status);
- icu_check_status(status);
-
- if(!U_SUCCESS(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);
- }
-
-
- // do the sorting
- qsort(list, src_list_len,
- sizeof(struct icu_termmap *), icu_termmap_cmp);
-
- // checking correct sorting
- for (i = 0; i < src_list_len; i++){
- if (0 != strcmp(list[i]->disp_term, chk_list[i])){
- success = 0;
+ 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");
}
- }
-
- 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("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");
- }
- ucol_close(coll);
+ ucol_close(coll);
- icu_buf_utf8_destroy(buf8);
- icu_buf_utf16_destroy(buf16);
+ icu_buf_utf8_destroy(buf8);
+ icu_buf_utf16_destroy(buf16);
}
-#endif
-
-
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
void test_icu_I18N_sortmap(int argc, char **argv)
}
-#endif
+#endif // HAVE_ICU
// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
test_icu_I18N_casemap(argc, argv);
test_icu_I18N_sortmap(argc, argv);
-#else
+#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
+
+
+// CRAP to follow
+#if 0
+
+// DO NOT EDIT THIS FILE IF YOUR EDITOR DOES NOT SUPPORT UTF-8
+
+void test_icu_I18N_casemap_failures(int argc, char **argv)
+{
+
+ size_t buf_cap = 128;
+ char buf[buf_cap];
+ size_t dest8_len = 0;
+ NMEM nmem = nmem_create();
+ char * dest8 = 0;
+
+ 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);
+}
+
+
+
+#endif
+
+
+
/*
* Local variables:
* c-basic-offset: 4
projects / pazpar2-moved-to-github.git / commitdiff