// Make command on debian 64 bit testing dist
/*
-gcc -g -Wall `icu-config --cppflags` `icu-config --ldflags` -o icu_bug icu_bug.c
+gcc -g -Wall `icu-config --cppflags` `icu-config --ldflags` -o icu_bug_2 icu_bug_2.c
snatched from http://www.icu-project.org/userguide/Collate_API.html
-and corrected for compile errors
+and changed.
added a struct icu_termmap such that I actually can see the output
*/
#define MAX_KEY_SIZE 256
-//#define MAX_BUFFER_SIZE 10000
-//#define MAX_LIST_LENGTH 5
-
struct icu_buf_utf16
{
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)
{
- printf("buf16 resize: %d\n", (int)capacity);
if (buf16){
if (capacity > 0){
if (0 == buf16->utf16)
else
buf16->utf16
= (UChar *) realloc(buf16->utf16, sizeof(UChar) * capacity);
+ buf16->utf16[0] = (UChar) 0;
buf16->utf16_len = 0;
buf16->utf16_cap = capacity;
}
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)
{
- printf("buf8 resize: %d\n", (int)capacity);
if (buf8){
if (capacity > 0){
if (0 == buf8->utf8)
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;
}
struct icu_buf_utf8 * src8,
UErrorCode * status)
{
- if(!U_SUCCESS(*status))
- return *status;
+ printf("icu_utf16_from_utf8 - needs correcting, see icu_utf16_from_utf8_cstr\n");
u_strFromUTF8(dest16->utf16, dest16->utf16_cap, &(dest16->utf16_len),
(const char *) src8->utf8, src8->utf8_len, status);
// check for buffer overflow, resize and retry
if (dest16->utf16_len > dest16->utf16_cap){
+ printf("icu_utf16_from_utf8 need resize\n");
icu_buf_utf16_resize(dest16, dest16->utf16_len * 2);
*status = U_ZERO_ERROR;
u_strFromUTF8(dest16->utf16, dest16->utf16_cap, &(dest16->utf16_len),
UErrorCode * status)
{
size_t src8cstr_len = 0;
-
- if(!U_SUCCESS(status))
- return *status;
-
+ int32_t utf16_len = 0;
+
src8cstr_len = strlen(src8cstr);
- u_strFromUTF8(dest16->utf16, dest16->utf16_cap, &(dest16->utf16_len),
+ u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
+ &utf16_len,
+ //&(dest16->utf16_len),
src8cstr, src8cstr_len, status);
// check for buffer overflow, resize and retry
- if (dest16->utf16_len > dest16->utf16_cap){
- icu_buf_utf16_resize(dest16, dest16->utf16_len * 2);
+ 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, &(dest16->utf16_len),
+ u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
+ &utf16_len,
+ //&(dest16->utf16_len),
src8cstr, src8cstr_len, status);
}
+
+ if (*status != U_BUFFER_OVERFLOW_ERROR
+ && utf16_len < dest16->utf16_cap)
+ dest16->utf16_len = utf16_len;
+ else {
+ dest16->utf16[0] = (UChar) 0;
+ dest16->utf16_len = 0;
+ }
return *status;
};
{
int32_t sortkey_len = 0;
- if(!U_SUCCESS(status))
- return *status;
-
+
sortkey_len = ucol_getSortKey(coll, src16->utf16, src16->utf16_len,
- dest8->utf8, dest8->utf8_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);
- ucol_getSortKey(coll, src16->utf16, src16->utf16_len,
- dest8->utf8, dest8->utf8_len);
+ sortkey_len = ucol_getSortKey(coll, src16->utf16, src16->utf16_len,
+ dest8->utf8, dest8->utf8_cap);
}
+ if (sortkey_len > 0)
+ dest8->utf8_len = sortkey_len;
+
return *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);
}
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");