/*
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
*/
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)
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)
struct icu_buf_utf8 * src8,
UErrorCode * status)
{
- //if(!U_SUCCESS(*status))
- // return *status;
- printf("icu_utf16_from_utf8 working\n");
+ 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);
size_t src8cstr_len = 0;
int32_t utf16_len = 0;
- //if(!U_SUCCESS(status))
- // return *status;
-
- printf("icu_utf16_from_utf8_cstr working\n");
src8cstr_len = strlen(src8cstr);
u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
if (*status == U_BUFFER_OVERFLOW_ERROR
//|| dest16->utf16_len > dest16->utf16_cap
){
- printf("icu_utf16_from_utf8_cstr need resize\n");
icu_buf_utf16_resize(dest16, utf16_len * 2);
*status = U_ZERO_ERROR;
u_strFromUTF8(dest16->utf16, dest16->utf16_cap,
{
int32_t sortkey_len = 0;
- //if(!U_SUCCESS(status))
- // return *status;
- printf("icu_sortkey8_from_utf16 working\n");
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) {
- printf("icu_sortkey8_from_utf16 need resize\n");
icu_buf_utf8_resize(dest8, sortkey_len * 2);
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");
const char * en_1_src[6] = {"z", "K", "a", "A", "Z", "k"};
const char * en_1_cck[6] = {"a", "A", "K", "k", "z", "Z"};
icu_coll_sort("en", en_1_len, en_1_src, en_1_cck);
-
-#if 0
icu_coll_sort("en_AU", en_1_len, en_1_src, en_1_cck);
icu_coll_sort("en_CA", en_1_len, en_1_src, en_1_cck);
icu_coll_sort("en_GB", en_1_len, en_1_src, en_1_cck);
icu_coll_sort("de", de_1_len, de_1_src, de_1_cck);
icu_coll_sort("de_AT", de_1_len, de_1_src, de_1_cck);
icu_coll_sort("de_DE", de_1_len, de_1_src, de_1_cck);
-#endif
+
return 0;
};