/* This file is part of the YAZ toolkit.
- * Copyright (C) 1995-2013 Index Data
+ * Copyright (C) Index Data
* See the file LICENSE for details.
*/
if (!rule && strcmp((const char *) node->name, "display"))
{
- yaz_log(YLOG_WARN, "Missing attribute rule for element %s",
+ yaz_log(YLOG_WARN, "Missing attribute 'rule' for element %s",
(const char *) node->name);
no_errors++;
continue;
}
if (step && U_FAILURE(*status))
{
+ yaz_log(YLOG_WARN, "ICU Error %d %s for element %s, rule %s",
+ *status, u_errorName(*status), node->name, rule ?
+ rule : "");
no_errors++;
break;
}
struct icu_iter {
struct icu_chain *chain;
struct icu_buf_utf16 *last;
+ struct icu_buf_utf16 *org;
UErrorCode status;
struct icu_buf_utf8 *display;
struct icu_buf_utf8 *sort8;
struct icu_buf_utf8 *result;
- struct icu_buf_utf16 *input;
int token_count;
+ size_t org_start;
+ size_t org_len;
struct icu_chain_step *steps;
};
}
dst = icu_buf_utf16_create(0);
iter->status = U_ZERO_ERROR;
- if (!icu_tokenizer_next_token(step->u.tokenizer, dst, &iter->status))
+ if (!icu_tokenizer_next_token(step->u.tokenizer, dst, &iter->status,
+ &iter->org_start, &iter->org_len))
{
icu_buf_utf16_destroy(dst);
dst = 0;
iter->display = icu_buf_utf8_create(0);
iter->sort8 = icu_buf_utf8_create(0);
iter->result = icu_buf_utf8_create(0);
+ iter->org = icu_buf_utf16_create(0);
iter->last = 0; /* no last returned string (yet) */
iter->steps = icu_chain_step_clone(chain->csteps);
- iter->input = 0;
+ iter->token_count = 0;
return iter;
}
void icu_iter_first(yaz_icu_iter_t iter, const char *src8cstr)
{
- if (iter->input)
- icu_buf_utf16_destroy(iter->input);
- iter->input = icu_buf_utf16_create(0);
+ struct icu_buf_utf16 *src = icu_buf_utf16_create(0);
+ icu_utf16_from_utf8_cstr(src, src8cstr, &iter->status);
+ icu_buf_utf16_copy(iter->org, src);
iter->token_count = 0;
- /* fill and assign input string.. It will be 0 after
- first iteration */
- icu_utf16_from_utf8_cstr(iter->input, src8cstr, &iter->status);
+ iter->org_start = 0;
+ iter->org_len = src->utf16_len;
+ iter->last = icu_iter_invoke(iter, iter->steps, src);
}
void icu_iter_destroy(yaz_icu_iter_t iter)
icu_buf_utf8_destroy(iter->display);
icu_buf_utf8_destroy(iter->sort8);
icu_buf_utf8_destroy(iter->result);
- if (iter->input)
- icu_buf_utf16_destroy(iter->input);
+ icu_buf_utf16_destroy(iter->org);
icu_chain_step_destroy(iter->steps);
xfree(iter);
}
int icu_iter_next(yaz_icu_iter_t iter)
{
- if (!iter->input && iter->last == 0)
+ if (iter->token_count && iter->last)
+ iter->last = icu_iter_invoke(iter, iter->steps, 0);
+ if (!iter->last)
return 0;
else
{
- /* on first call, iter->input is the input string. Thereafter: 0. */
- assert(iter->steps || !iter->chain->csteps);
- iter->last = icu_iter_invoke(iter, iter->steps, iter->input);
- iter->input = 0;
-
- if (!iter->last)
- return 0;
-
iter->token_count++;
-
if (iter->chain->sort)
{
icu_sortkey8_from_utf16(iter->chain->coll,
return iter->token_count;
}
+
+void icu_iter_get_org_info(yaz_icu_iter_t iter, size_t *start, size_t *len)
+{
+ /* save full length of org since we're gonna cut it */
+ int32_t save_len = iter->org->utf16_len;
+
+ struct icu_buf_utf8 *tmp = icu_buf_utf8_create(0);
+ UErrorCode status = U_ZERO_ERROR;
+
+ iter->org->utf16_len = iter->org_start;
+ icu_utf16_to_utf8(tmp, iter->org, &status);
+ if (U_SUCCESS(status))
+ *start = tmp->utf8_len;
+ else
+ *start = 0;
+ iter->org->utf16_len = iter->org_start + iter->org_len;
+ icu_utf16_to_utf8(tmp, iter->org, &status);
+ if (U_SUCCESS(status))
+ *len = tmp->utf8_len - *start;
+ else
+ *len = 0;
+ icu_buf_utf8_destroy(tmp);
+ iter->org->utf16_len = save_len;
+}
+
int icu_chain_assign_cstr(struct icu_chain *chain, const char *src8cstr,
UErrorCode *status)
{
return 0;
}
+void icu_chain_get_org_info(struct icu_chain *chain, size_t *start, size_t *len)
+{
+ if (chain->iter)
+ icu_iter_get_org_info(chain->iter, start, len);
+}
+
+
#endif /* YAZ_HAVE_ICU */
/*