+
+WRBUF zebra_replace(zebra_map_t zm, const char *ex_list,
+ const char *input_str, int input_len)
+{
+ wrbuf_rewind(zm->zebra_maps->wrbuf_1);
+ wrbuf_write(zm->zebra_maps->wrbuf_1, input_str, input_len);
+ return zm->zebra_maps->wrbuf_1;
+}
+
+#define SE_CHARS ";,.()-/?<> \r\n\t"
+
+static int tokenize_simple(zebra_map_t zm,
+ const char **result_buf, size_t *result_len)
+{
+ char *buf = wrbuf_buf(zm->input_str);
+ size_t len = wrbuf_len(zm->input_str);
+ size_t i = zm->simple_off;
+ size_t start;
+
+ while (i < len && strchr(SE_CHARS, buf[i]))
+ i++;
+ start = i;
+ while (i < len && !strchr(SE_CHARS, buf[i]))
+ {
+ if (buf[i] > 32 && buf[i] < 127)
+ buf[i] = tolower(buf[i]);
+ i++;
+ }
+
+ zm->simple_off = i;
+ if (start != i)
+ {
+ *result_buf = buf + start;
+ *result_len = i - start;
+ return 1;
+ }
+ return 0;
+ }
+
+
+int zebra_map_tokenize_next(zebra_map_t zm,
+ const char **result_buf, size_t *result_len,
+ const char **display_buf, size_t *display_len)
+{
+ assert(zm->use_chain);
+
+#if YAZ_HAVE_ICU
+ if (!zm->icu_chain)
+ return tokenize_simple(zm, result_buf, result_len);
+ else
+ {
+ UErrorCode status;
+ while (icu_chain_next_token(zm->icu_chain, &status))
+ {
+ if (!U_SUCCESS(status))
+ return 0;
+ *result_buf = icu_chain_token_sortkey(zm->icu_chain);
+ assert(*result_buf);
+
+ *result_len = strlen(*result_buf);
+
+ if (display_buf)
+ {
+ *display_buf = icu_chain_token_display(zm->icu_chain);
+ if (display_len)
+ *display_len = strlen(*display_buf);
+ }
+ if (zm->debug)
+ {
+ wrbuf_rewind(zm->print_str);
+ wrbuf_write_escaped(zm->print_str, *result_buf, *result_len);
+ yaz_log(YLOG_LOG, "output %s", wrbuf_cstr(zm->print_str));
+ }
+
+ if (**result_buf != '\0')
+ return 1;
+ }
+ }
+ return 0;
+#else
+ return tokenize_simple(zm, result_buf, result_len);
+#endif
+}
+
+int zebra_map_tokenize_start(zebra_map_t zm,
+ const char *buf, size_t len)
+{
+ assert(zm->use_chain);
+
+ wrbuf_rewind(zm->input_str);
+ wrbuf_write(zm->input_str, buf, len);
+ zm->simple_off = 0;
+#if YAZ_HAVE_ICU
+ if (zm->icu_chain)
+ {
+ UErrorCode status;
+ if (zm->debug)
+ {
+ wrbuf_rewind(zm->print_str);
+ wrbuf_write_escaped(zm->print_str, wrbuf_buf(zm->input_str),
+ wrbuf_len(zm->input_str));
+
+ yaz_log(YLOG_LOG, "input %s",
+ wrbuf_cstr(zm->print_str));
+ }
+ icu_chain_assign_cstr(zm->icu_chain,
+ wrbuf_cstr(zm->input_str),
+ &status);
+ if (!U_SUCCESS(status))
+ {
+ if (zm->debug)
+ {
+ yaz_log(YLOG_WARN, "bad encoding for input");
+ }
+ return -1;
+ }
+ }
+#endif
+ return 0;
+}
+
+int zebra_maps_is_icu(zebra_map_t zm)
+{
+ assert(zm);
+#if YAZ_HAVE_ICU
+ return zm->use_chain;
+#else
+ return 0;
+#endif
+}
+
+
+/*
+ * Local variables:
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ * vim: shiftwidth=4 tabstop=8 expandtab
+ */
+