-/* $Id: rpnscan.c,v 1.19 2007-11-01 14:56:07 adam Exp $
+/* $Id: rpnscan.c,v 1.25 2008-01-24 16:17:03 adam Exp $
Copyright (C) 1995-2007
Index Data ApS
#define RPN_MAX_ORDS 32
-static int log_scan = YLOG_LOG;
-
/* convert APT SCAN term to internal cmap */
static ZEBRA_RES trans_scan_term(ZebraHandle zh, Z_AttributesPlusTerm *zapt,
char *termz, zebra_map_t zm)
{
- char termz0[IT_MAX_WORD];
+ char term_utf8[IT_MAX_WORD];
- if (zapt_term_to_utf8(zh, zapt, termz0) == ZEBRA_FAIL)
+ if (zapt_term_to_utf8(zh, zapt, term_utf8) == ZEBRA_FAIL)
return ZEBRA_FAIL; /* error */
+ else if (zebra_maps_is_icu(zm))
+ {
+ const char *res_buf;
+ size_t res_len;
+ zebra_map_tokenize_start(zm, term_utf8, strlen(term_utf8));
+
+ if (zebra_map_tokenize_next(zm, &res_buf, &res_len, 0, 0))
+ {
+ memcpy(termz, res_buf, res_len);
+ termz[res_len] = '\0';
+ }
+ else
+ termz[0] = '\0';
+ }
else
{
const char **map;
- const char *cp = (const char *) termz0;
+ const char *cp = (const char *) term_utf8;
const char *cp_end = cp + strlen(cp);
const char *src;
int i = 0;
return ZEBRA_OK;
}
-static void count_set(ZebraHandle zh, RSET rset, zint *count, zint approx_limit)
-{
- zint psysno = 0;
- struct it_key key;
- RSFD rfd;
-
- yaz_log(YLOG_DEBUG, "count_set");
-
- rset->hits_limit = approx_limit;
-
- *count = 0;
- rfd = rset_open(rset, RSETF_READ);
- while (rset_read(rfd, &key,0 /* never mind terms */))
- {
- if (key.mem[0] != psysno)
- {
- psysno = key.mem[0];
- if (rfd->counted_items >= rset->hits_limit)
- break;
- }
- }
- rset_close(rfd);
- *count = rset->hits_count;
-}
-
static void get_first_snippet_from_rset(ZebraHandle zh,
RSET rset, zebra_snippets *snippets,
zint *sysno)
rset = rset_create_and(nmem, kc, kc->scope, 2, rsets);
}
/* count it */
- count_set(zh, rset, &count, approx_limit);
+ zebra_count_set(zh, rset, &count, approx_limit);
if (pos != -1)
{
zint sysno;
- int code = -1;
- zebra_snippets *rec_snippets = zebra_snippets_create();
zebra_snippets *hit_snippets = zebra_snippets_create();
glist[pos].term = 0;
get_first_snippet_from_rset(zh, rset, hit_snippets, &sysno);
if (sysno)
- code = zebra_get_rec_snippets(zh, sysno, rec_snippets);
-
- if (code == 0)
{
- const struct zebra_snippet_word *w =
- zebra_snippets_lookup(rec_snippets, hit_snippets);
- if (w)
+ zebra_snippets *rec_snippets = zebra_snippets_create();
+ int code = zebra_get_rec_snippets(zh, sysno, rec_snippets);
+ if (code == 0)
{
- glist[pos].display_term = odr_strdup(stream, w->term);
+ const struct zebra_snippet_word *w =
+ zebra_snippets_lookup(rec_snippets, hit_snippets);
+ if (w)
+ {
+ glist[pos].display_term = odr_strdup(stream, w->term);
+ }
+ else
+ {
+ yaz_log(YLOG_WARN, "zebra_snippets_lookup failed for pos=%d", pos);
+ }
}
+ zebra_snippets_destroy(rec_snippets);
+ }
+ if (zebra_term_untrans_iconv(zh, stream->mem, index_type,
+ &glist[pos].term, term))
+ {
+ /* failed.. use display_term instead (which could be 0) */
+ glist[pos].term = glist[pos].display_term;
}
+
if (!glist[pos].term)
- zebra_term_untrans_iconv(zh, stream->mem, index_type,
- &glist[pos].term, term);
+ {
+ yaz_log(YLOG_WARN, "Could not generate scan term for pos=%d",
+ pos);
+ glist[pos].term = "None";
+ }
glist[pos].occurrences = count;
- zebra_snippets_destroy(rec_snippets);
zebra_snippets_destroy(hit_snippets);
}
rset_delete(rset);
{
struct scan2_info_entry *ar = nmem_malloc(nmem, sizeof(*ar) * ord_no);
struct rpn_char_map_info rcmi;
- zebra_map_t zm = zebra_map_get(zh->reg->zebra_maps, index_type);
+ zebra_map_t zm = zebra_map_get_or_add(zh->reg->zebra_maps, index_type);
int i, dif;
int after_pos;
int pos = 0;
const Odr_oid *attributeset,
int num_bases, char **basenames,
int *position, int *num_entries, ZebraScanEntry **list,
- int *is_partial, RSET limit_set)
+ int *is_partial, const char *set_name)
{
int base_no;
int ords[RPN_MAX_ORDS], ord_no = 0;
NMEM nmem;
ZEBRA_RES res;
struct rset_key_control *kc = 0;
+ RSET limit_set = 0;
*list = 0;
*is_partial = 0;
if (!attributeset)
attributeset = yaz_oid_attset_bib_1;
- if (!limit_set) /* no limit set given already */
+ if (!set_name)
{
/* see if there is a @attr 8=set */
AttrType termset;
int termset_value_numeric;
- const char *termset_value_string;
+ const char *termset_value_string = 0;
attr_init_APT(&termset, zapt, 8);
termset_value_numeric =
attr_find_ex(&termset, NULL, &termset_value_string);
if (termset_value_numeric != -1)
{
- char resname[32];
- const char *termset_name = 0;
-
if (termset_value_numeric != -2)
{
+ char resname[32];
sprintf(resname, "%d", termset_value_numeric);
- termset_name = resname;
+ set_name = odr_strdup(stream, resname);
}
else
- termset_name = termset_value_string;
-
- limit_set = resultSetRef(zh, termset_name);
-
- if (!limit_set)
- {
- zebra_setError(zh,
- YAZ_BIB1_SPECIFIED_RESULT_SET_DOES_NOT_EXIST,
- termset_name);
- return ZEBRA_FAIL;
- }
+ set_name = odr_strdup(stream, termset_value_string);
}
}
+
+ if (set_name)
+ {
+ limit_set = resultSetRef(zh, set_name);
+ if (!limit_set)
+ {
+ zebra_setError(zh,
+ YAZ_BIB1_SPECIFIED_RESULT_SET_DOES_NOT_EXIST,
+ set_name);
+ return ZEBRA_FAIL;
+ }
+ }
+
yaz_log(YLOG_DEBUG, "position = %d, num = %d",
*position, *num_entries);
zebra_setError(zh, YAZ_BIB1_TOO_MANY_DATABASES_SPECIFIED, 0);
return ZEBRA_FAIL;
}
-
for (base_no = 0; base_no < num_bases; base_no++)
{
int ord;
nmem = nmem_create();
kc = zebra_key_control_create(zh);
- if (sort_flag)
- res = rpn_facet(zh, stream, nmem, kc, zapt, position, num_entries,
- list,
- is_partial, limit_set, index_type, ord_no, ords);
- else
- res = rpn_scan_norm(zh, stream, nmem, kc, zapt, position, num_entries,
- list,
- is_partial, limit_set, index_type, ord_no, ords);
+ res = rpn_scan_norm(zh, stream, nmem, kc, zapt, position, num_entries,
+ list, is_partial, limit_set,
+ index_type, ord_no, ords);
nmem_destroy(nmem);
(*kc->dec)(kc);
return res;