/*
- * Copyright (C) 1995-1999, Index Data
+ * Copyright (C) 1995-2000, Index Data
* All rights reserved.
* Sebastian Hammer, Adam Dickmeiss
*
* $Log: zrpn.c,v $
- * Revision 1.100 1999-12-28 15:48:12 adam
+ * Revision 1.104 2000-04-05 09:49:35 adam
+ * On Unix, zebra/z'mbol uses automake.
+ *
+ * Revision 1.103 2000/03/20 19:08:36 adam
+ * Added remote record import using Z39.50 extended services and Segment
+ * Requests.
+ *
+ * Revision 1.102 2000/03/15 15:00:31 adam
+ * First work on threaded version.
+ *
+ * Revision 1.101 2000/03/02 14:35:03 adam
+ * Fixed proximity handling.
+ *
+ * Revision 1.100 1999/12/28 15:48:12 adam
* Minor Fix.
*
* Revision 1.99 1999/12/23 09:03:32 adam
static void rpn_char_map_prepare (ZebraHandle zh, int reg_type,
struct rpn_char_map_info *map_info)
{
- map_info->zm = zh->zebra_maps;
+ map_info->zm = zh->service->zebra_maps;
map_info->reg_type = reg_type;
- dict_grep_cmap (zh->dict, map_info, rpn_char_map_handler);
+ dict_grep_cmap (zh->service->dict, map_info, rpn_char_map_handler);
}
typedef struct {
{
while (*src)
{
- const char *cp = zebra_maps_output (zh->zebra_maps, reg_type, &src);
+ const char *cp = zebra_maps_output (zh->service->zebra_maps,
+ reg_type, &src);
if (!cp)
*dst++ = *src++;
else
switch (relation_value)
{
case 1:
- if (!term_100 (zh->zebra_maps, reg_type, term_sub, term_component,
+ if (!term_100 (zh->service->zebra_maps, reg_type,
+ term_sub, term_component,
space_split, term_dst))
return 0;
logf (LOG_DEBUG, "Relation <");
*term_tmp = '\0';
break;
case 2:
- if (!term_100 (zh->zebra_maps, reg_type, term_sub, term_component,
+ if (!term_100 (zh->service->zebra_maps, reg_type,
+ term_sub, term_component,
space_split, term_dst))
return 0;
logf (LOG_DEBUG, "Relation <=");
*term_tmp = '\0';
break;
case 5:
- if (!term_100 (zh->zebra_maps, reg_type, term_sub, term_component,
- space_split, term_dst))
+ if (!term_100 (zh->service->zebra_maps, reg_type,
+ term_sub, term_component, space_split, term_dst))
return 0;
logf (LOG_DEBUG, "Relation >");
*term_tmp = '\0';
break;
case 4:
- if (!term_100 (zh->zebra_maps, reg_type, term_sub, term_component,
- space_split, term_dst))
+ if (!term_100 (zh->service->zebra_maps, reg_type, term_sub,
+ term_component, space_split, term_dst))
return 0;
logf (LOG_DEBUG, "Relation >=");
case 3:
default:
logf (LOG_DEBUG, "Relation =");
- if (!term_100 (zh->zebra_maps, reg_type, term_sub, term_component,
- space_split, term_dst))
+ if (!term_100 (zh->service->zebra_maps, reg_type, term_sub,
+ term_component, space_split, term_dst))
return 0;
strcat (term_tmp, "(");
strcat (term_tmp, term_component);
}
return -1;
}
- if (zebraExplain_curDatabase (zh->zei, basenames[base_no]))
+ if (zebraExplain_curDatabase (zh->service->zei, basenames[base_no]))
{
zh->errCode = 109; /* Database unavailable */
zh->errString = basenames[base_no];
char ord_buf[32];
int i, ord_len;
- ord = zebraExplain_lookupSU (zh->zei, attp.attset_ordinal,
+ ord = zebraExplain_lookupSU (zh->service->zei, attp.attset_ordinal,
local_attr->local);
if (ord < 0)
continue;
reg_type, space_split, term_dst))
return 0;
logf (LOG_DEBUG, "dict_lookup_grep: %s", term_dict+prefix_len);
- r = dict_lookup_grep (zh->dict, term_dict, 0, grep_info, &max_pos,
- 0, grep_handle);
+ r = dict_lookup_grep (zh->service->dict, term_dict, 0,
+ grep_info, &max_pos, 0, grep_handle);
if (r)
logf (LOG_WARN, "dict_lookup_grep fail, rel=gt: %d", r);
break;
case 1: /* right truncation */
term_dict[j++] = '(';
- if (!term_100 (zh->zebra_maps, reg_type,
+ if (!term_100 (zh->service->zebra_maps, reg_type,
&termp, term_dict + j, space_split, term_dst))
return 0;
strcat (term_dict, ".*)");
- dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ dict_lookup_grep (zh->service->dict, term_dict, 0, grep_info,
&max_pos, 0, grep_handle);
break;
case 2: /* keft truncation */
term_dict[j++] = '('; term_dict[j++] = '.'; term_dict[j++] = '*';
- if (!term_100 (zh->zebra_maps, reg_type,
+ if (!term_100 (zh->service->zebra_maps, reg_type,
&termp, term_dict + j, space_split, term_dst))
return 0;
strcat (term_dict, ")");
- dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ dict_lookup_grep (zh->service->dict, term_dict, 0, grep_info,
&max_pos, 0, grep_handle);
break;
case 3: /* left&right truncation */
term_dict[j++] = '('; term_dict[j++] = '.'; term_dict[j++] = '*';
- if (!term_100 (zh->zebra_maps, reg_type,
+ if (!term_100 (zh->service->zebra_maps, reg_type,
&termp, term_dict + j, space_split, term_dst))
return 0;
strcat (term_dict, ".*)");
- dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ dict_lookup_grep (zh->service->dict, term_dict, 0, grep_info,
&max_pos, 0, grep_handle);
break;
zh->errCode = 120;
return -1;
case 101: /* process # in term */
term_dict[j++] = '(';
- if (!term_101 (zh->zebra_maps, reg_type,
+ if (!term_101 (zh->service->zebra_maps, reg_type,
&termp, term_dict + j, space_split, term_dst))
return 0;
strcat (term_dict, ")");
- r = dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ r = dict_lookup_grep (zh->service->dict, term_dict, 0, grep_info,
&max_pos, 0, grep_handle);
if (r)
logf (LOG_WARN, "dict_lookup_grep err, trunc=#: %d", r);
break;
case 102: /* Regexp-1 */
term_dict[j++] = '(';
- if (!term_102 (zh->zebra_maps, reg_type,
+ if (!term_102 (zh->service->zebra_maps, reg_type,
&termp, term_dict + j, space_split, term_dst))
return 0;
strcat (term_dict, ")");
logf (LOG_DEBUG, "Regexp-1 tolerance=%d", r);
- r = dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ r = dict_lookup_grep (zh->service->dict, term_dict, 0, grep_info,
&max_pos, 0, grep_handle);
if (r)
logf (LOG_WARN, "dict_lookup_grep err, trunc=regular: %d",
case 103: /* Regexp-2 */
r = 1;
term_dict[j++] = '(';
- if (!term_103 (zh->zebra_maps, reg_type,
+ if (!term_103 (zh->service->zebra_maps, reg_type,
&termp, term_dict + j, &r, space_split, term_dst))
return 0;
strcat (term_dict, ")");
logf (LOG_DEBUG, "Regexp-2 tolerance=%d", r);
- r = dict_lookup_grep (zh->dict, term_dict, r, grep_info,
+ r = dict_lookup_grep (zh->service->dict, term_dict, r, grep_info,
&max_pos, 2, grep_handle);
if (r)
logf (LOG_WARN, "dict_lookup_grep err, trunc=eregular: %d",
break;
case 104: /* process # and ! in term */
term_dict[j++] = '(';
- if (!term_104 (zh->zebra_maps, reg_type,
+ if (!term_104 (zh->service->zebra_maps, reg_type,
&termp, term_dict + j, space_split, term_dst))
return 0;
strcat (term_dict, ")");
- r = dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ r = dict_lookup_grep (zh->service->dict, term_dict, 0, grep_info,
&max_pos, 0, grep_handle);
if (r)
logf (LOG_WARN, "dict_lookup_grep err, trunc=#/!: %d", r);
break;
case 105: /* process * and ! in term */
term_dict[j++] = '(';
- if (!term_105 (zh->zebra_maps, reg_type,
+ if (!term_105 (zh->service->zebra_maps, reg_type,
&termp, term_dict + j, space_split, term_dst, 1))
return 0;
strcat (term_dict, ")");
- r = dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ r = dict_lookup_grep (zh->service->dict, term_dict, 0, grep_info,
&max_pos, 0, grep_handle);
if (r)
logf (LOG_WARN, "dict_lookup_grep err, trunc=*/!: %d", r);
break;
case 106: /* process * and ! in term */
term_dict[j++] = '(';
- if (!term_105 (zh->zebra_maps, reg_type,
+ if (!term_105 (zh->service->zebra_maps, reg_type,
&termp, term_dict + j, space_split, term_dst, 0))
return 0;
strcat (term_dict, ")");
- r = dict_lookup_grep (zh->dict, term_dict, 0, grep_info,
+ r = dict_lookup_grep (zh->service->dict, term_dict, 0, grep_info,
&max_pos, 0, grep_handle);
if (r)
logf (LOG_WARN, "dict_lookup_grep err, trunc=*/!: %d", r);
while ((len = (cp_end - cp)) > 0)
{
- map = zebra_maps_input (zh->zebra_maps, reg_type, &cp, len);
+ map = zebra_maps_input (zh->service->zebra_maps, reg_type, &cp, len);
if (**map == *CHR_SPACE)
space_map = *map;
else
termz[i] = '\0';
}
-static RSET rpn_proximity (ZebraHandle zh, RSET rset1, RSET rset2,
- int ordered,
- int exclusion, int relation, int distance)
-{
- int i;
- RSFD rsfd1, rsfd2;
- int more1, more2;
- struct it_key buf1, buf2;
- RSFD rsfd_result;
- RSET result;
- rset_temp_parms parms;
- int term_index;
-
- rsfd1 = rset_open (rset1, RSETF_READ);
- more1 = rset_read (rset1, rsfd1, &buf1, &term_index);
-
- rsfd2 = rset_open (rset2, RSETF_READ);
- more2 = rset_read (rset2, rsfd2, &buf2, &term_index);
-
- parms.key_size = sizeof (struct it_key);
- parms.temp_path = res_get (zh->res, "setTmpDir");
- result = rset_create (rset_kind_temp, &parms);
- rsfd_result = rset_open (result, RSETF_WRITE);
-
- logf (LOG_DEBUG, "rpn_proximity excl=%d ord=%d rel=%d dis=%d",
- exclusion, ordered, relation, distance);
- while (more1 && more2)
- {
- int cmp = key_compare_it (&buf1, &buf2);
- if (cmp < -1)
- more1 = rset_read (rset1, rsfd1, &buf1, &term_index);
- else if (cmp > 1)
- more2 = rset_read (rset2, rsfd2, &buf2, &term_index);
- else
- {
- int sysno = buf1.sysno;
- int seqno[500];
- int n = 0;
-
- seqno[n++] = buf1.seqno;
- while ((more1 = rset_read (rset1, rsfd1, &buf1, &term_index)) &&
- sysno == buf1.sysno)
- if (n < 500)
- seqno[n++] = buf1.seqno;
- do
- {
- for (i = 0; i<n; i++)
- {
- int diff = buf2.seqno - seqno[i];
- int excl = exclusion;
- if (!ordered && diff < 0)
- diff = -diff;
- switch (relation)
- {
- case 1: /* < */
- if (diff < distance)
- excl = !excl;
- break;
- case 2: /* <= */
- if (diff <= distance)
- excl = !excl;
- break;
- case 3: /* == */
- if (diff == distance)
- excl = !excl;
- break;
- case 4: /* >= */
- if (diff >= distance)
- excl = !excl;
- break;
- case 5: /* > */
- if (diff > distance)
- excl = !excl;
- break;
- case 6: /* != */
- if (diff != distance)
- excl = !excl;
- break;
- }
- if (excl)
- rset_write (result, rsfd_result, &buf2);
- }
- } while ((more2 = rset_read (rset2, rsfd2, &buf2, &term_index)) &&
- sysno == buf2.sysno);
- }
- }
- rset_close (result, rsfd_result);
- rset_close (rset1, rsfd1);
- rset_close (rset2, rsfd2);
- return result;
-}
-
-static RSET rpn_prox (ZebraHandle zh, RSET *rset, int rset_no)
+static RSET rpn_prox (ZebraHandle zh, RSET *rset, int rset_no,
+ int ordered, int exclusion, int relation, int distance)
{
int i;
RSFD *rsfd;
}
for (i = 0; i<rset_no; i++)
{
+ buf[i] = 0;
+ rsfd[i] = 0;
+ }
+ for (i = 0; i<rset_no; i++)
+ {
buf[i] = (struct it_key *) xmalloc (sizeof(**buf));
rsfd[i] = rset_open (rset[i], RSETF_READ);
if (!(more[i] = rset_read (rset[i], rsfd[i], buf[i], &term_index)))
}
if (i != rset_no)
{
+ /* at least one is empty ... return null set */
rset_null_parms parms;
-
- while (i >= 0)
- {
- rset_close (rset[i], rsfd[i]);
- xfree (buf[i]);
- --i;
- }
+
parms.rset_term = rset_term_create (prox_term, length_prox_term,
flags);
parms.rset_term->nn = 0;
result = rset_create (rset_kind_null, &parms);
}
- else
+ else if (ordered && relation == 3 && exclusion == 0 && distance == 1)
{
+ /* special proximity case = phrase search ... */
rset_temp_parms parms;
RSFD rsfd_result;
flags);
parms.rset_term->nn = min_nn;
parms.key_size = sizeof (struct it_key);
- parms.temp_path = res_get (zh->res, "setTmpDir");
+ parms.temp_path = res_get (zh->service->res, "setTmpDir");
result = rset_create (rset_kind_temp, &parms);
rsfd_result = rset_open (result, RSETF_WRITE);
more[0] = rset_read (*rset, *rsfd, *buf, &term_index);
}
}
-
- for (i = 0; i<rset_no; i++)
+ rset_close (result, rsfd_result);
+ }
+ else if (rset_no == 2)
+ {
+ /* generic proximity case (two input sets only) ... */
+ rset_temp_parms parms;
+ RSFD rsfd_result;
+
+ logf (LOG_LOG, "generic prox, dist = %d, relation = %d, ordered =%d, exclusion=%d",
+ distance, relation, ordered, exclusion);
+ parms.rset_term = rset_term_create (prox_term, length_prox_term,
+ flags);
+ parms.rset_term->nn = min_nn;
+ parms.key_size = sizeof (struct it_key);
+ parms.temp_path = res_get (zh->service->res, "setTmpDir");
+ result = rset_create (rset_kind_temp, &parms);
+ rsfd_result = rset_open (result, RSETF_WRITE);
+
+ while (more[0] && more[1])
{
- rset_close (rset[i], rsfd[i]);
- xfree (buf[i]);
+ int cmp = key_compare_it (buf[0], buf[1]);
+ if (cmp < -1)
+ more[0] = rset_read (rset[0], rsfd[0], buf[0], &term_index);
+ else if (cmp > 1)
+ more[1] = rset_read (rset[1], rsfd[1], buf[1], &term_index);
+ else
+ {
+ int sysno = buf[0]->sysno;
+ int seqno[500];
+ int n = 0;
+
+ seqno[n++] = buf[0]->seqno;
+ while ((more[0] = rset_read (rset[0], rsfd[0], buf[0],
+ &term_index)) &&
+ sysno == buf[0]->sysno)
+ if (n < 500)
+ seqno[n++] = buf[0]->seqno;
+ do
+ {
+ for (i = 0; i<n; i++)
+ {
+ int diff = buf[1]->seqno - seqno[i];
+ int excl = exclusion;
+ if (!ordered && diff < 0)
+ diff = -diff;
+ switch (relation)
+ {
+ case 1: /* < */
+ if (diff < distance && diff >= 0)
+ excl = !excl;
+ break;
+ case 2: /* <= */
+ if (diff <= distance && diff >= 0)
+ excl = !excl;
+ break;
+ case 3: /* == */
+ if (diff == distance && diff >= 0)
+ excl = !excl;
+ break;
+ case 4: /* >= */
+ if (diff >= distance && diff >= 0)
+ excl = !excl;
+ break;
+ case 5: /* > */
+ if (diff > distance && diff >= 0)
+ excl = !excl;
+ break;
+ case 6: /* != */
+ if (diff != distance && diff >= 0)
+ excl = !excl;
+ break;
+ }
+ if (excl)
+ {
+ rset_write (result, rsfd_result, buf[1]);
+ break;
+ }
+ }
+ } while ((more[1] = rset_read (rset[1], rsfd[1], buf[1],
+ &term_index)) &&
+ sysno == buf[1]->sysno);
+ }
}
rset_close (result, rsfd_result);
}
+ else
+ {
+ rset_null_parms parms;
+
+ parms.rset_term = rset_term_create (prox_term, length_prox_term,
+ flags);
+ parms.rset_term->nn = 0;
+ result = rset_create (rset_kind_null, &parms);
+ }
+ for (i = 0; i<rset_no; i++)
+ {
+ if (rsfd[i])
+ rset_close (rset[i], rsfd[i]);
+ xfree (buf[i]);
+ }
xfree (buf);
xfree (more);
xfree (rsfd);
break;
}
if (ex_list)
- wrbuf = zebra_replace(zh->zebra_maps, reg_id, ex_list,
+ wrbuf = zebra_replace(zh->service->zebra_maps, reg_id, ex_list,
termz, strlen(termz));
if (!wrbuf)
return nmem_strdup(stream, termz);
}
else if (rset_no == 1)
return (rset[0]);
- result = rpn_prox (zh, rset, rset_no);
+ result = rpn_prox (zh, rset, rset_no, 1, 0, 3, 1);
for (i = 0; i<rset_no; i++)
rset_delete (rset[i]);
return result;
logf (LOG_DEBUG, "numeric relation value=%d", relation_value);
- if (!term_100 (zh->zebra_maps, reg_type, term_sub, term_tmp, 1,
+ if (!term_100 (zh->service->zebra_maps, reg_type, term_sub, term_tmp, 1,
term_dst))
return 0;
term_value = atoi (term_tmp);
sprintf (term_tmp, "(0*%d)", term_value);
}
logf (LOG_DEBUG, "dict_lookup_grep: %s", term_tmp);
- r = dict_lookup_grep (zh->dict, term_dict, 0, grep_info, max_pos,
+ r = dict_lookup_grep (zh->service->dict, term_dict, 0, grep_info, max_pos,
0, grep_handle);
if (r)
logf (LOG_WARN, "dict_lookup_grep fail, rel=gt: %d", r);
zh->errCode = 121;
return -1;
}
- if (zebraExplain_curDatabase (zh->zei, basenames[base_no]))
+ if (zebraExplain_curDatabase (zh->service->zei, basenames[base_no]))
{
zh->errCode = 109; /* Database unavailable */
zh->errString = basenames[base_no];
char ord_buf[32];
int i, ord_len;
- ord = zebraExplain_lookupSU (zh->zei, attp.attset_ordinal,
+ ord = zebraExplain_lookupSU (zh->service->zei, attp.attset_ordinal,
local_attr->local);
if (ord < 0)
continue;
parms.rset_term = rset_term_create (termz, -1, rank_type);
parms.key_size = sizeof (struct it_key);
- parms.temp_path = res_get (zh->res, "setTmpDir");
+ parms.temp_path = res_get (zh->service->res, "setTmpDir");
result = rset_create (rset_kind_temp, &parms);
rsfd = rset_open (result, RSETF_WRITE);
int sort_flag;
char termz[IT_MAX_WORD+1];
- zebra_maps_attr (zh->zebra_maps, zapt, ®_id, &search_type,
+ zebra_maps_attr (zh->service->zebra_maps, zapt, ®_id, &search_type,
&rank_type, &complete_flag, &sort_flag);
logf (LOG_DEBUG, "reg_id=%c", reg_id);
return NULL;
}
#endif
- r = rpn_proximity (zh, bool_parms.rset_l, bool_parms.rset_r,
- *zop->u.prox->ordered,
- (!zop->u.prox->exclusion ? 0 :
- *zop->u.prox->exclusion),
- *zop->u.prox->relationType,
- *zop->u.prox->distance);
+ else
+ {
+ RSET rsets[2];
+
+ rsets[0] = bool_parms.rset_l;
+ rsets[1] = bool_parms.rset_r;
+
+ r = rpn_prox (zh, rsets, 2,
+ *zop->u.prox->ordered,
+ (!zop->u.prox->exclusion ? 0 :
+ *zop->u.prox->exclusion),
+ *zop->u.prox->relationType,
+ *zop->u.prox->distance);
+ rset_delete (rsets[0]);
+ rset_delete (rsets[1]);
+ }
break;
default:
zh->errCode = 110;
attr_init (&use, zapt, 1);
use_value = attr_find (&use, &attributeset);
- if (zebra_maps_attr (zh->zebra_maps, zapt, ®_id, &search_type,
+ if (zebra_maps_attr (zh->service->zebra_maps, zapt, ®_id, &search_type,
&rank_type, &complete_flag, &sort_flag))
{
zh->errCode = 113;
*num_entries = 0;
return;
}
- if (zebraExplain_curDatabase (zh->zei, basenames[base_no]))
+ if (zebraExplain_curDatabase (zh->service->zei, basenames[base_no]))
{
zh->errString = basenames[base_no];
zh->errCode = 109; /* Database unavailable */
{
int ord;
- ord = zebraExplain_lookupSU (zh->zei, attp.attset_ordinal,
+ ord = zebraExplain_lookupSU (zh->service->zei, attp.attset_ordinal,
local_attr->local);
if (ord > 0)
ords[ord_no++] = ord;
trans_scan_term (zh, zapt, termz+prefix_len, reg_id);
- dict_scan (zh->dict, termz, &before_tmp, &after_tmp, scan_info,
- scan_handle);
+ dict_scan (zh->service->dict, termz, &before_tmp, &after_tmp,
+ scan_info, scan_handle);
}
glist = (ZebraScanEntry *)
odr_malloc (stream, (before+after)*sizeof(*glist));
projects / idzebra-moved-to-github.git / blobdiff