-/*
- * Copyright (C) 1994-1999, Index Data
- * All rights reserved.
- * Sebastian Hammer, Adam Dickmeiss
- *
- * $Log: zebramap.c,v $
- * Revision 1.16 1999-09-07 07:19:21 adam
- * Work on character mapping. Implemented replace rules.
- *
- * Revision 1.15 1999/05/26 07:49:14 adam
- * C++ compilation.
- *
- * Revision 1.14 1999/02/19 10:37:40 adam
- * Minor fix.
- *
- * Revision 1.13 1999/02/18 15:01:04 adam
- * Structure=key uses register type 0.
- *
- * Revision 1.12 1999/02/12 13:29:25 adam
- * Implemented position-flag for registers.
- *
- * Revision 1.11 1998/10/13 20:09:19 adam
- * Changed call to readconf_line.
- *
- * Revision 1.10 1998/06/23 15:33:37 adam
- * Added feature to specify sort criteria in query (type 7 specifies
- * sort flags).
- *
- * Revision 1.9 1998/04/02 14:35:30 adam
- * First version of Zebra that works with compiled ASN.1.
- *
- * Revision 1.8 1998/03/05 08:42:44 adam
- * Minor changes to zebramap data structures. Query mapping rules changed.
- *
- * Revision 1.7 1998/02/10 12:03:07 adam
- * Implemented Sort.
- *
- * Revision 1.6 1998/01/29 13:36:01 adam
- * Structure word-list, free-form-text and document-text all
- * trigger ranked search.
- *
- * Revision 1.5 1997/11/19 10:22:14 adam
- * Bug fix (introduced by previous commit).
- *
- * Revision 1.4 1997/11/18 10:05:08 adam
- * Changed character map facility so that admin can specify character
- * mapping files for each register type, w, p, etc.
- *
- * Revision 1.3 1997/11/17 15:35:26 adam
- * Bug fix. Relation=relevance wasn't observed.
- *
- * Revision 1.2 1997/10/31 12:39:30 adam
- * Changed log message.
- *
- * Revision 1.1 1997/10/27 14:33:06 adam
- * Moved towards generic character mapping depending on "structure"
- * field in abstract syntax file. Fixed a few memory leaks. Fixed
- * bug with negative integers when doing searches with relational
- * operators.
- *
- */
+/* $Id: zebramap.c,v 1.42 2005-06-15 21:31:45 adam Exp $
+ Copyright (C) 1995-2005
+ Index Data ApS
+
+This file is part of the Zebra server.
+
+Zebra is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 2, or (at your option) any later
+version.
+
+Zebra is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+for more details.
+
+You should have received a copy of the GNU General Public License
+along with Zebra; see the file LICENSE.zebra. If not, write to the
+Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA
+02111-1307, USA.
+*/
#include <assert.h>
+#include <stdlib.h>
#include <ctype.h>
-#include <yaz-util.h>
#include <charmap.h>
-#include <zebramap.h>
+#include <yaz/yaz-util.h>
+
+#include <idzebra/zebramap.h>
#define ZEBRA_MAP_TYPE_SORT 1
#define ZEBRA_MAP_TYPE_INDEX 2
-struct zm_token {
- char *token_from;
- char *token_to;
- int token_min;
- struct zm_token *next;
-};
+#define ZEBRA_REPLACE_ANY 300
struct zebra_map {
unsigned reg_id;
chrmaptab maptab;
const char *maptab_name;
struct zebra_map *next;
- struct zm_token *replace_tokens;
};
struct zebra_maps {
char *tabpath;
+ char *tabroot;
NMEM nmem;
struct zebra_map *map_list;
char temp_map_str[2];
const char *temp_map_ptr[2];
struct zebra_map **lookup_array;
- WRBUF wrbuf_1, wrbuf_2;
+ WRBUF wrbuf_1;
};
void zebra_maps_close (ZebraMaps zms)
zm = zm->next;
}
wrbuf_free (zms->wrbuf_1, 1);
- wrbuf_free (zms->wrbuf_2, 1);
nmem_destroy (zms->nmem);
xfree (zms);
}
int lineno = 0;
struct zebra_map **zm = 0, *zp;
- if (!(f = yaz_path_fopen(zms->tabpath, name, "r")))
+ if (!(f = yaz_fopen(zms->tabpath, name, "r", zms->tabroot)))
{
- logf(LOG_WARN|LOG_ERRNO, "%s", name);
+ yaz_log(YLOG_WARN|YLOG_ERRNO, "%s", name);
return ;
}
while ((argc = readconf_line(f, &lineno, line, 512, argv, 10)))
(*zm)->type = ZEBRA_MAP_TYPE_INDEX;
(*zm)->completeness = 0;
(*zm)->positioned = 1;
- (*zm)->replace_tokens = 0;
}
else if (!yaz_matchstr (argv[0], "sort") && argc == 2)
{
if ((*zm)->type == ZEBRA_MAP_TYPE_SORT)
(*zm)->u.sort.entry_size = atoi (argv[1]);
}
- else if (zm && !yaz_matchstr (argv[0], "replace") && argc >= 2)
- {
- struct zm_token *token = nmem_malloc (zms->nmem, sizeof(*token));
- char *cp, *dp;
- token->next = (*zm)->replace_tokens;
- (*zm)->replace_tokens = token;
- dp = token->token_from = nmem_strdup (zms->nmem, cp = argv[1]);
- while (*cp)
- {
- if (*cp == '$')
- {
- *dp++ = ' ';
- cp++;
- }
- else
- *dp++ = zebra_prim(&cp);
- }
- *dp = '\0';
-
- if (argc >= 3)
- {
- dp = token->token_to = nmem_strdup (zms->nmem, cp = argv[2]);
- while (*cp)
- {
- if (*cp == '$')
- {
- *dp++ = ' ';
- cp++;
- }
- else
- *dp++ = zebra_prim(&cp);
- }
- *dp = '\0';
- }
- else
- token->token_to = 0;
- }
}
if (zm)
(*zm)->next = NULL;
- fclose (f);
+ yaz_fclose (f);
for (zp = zms->map_list; zp; zp = zp->next)
zms->lookup_array[zp->reg_id] = zp;
zebra_map_read (zms, value);
}
-ZebraMaps zebra_maps_open (Res res)
+ZebraMaps zebra_maps_open (Res res, const char *base)
{
ZebraMaps zms = (ZebraMaps) xmalloc (sizeof(*zms));
int i;
zms->nmem = nmem_create ();
zms->tabpath = nmem_strdup (zms->nmem,
- res_get_def (res, "profilePath", "."));
+ res_get_def (res, "profilePath",
+ DEFAULT_PROFILE_PATH));
+ zms->tabroot = 0;
+ if (base)
+ zms->tabroot = nmem_strdup (zms->nmem, base);
zms->map_list = NULL;
zms->temp_map_str[0] = '\0';
zebra_map_read (zms, "default.idx");
zms->wrbuf_1 = wrbuf_alloc();
- zms->wrbuf_2 = wrbuf_alloc();
+
return zms;
}
if (!zm)
{
zm = (struct zebra_map *) nmem_malloc (zms->nmem, sizeof(*zm));
- logf (LOG_WARN, "Unknown register type: %c", reg_id);
+ yaz_log (YLOG_WARN, "Unknown register type: %c", reg_id);
zm->reg_id = reg_id;
- zm->maptab_name = NULL;
+ zm->maptab_name = nmem_strdup (zms->nmem, "@");
zm->maptab = NULL;
zm->type = ZEBRA_MAP_TYPE_INDEX;
zm->completeness = 0;
if (!zm->maptab_name || !yaz_matchstr (zm->maptab_name, "@"))
return NULL;
if (!(zm->maptab = chrmaptab_create (zms->tabpath,
- zm->maptab_name, 0)))
- logf(LOG_WARN, "Failed to read character table %s",
+ zm->maptab_name,
+ zms->tabroot)))
+ yaz_log(YLOG_WARN, "Failed to read character table %s",
zm->maptab_name);
else
- logf(LOG_DEBUG, "Read character table %s", zm->maptab_name);
+ yaz_log(YLOG_DEBUG, "Read character table %s", zm->maptab_name);
}
return zm->maptab;
}
const char **zebra_maps_input (ZebraMaps zms, unsigned reg_id,
- const char **from, int len)
+ const char **from, int len, int first)
{
chrmaptab maptab;
maptab = zebra_charmap_get (zms, reg_id);
if (maptab)
- return chr_map_input(maptab, from, len);
+ return chr_map_input(maptab, from, len, first);
zms->temp_map_str[0] = **from;
return zms->temp_map_ptr;
}
-#if 0
-int zebra_maps_input_tokens (ZebraMaps zms, unsigned reg_id,
- const char *input_str, int input_len,
- WRBUF wrbuf)
+const char **zebra_maps_search(ZebraMaps zms, unsigned reg_id,
+ const char **from, int len, int *q_map_match)
{
- chrmaptab maptab = zebra_charmap_get (zms, reg_id);
- int len[4];
- char *str[3];
- int input_i = 0;
- int first = 1;
- const char **out;
-
- if (!maptab)
- {
- wrbuf_write (wrbuf, input_str, input_len);
- return -1;
- }
- str[0] = " ";
- len[0] = 1;
- str[1] = input_str;
- len[1] = input_len;
- str[2] = " ";
- len[2] = 1;
- len[3] = -1;
+ chrmaptab maptab;
- out = chr_map_input (maptab, str, len);
- while (len[1] > 0)
+ *q_map_match = 0;
+ maptab = zebra_charmap_get (zms, reg_id);
+ if (maptab)
{
- while (out && *out && **out == *CHR_SPACE)
- out = chr_map_input (maptab, str, len);
+ const char **map;
+ map = chr_map_q_input(maptab, from, len, 0);
+ if (map && map[0])
+ {
+ *q_map_match = 1;
+ return map;
+ }
+ map = chr_map_input(maptab, from, len, 0);
+ if (map)
+ return map;
}
+ zms->temp_map_str[0] = **from;
+
+ (*from)++;
+ return zms->temp_map_ptr;
}
-#endif
const char *zebra_maps_output(ZebraMaps zms, unsigned reg_id,
const char **from)
static void attr_init_APT (AttrType *src, Z_AttributesPlusTerm *zapt, int type)
{
-#ifdef ASN_COMPILED
src->attributeList = zapt->attributes->attributes;
src->num_attributes = zapt->attributes->num_attributes;
-#else
- src->attributeList = zapt->attributeList;
- src->num_attributes = zapt->num_attributes;
-#endif
src->type = type;
src->major = 0;
src->minor = 0;
return 0;
}
-int zebra_maps_sort (ZebraMaps zms, Z_SortAttributes *sortAttributes)
+int zebra_maps_sort (ZebraMaps zms, Z_SortAttributes *sortAttributes,
+ int *numerical)
{
AttrType use;
+ AttrType structure;
+ int structure_value;
attr_init_AttrList (&use, sortAttributes->list, 1);
+ attr_init_AttrList (&structure, sortAttributes->list, 4);
+ *numerical = 0;
+ structure_value = attr_find (&structure, 0);
+ if (structure_value == 109)
+ *numerical = 1;
return attr_find (&use, NULL);
}
int zebra_maps_attr (ZebraMaps zms, Z_AttributesPlusTerm *zapt,
- unsigned *reg_id, char **search_type, char **rank_type,
+ unsigned *reg_id, char **search_type, char *rank_type,
int *complete_flag, int *sort_flag)
{
AttrType completeness;
AttrType structure;
AttrType relation;
AttrType sort_relation;
+ AttrType weight;
+ AttrType use;
int completeness_value;
int structure_value;
int relation_value;
int sort_relation_value;
+ int weight_value;
+ int use_value;
attr_init_APT (&structure, zapt, 4);
attr_init_APT (&completeness, zapt, 6);
attr_init_APT (&relation, zapt, 2);
attr_init_APT (&sort_relation, zapt, 7);
+ attr_init_APT (&weight, zapt, 9);
+ attr_init_APT (&use, zapt, 1);
completeness_value = attr_find (&completeness, NULL);
structure_value = attr_find (&structure, NULL);
relation_value = attr_find (&relation, NULL);
sort_relation_value = attr_find (&sort_relation, NULL);
+ weight_value = attr_find (&weight, NULL);
+ use_value = attr_find(&use, NULL);
if (completeness_value == 2 || completeness_value == 3)
*complete_flag = 1;
*sort_flag = (sort_relation_value > 0) ? 1 : 0;
*search_type = "phrase";
- *rank_type = "void";
+ strcpy (rank_type, "void");
if (relation_value == 102)
- *rank_type = "rank";
-
+ {
+ if (weight_value == -1)
+ weight_value = 34;
+ sprintf (rank_type, "rank,w=%d,u=%d", weight_value, use_value);
+ }
+ if (relation_value == 103)
+ {
+ *search_type = "always";
+ return 0;
+ }
if (*complete_flag)
*reg_id = 'p';
else
*reg_id = '0';
*search_type = "phrase";
break;
+ case 4: /* year */
+ *reg_id = 'y';
+ *search_type = "phrase";
+ break;
+ case 5: /* date */
+ *reg_id = 'd';
+ *search_type = "phrase";
+ break;
default:
return -1;
}
return 0;
}
-int zebra_replace_sub(ZebraMaps zms, unsigned reg_id, const char *ex_list,
- const char *input_str, int input_len, WRBUF wrbuf);
-
WRBUF zebra_replace(ZebraMaps zms, unsigned reg_id, const char *ex_list,
const char *input_str, int input_len)
{
- struct zebra_map *zm = zebra_map_get (zms, reg_id);
-
wrbuf_rewind(zms->wrbuf_1);
wrbuf_write(zms->wrbuf_1, input_str, input_len);
- if (!zm->replace_tokens)
- return zms->wrbuf_1;
-
-#if 0
- logf (LOG_LOG, "zebra_replace");
- logf (LOG_LOG, "in:%.*s:", wrbuf_len(zms->wrbuf_1),
- wrbuf_buf(zms->wrbuf_1));
-#endif
- for (;;)
- {
- if (!zebra_replace_sub(zms, reg_id, ex_list, wrbuf_buf(zms->wrbuf_1),
- wrbuf_len(zms->wrbuf_1), zms->wrbuf_2))
- return zms->wrbuf_2;
- if (!zebra_replace_sub(zms, reg_id, ex_list, wrbuf_buf(zms->wrbuf_2),
- wrbuf_len(zms->wrbuf_2), zms->wrbuf_1))
- return zms->wrbuf_1;
- }
- return 0;
+ return zms->wrbuf_1;
}
-int zebra_replace_sub(ZebraMaps zms, unsigned reg_id, const char *ex_list,
- const char *input_str, int input_len, WRBUF wrbuf)
-{
- int i = -1;
- int no_replaces = 0;
- struct zebra_map *zm = zebra_map_get (zms, reg_id);
-
- wrbuf_rewind(wrbuf);
- for (i = -1; i <= input_len; )
- {
- struct zm_token *token;
- char replace_string[128];
- int replace_out;
- int replace_in = 0;
-
- for (token = zm->replace_tokens; !replace_in && token;
- token = token->next)
- {
- int j = 0;
- int replace_done = 0;
- replace_out = 0;
- for (;; j++)
- {
- int c;
- if (!token->token_from[j])
- {
- replace_in = j;
- break;
- }
- if (ex_list && strchr (ex_list, token->token_from[j]))
- break;
- if (i+j < 0 || j+i >= input_len)
- c = ' ';
- else
- c = tolower(input_str[j+i]);
- if (token->token_from[j] == '.')
- {
- if (c == ' ')
- break;
- replace_string[replace_out++] = c;
- }
- else
- {
- if (c != token->token_from[j])
- break;
- if (!replace_done)
- {
- const char *cp = token->token_to;
- replace_done = 1;
- for (; cp && *cp; cp++)
- replace_string[replace_out++] = *cp;
- }
- }
- }
- }
- if (!replace_in)
- {
- if (i >= 0 && i < input_len)
- wrbuf_putc(wrbuf, input_str[i]);
- i++;
- }
- else
- {
- no_replaces++;
- if (replace_out)
- wrbuf_write(wrbuf, replace_string, replace_out);
- i += replace_in;
- }
- }
-#if 0
- logf (LOG_LOG, "out:%.*s:", wrbuf_len(wrbuf), wrbuf_buf(wrbuf));
-#endif
- return no_replaces;
-}