-/* $Id: zebramap.c,v 1.33 2004-09-14 14:38:08 quinn Exp $
- Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002,2003,2004
- Index Data Aps
+/* $Id: zebramap.c,v 1.43 2005-08-30 12:23:02 adam Exp $
+ Copyright (C) 1995-2005
+ Index Data ApS
This file is part of the Zebra server.
02111-1307, USA.
*/
-
-
#include <assert.h>
+#include <stdlib.h>
#include <ctype.h>
-#include <yaz/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
#define ZEBRA_REPLACE_ANY 300
-struct zm_token {
- int *token_from;
- char *token_to;
- int token_min;
- struct zm_token *next;
-};
-
struct zebra_map {
unsigned reg_id;
int completeness;
chrmaptab maptab;
const char *maptab_name;
struct zebra_map *next;
- struct zm_token *replace_tokens;
};
struct zebra_maps {
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);
}
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)
{
(*zm)->maptab = NULL;
(*zm)->completeness = 0;
(*zm)->positioned = 0;
- (*zm)->replace_tokens = 0;
}
else if (zm && !yaz_matchstr (argv[0], "charmap") && 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));
- token->next = (*zm)->replace_tokens;
- (*zm)->replace_tokens = token;
-#if 0
- logf (LOG_LOG, "replace %s", argv[1]);
-#endif
- token->token_from = 0;
- if (argc >= 2)
- {
- char *cp = argv[1];
- int *dp = token->token_from = (int *)
- nmem_malloc (zms->nmem, (1+strlen(cp))*sizeof(int));
- while (*cp)
- if (*cp == '$')
- {
- *dp++ = ' ';
- cp++;
- }
- else if (*cp == '.')
- {
- *dp++ = ZEBRA_REPLACE_ANY;
- cp++;
- }
- else
- {
- *dp++ = zebra_prim(&cp);
-#if 0
- logf (LOG_LOG, " char %2X %c", dp[-1], dp[-1]);
-#endif
- }
- *dp = '\0';
- }
- if (argc >= 3)
- {
- char *cp = argv[2];
- char *dp = token->token_to =
- nmem_malloc (zms->nmem, strlen(cp)+1);
- while (*cp)
- if (*cp == '$')
- {
- *dp++ = ' ';
- cp++;
- }
- else
- *dp++ = zebra_prim(&cp);
- *dp = '\0';
- }
- else
- token->token_to = 0;
- }
}
if (zm)
(*zm)->next = NULL;
zebra_map_read (zms, "default.idx");
zms->wrbuf_1 = wrbuf_alloc();
- zms->wrbuf_2 = wrbuf_alloc();
return zms;
}
struct zebra_map *zebra_map_get (ZebraMaps zms, unsigned reg_id)
{
+ assert(reg_id >= 0 && reg_id <= 255);
return zms->lookup_array[reg_id];
}
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 = nmem_strdup (zms->nmem, "@");
zm->type = ZEBRA_MAP_TYPE_INDEX;
zm->completeness = 0;
zm->next = zms->map_list;
- zm->replace_tokens = 0;
zms->map_list = zm->next;
zms->lookup_array[zm->reg_id & 255] = zm;
if (!zm->maptab_name || !yaz_matchstr (zm->maptab_name, "@"))
return NULL;
if (!(zm->maptab = chrmaptab_create (zms->tabpath,
- zm->maptab_name, 0,
+ zm->maptab_name,
zms->tabroot)))
- logf(LOG_WARN, "Failed to read character table %s",
+ 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;
}
return zms->temp_map_ptr;
}
+const char **zebra_maps_search(ZebraMaps zms, unsigned reg_id,
+ const char **from, int len, int *q_map_match)
+{
+ chrmaptab maptab;
+
+ *q_map_match = 0;
+ maptab = zebra_charmap_get (zms, reg_id);
+ if (maptab)
+ {
+ 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;
+}
+
const char *zebra_maps_output(ZebraMaps zms, unsigned reg_id,
const char **from)
{
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 || !zm->replace_tokens)
- return zms->wrbuf_1;
-
-#if 0
- 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 = 0;
- 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 = input_str[j+i] & 255;
- if (token->token_from[j] == ZEBRA_REPLACE_ANY)
- {
- 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;
-}