-/* $Id: lookgrep.c,v 1.28 2004-12-08 12:23:08 adam Exp $
- Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002
- 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.
+/* This file is part of the Zebra server.
+ Copyright (C) 1994-2011 Index Data
+
+ 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 this program; if not, write to the Free Software
+ Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+
*/
#define WORD_BITS 32
#define MAX_LENGTH 1024
+/* This code is based
+ * Sun Wu and Udi Manber: Fast Text Searching Allowing Errors.
+ * Communications of the ACM, pp. 83-91, Vol. 35, No. 10, Oct. 1992, USA.
+ * PostScript version of the paper in its submitted form: agrep1.ps)
+ * recommended reading to understand AGREP !
+ *
+ * http://www.tgries.de/agrep/#AGREP1PS
+ * http://www.tgries.de/agrep/doc/agrep1ps.zip
+ */
+
typedef struct {
- int n; /* no of MatchWord needed */
- int range; /* max no. of errors */
- int fact; /* (range+1)*n */
- MatchWord *match_mask; /* match_mask */
+ int n; /* no of MatchWord needed */
+ int range; /* max no. of errors */
+ int fact; /* (range+1)*n */
+ MatchWord *match_mask; /* match_mask */
} MatchContext;
#define INLINE
-static INLINE void set_bit (MatchContext *mc, MatchWord *m, int ch, int state)
+static INLINE void set_bit(MatchContext *mc, MatchWord *m, int ch, int state)
{
int off = state & (WORD_BITS-1);
int wno = state / WORD_BITS;
m[mc->n * ch + wno] |= 1<<off;
}
-static INLINE MatchWord get_bit (MatchContext *mc, MatchWord *m, int ch,
- int state)
+static INLINE MatchWord get_bit(MatchContext *mc, MatchWord *m, int ch,
+ int state)
{
int off = state & (WORD_BITS-1);
int wno = state / WORD_BITS;
return m[mc->n * ch + wno] & (1<<off);
}
-static MatchContext *mk_MatchContext (struct DFA *dfa, int range)
+static MatchContext *mk_MatchContext(struct DFA *dfa, int range)
{
- MatchContext *mc = (MatchContext *) xmalloc (sizeof(*mc));
+ MatchContext *mc = (MatchContext *) xmalloc(sizeof(*mc));
int s;
mc->n = (dfa->no_states+WORD_BITS) / WORD_BITS;
mc->range = range;
mc->fact = (range+1)*mc->n;
- mc->match_mask = (MatchWord *) xcalloc (mc->n, sizeof(*mc->match_mask));
+ mc->match_mask = (MatchWord *) xcalloc(mc->n, sizeof(*mc->match_mask));
for (s = 0; s<dfa->no_states; s++)
if (dfa->states[s]->rule_no)
- set_bit (mc, mc->match_mask, 0, s);
+ set_bit(mc, mc->match_mask, 0, s);
return mc;
}
-static void rm_MatchContext (MatchContext **mc)
+static void rm_MatchContext(MatchContext **mc)
{
- xfree ((*mc)->match_mask);
- xfree (*mc);
+ xfree((*mc)->match_mask);
+ xfree(*mc);
*mc = NULL;
}
-static void mask_shift (MatchContext *mc, MatchWord *Rdst, MatchWord *Rsrc,
- struct DFA *dfa, int ch)
+static void mask_shift(MatchContext *mc, MatchWord *Rdst, MatchWord *Rsrc,
+ struct DFA *dfa, int ch)
{
int j, s = 0;
MatchWord *Rsrc_p = Rsrc, mask;
while (--i >= 0)
if (ch >= state->trans[i].ch[0] &&
ch <= state->trans[i].ch[1])
- set_bit (mc, Rdst, 0, state->trans[i].to);
+ set_bit(mc, Rdst, 0, state->trans[i].to);
}
if (mask & 2)
{
while (--i >= 0)
if (ch >= state->trans[i].ch[0] &&
ch <= state->trans[i].ch[1])
- set_bit (mc, Rdst, 0, state->trans[i].to);
+ set_bit(mc, Rdst, 0, state->trans[i].to);
}
if (mask & 4)
{
while (--i >= 0)
if (ch >= state->trans[i].ch[0] &&
ch <= state->trans[i].ch[1])
- set_bit (mc, Rdst, 0, state->trans[i].to);
+ set_bit(mc, Rdst, 0, state->trans[i].to);
}
if (mask & 8)
{
while (--i >= 0)
if (ch >= state->trans[i].ch[0] &&
ch <= state->trans[i].ch[1])
- set_bit (mc, Rdst, 0, state->trans[i].to);
+ set_bit(mc, Rdst, 0, state->trans[i].to);
}
}
s += 4;
}
}
-static void shift (MatchContext *mc, MatchWord *Rdst, MatchWord *Rsrc,
- struct DFA *dfa)
+static void shift(MatchContext *mc, MatchWord *Rdst, MatchWord *Rsrc,
+ struct DFA *dfa)
{
int j, s = 0;
MatchWord *Rsrc_p = Rsrc, mask;
struct DFA_state *state = dfa->states[s];
int i = state->tran_no;
while (--i >= 0)
- set_bit (mc, Rdst, 0, state->trans[i].to);
+ set_bit(mc, Rdst, 0, state->trans[i].to);
}
if (mask & 2)
{
struct DFA_state *state = dfa->states[s+1];
int i = state->tran_no;
while (--i >= 0)
- set_bit (mc, Rdst, 0, state->trans[i].to);
+ set_bit(mc, Rdst, 0, state->trans[i].to);
}
if (mask & 4)
{
struct DFA_state *state = dfa->states[s+2];
int i = state->tran_no;
while (--i >= 0)
- set_bit (mc, Rdst, 0, state->trans[i].to);
+ set_bit(mc, Rdst, 0, state->trans[i].to);
}
if (mask & 8)
{
struct DFA_state *state = dfa->states[s+3];
int i = state->tran_no;
while (--i >= 0)
- set_bit (mc, Rdst, 0, state->trans[i].to);
+ set_bit(mc, Rdst, 0, state->trans[i].to);
}
}
s += 4;
}
}
-static void or (MatchContext *mc, MatchWord *Rdst,
- MatchWord *Rsrc1, MatchWord *Rsrc2)
+static void or(MatchContext *mc, MatchWord *Rdst,
+ MatchWord *Rsrc1, MatchWord *Rsrc2)
{
int i;
for (i = 0; i<mc->n; i++)
Rdst[i] = Rsrc1[i] | Rsrc2[i];
}
-static INLINE int move (MatchContext *mc, MatchWord *Rj1, MatchWord *Rj,
- Dict_char ch, struct DFA *dfa, MatchWord *Rtmp,
- int range)
+static INLINE int move(MatchContext *mc, MatchWord *Rj1, MatchWord *Rj,
+ Dict_char ch, struct DFA *dfa, MatchWord *Rtmp,
+ int range)
{
int d;
MatchWord *Rtmp_2 = Rtmp + mc->n;
- mask_shift (mc, Rj1, Rj, dfa, ch);
+ mask_shift(mc, Rj1, Rj, dfa, ch);
for (d = 1; d <= mc->range; d++)
{
- or (mc, Rtmp, Rj, Rj1); /* 2,3 */
+ or(mc, Rtmp, Rj, Rj1); /* 2,3 */
- shift (mc, Rtmp_2, Rtmp, dfa);
+ shift(mc, Rtmp_2, Rtmp, dfa);
- mask_shift (mc, Rtmp, Rj+mc->n, dfa, ch); /* 1 */
+ mask_shift(mc, Rtmp, Rj+mc->n, dfa, ch); /* 1 */
- or (mc, Rtmp, Rtmp_2, Rtmp); /* 1,2,3*/
+ or(mc, Rtmp, Rtmp_2, Rtmp); /* 1,2,3*/
Rj1 += mc->n;
- or (mc, Rj1, Rtmp, Rj); /* 1,2,3,4 */
+ or(mc, Rj1, Rtmp, Rj); /* 1,2,3,4 */
Rj += mc->n;
}
}
-static int dict_grep (Dict dict, Dict_ptr ptr, MatchContext *mc,
- MatchWord *Rj, int pos, void *client,
- int (*userfunc)(char *, const char *, void *),
- Dict_char *prefix, struct DFA *dfa,
- int *max_pos, int init_pos)
+static int grep(Dict dict, Dict_ptr ptr, MatchContext *mc,
+ MatchWord *Rj, int pos, void *client,
+ int (*userfunc)(char *, const char *, void *),
+ Dict_char *prefix, struct DFA *dfa,
+ int *max_pos, int init_pos)
{
int lo, hi, d;
void *p;
short *indxp;
char *info;
- dict_bf_readp (dict->dbf, ptr, &p);
+ dict_bf_readp(dict->dbf, ptr, &p);
lo = 0;
hi = DICT_nodir(p)-1;
indxp = (short*) ((char*) p+DICT_bsize(p)-sizeof(short));
MatchWord *Rj_tmp = Rj + (j+2)*mc->fact;
int range;
- memcpy (&ch, info+j*sizeof(Dict_char), sizeof(Dict_char));
+ memcpy(&ch, info+j*sizeof(Dict_char), sizeof(Dict_char));
prefix[pos+j] = ch;
if (pos+j > *max_pos)
*max_pos = pos+j;
if (ch == DICT_EOS)
{
if (was_match)
- if ((*userfunc)((char*) prefix,
- info+(j+1)*sizeof(Dict_char), client))
- return 1;
+ {
+ int ret = userfunc((char*) prefix,
+ info+(j+1)*sizeof(Dict_char), client);
+ if (ret)
+ return ret;
+ }
break;
}
if (pos+j >= init_pos)
range = mc->range;
else
range = 0;
- move (mc, Rj1, Rj0, ch, dfa, Rj_tmp, range);
+ move(mc, Rj1, Rj0, ch, dfa, Rj_tmp, range);
for (d = mc->n; --d >= 0; )
if (Rj1[range*mc->n + d])
break;
/* unsigned char length of information */
/* char * information */
info = (char*)p - indxp[-lo];
- memcpy (&ch, info+sizeof(Dict_ptr), sizeof(Dict_char));
+ memcpy(&ch, info+sizeof(Dict_ptr), sizeof(Dict_char));
prefix[pos] = ch;
if (pos > *max_pos)
range = mc->range;
else
range = 0;
- move (mc, Rj1, Rj, ch, dfa, Rj_tmp, range);
+ move(mc, Rj1, Rj, ch, dfa, Rj_tmp, range);
for (d = mc->n; --d >= 0; )
if (Rj1[range*mc->n + d])
break;
for (d = mc->n; --d >= 0; )
if (Rj1[range*mc->n + d] & mc->match_mask[d])
{
+ int ret;
prefix[pos+1] = DICT_EOS;
- if ((*userfunc)((char*) prefix,
- info+sizeof(Dict_ptr)+
- sizeof(Dict_char), client))
- return 1;
+ ret = userfunc((char*) prefix,
+ info+sizeof(Dict_ptr)+
+ sizeof(Dict_char), client);
+ if (ret)
+ return ret;
break;
}
}
- memcpy (&subptr, info, sizeof(Dict_ptr));
+ memcpy(&subptr, info, sizeof(Dict_ptr));
if (subptr)
{
- if (dict_grep (dict, subptr, mc, Rj1, pos+1,
+ int ret = grep(dict, subptr, mc, Rj1, pos+1,
client, userfunc, prefix, dfa, max_pos,
- init_pos))
- return 1;
- dict_bf_readp (dict->dbf, ptr, &p);
+ init_pos);
+ if (ret)
+ return ret;
+
+ dict_bf_readp(dict->dbf, ptr, &p);
indxp = (short*) ((char*) p+DICT_bsize(p)-sizeof(short));
}
}
return 0;
}
-int dict_lookup_grep (Dict dict, const char *pattern, int range, void *client,
- int *max_pos, int init_pos,
- int (*userfunc)(char *name, const char *info,
- void *client))
+int dict_lookup_grep(Dict dict, const char *pattern, int range, void *client,
+ int *max_pos, int init_pos,
+ int (*userfunc)(char *name, const char *info,
+ void *client))
{
MatchWord *Rj;
Dict_char prefix[MAX_LENGTH+1];
const char *this_pattern = pattern;
MatchContext *mc;
struct DFA *dfa = dfa_init();
- int i, d;
+ int i, d, ret = 0;
#if 0
debug_dfa_trav = 1;
dfa_verbose = 1;
#endif
- yaz_log (YLOG_DEBUG, "dict_lookup_grep range=%d", range);
+ dfa_anyset_includes_nl(dfa);
+
+ yaz_log(YLOG_DEBUG, "dict_lookup_grep range=%d", range);
for (i = 0; pattern[i]; i++)
{
- yaz_log (YLOG_DEBUG, " %3d %c", pattern[i],
- (pattern[i] > ' ' && pattern[i] < 127) ? pattern[i] : '?');
+ yaz_log(YLOG_DEBUG, " %2d %3d %c", i, pattern[i],
+ (pattern[i] > ' ' && pattern[i] < 127) ? pattern[i] : '?');
}
- dfa_set_cmap (dfa, dict->grep_cmap_data, dict->grep_cmap);
+ dfa_set_cmap(dfa, dict->grep_cmap_data, dict->grep_cmap);
- i = dfa_parse (dfa, &this_pattern);
+ i = dfa_parse(dfa, &this_pattern);
if (i || *this_pattern)
{
- yaz_log (YLOG_WARN, "dfa_parse fail=%d", i);
- dfa_delete (&dfa);
+ yaz_log(YLOG_WARN, "dfa_parse fail=%d", i);
+ dfa_delete(&dfa);
return -1;
}
- dfa_mkstate (dfa);
+ dfa_mkstate(dfa);
- mc = mk_MatchContext (dfa, range);
+ mc = mk_MatchContext(dfa, range);
- Rj = (MatchWord *) xcalloc ((MAX_LENGTH+1) * mc->n, sizeof(*Rj));
+ Rj = (MatchWord *) xcalloc((MAX_LENGTH+1) * mc->n, sizeof(*Rj));
set_bit (mc, Rj, 0, 0);
for (d = 1; d<=mc->range; d++)
{
int s;
- memcpy (Rj + mc->n * d, Rj + mc->n * (d-1), mc->n * sizeof(*Rj));
- for (s = 0; s<dfa->no_states; s++)
+ memcpy(Rj + mc->n * d, Rj + mc->n * (d-1), mc->n * sizeof(*Rj));
+ for (s = 0; s < dfa->no_states; s++)
{
- if (get_bit (mc, Rj, d-1, s))
+ if (get_bit(mc, Rj, d-1, s))
{
struct DFA_state *state = dfa->states[s];
int i = state->tran_no;
while (--i >= 0)
- set_bit (mc, Rj, d, state->trans[i].to);
+ set_bit(mc, Rj, d, state->trans[i].to);
}
}
}
*max_pos = 0;
if (dict->head.root)
- i = dict_grep (dict, dict->head.root, mc, Rj, 0, client,
- userfunc, prefix,
- dfa, max_pos, init_pos);
- else
- i = 0;
- yaz_log (YLOG_DEBUG, "max_pos = %d", *max_pos);
- dfa_delete (&dfa);
- xfree (Rj);
- rm_MatchContext (&mc);
- return i;
+ ret = grep(dict, dict->head.root, mc, Rj, 0, client,
+ userfunc, prefix,
+ dfa, max_pos, init_pos);
+ yaz_log(YLOG_DEBUG, "max_pos = %d", *max_pos);
+ dfa_delete(&dfa);
+ xfree(Rj);
+ rm_MatchContext(&mc);
+ return ret;
}
-void dict_grep_cmap (Dict dict, void *vp,
- const char **(*cmap)(void *vp, const char **from, int len))
+void dict_grep_cmap(Dict dict, void *vp,
+ const char **(*cmap)(void *vp, const char **from, int len))
{
dict->grep_cmap = cmap;
dict->grep_cmap_data = vp;
}
+/*
+ * Local variables:
+ * c-basic-offset: 4
+ * c-file-style: "Stroustrup"
+ * indent-tabs-mode: nil
+ * End:
+ * vim: shiftwidth=4 tabstop=8 expandtab
+ */
+
projects / idzebra-moved-to-github.git / blobdiff