/External/Mysql-5.0/include/my_trie.h

  1/* Copyright (C) 2005 MySQL AB
  2
  3   This program is free software; you can redistribute it and/or modify
  4   it under the terms of the GNU General Public License as published by
  5   the Free Software Foundation; version 2 of the License.
  6
  7   This program is distributed in the hope that it will be useful,
  8   but WITHOUT ANY WARRANTY; without even the implied warranty of
  9   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 10   GNU General Public License for more details.
 11
 12   You should have received a copy of the GNU General Public License
 13   along with this program; if not, write to the Free Software
 14   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA */
 15
 16#ifndef _trie_h
 17#define _trie_h
 18#ifdef	__cplusplus
 19extern "C" {
 20#endif
 21
 22typedef struct st_trie_node
 23{
 24  uint16 leaf;                /* Depth from root node if match, 0 else */
 25  uchar c;                    /* Label on this edge */
 26  struct st_trie_node *next;  /* Next label */
 27  struct st_trie_node *links; /* Array of edges leaving this node */
 28  struct st_trie_node *fail;  /* AC failure function */
 29} TRIE_NODE;
 30
 31typedef struct st_trie
 32{
 33  TRIE_NODE root;
 34  MEM_ROOT mem_root;
 35  CHARSET_INFO *charset;
 36  uint32 nnodes;
 37  uint32 nwords;
 38} TRIE;
 39
 40typedef struct st_ac_trie_state
 41{
 42  TRIE *trie;
 43  TRIE_NODE *node;
 44} AC_TRIE_STATE;
 45
 46extern TRIE *trie_init (TRIE *trie, CHARSET_INFO *charset);
 47extern void trie_free (TRIE *trie);
 48extern my_bool trie_insert (TRIE *trie, const uchar *key, uint keylen);
 49extern my_bool ac_trie_prepare (TRIE *trie);
 50extern void ac_trie_init (TRIE *trie, AC_TRIE_STATE *state);
 51
 52
 53/* `trie_goto' is internal function and shouldn't be used. */
 54
 55static inline TRIE_NODE *trie_goto (TRIE_NODE *root, TRIE_NODE *node, uchar c)
 56{
 57  TRIE_NODE *next;
 58  DBUG_ENTER("trie_goto");
 59  for (next= node->links; next; next= next->next)
 60    if (next->c == c)
 61      DBUG_RETURN(next);
 62  if (root == node)
 63    DBUG_RETURN(root);
 64  DBUG_RETURN(NULL);
 65}
 66
 67
 68/*
 69  SYNOPSIS
 70    int ac_trie_next (AC_TRIE_STATE *state, uchar *c);
 71    state - valid pointer to `AC_TRIE_STATE'
 72    c - character to lookup
 73
 74  DESCRIPTION
 75    Implementation of search using Aho-Corasick automaton.
 76    Performs char-by-char search.
 77
 78  RETURN VALUE
 79    `ac_trie_next' returns length of matched word or 0.
 80*/
 81
 82static inline int ac_trie_next (AC_TRIE_STATE *state, uchar *c)
 83{
 84  TRIE_NODE *root, *node;
 85  DBUG_ENTER("ac_trie_next");
 86  DBUG_ASSERT(state && c);
 87  root= &state->trie->root;
 88  node= state->node;
 89  while (! (state->node= trie_goto(root, node, *c)))
 90    node= node->fail;
 91  DBUG_RETURN(state->node->leaf);
 92}
 93
 94
 95/*
 96  SYNOPSIS
 97    my_bool trie_search (TRIE *trie, const uchar *key, uint keylen);
 98    trie - valid pointer to `TRIE'
 99    key - valid pointer to key to insert
100    keylen - non-0 key length
101
102  DESCRIPTION
103    Performs key lookup in trie.
104
105  RETURN VALUE
106    `trie_search' returns `true' if key is in `trie'. Otherwise,
107    `false' is returned.
108
109  NOTES
110    Consecutive search here is "best by test". arrays are very short, so
111    binary search or hashing would add too much complexity that would
112    overweight speed gain. Especially because compiler can optimize simple
113    consecutive loop better (tested)
114*/
115
116static inline my_bool trie_search (TRIE *trie, const uchar *key, uint keylen)
117{
118  TRIE_NODE *node;
119  uint k;
120  DBUG_ENTER("trie_search");
121  DBUG_ASSERT(trie && key && keylen);
122  node= &trie->root;
123
124  for (k= 0; k < keylen; k++)
125  {
126    uchar p;
127    if (! (node= node->links))
128      DBUG_RETURN(FALSE);
129    p= key[k];
130    while (p != node->c)
131      if (! (node= node->next))
132        DBUG_RETURN(FALSE);
133  }
134
135  DBUG_RETURN(node->leaf > 0);
136}
137
138#ifdef	__cplusplus
139}
140#endif
141#endif