summaryrefslogtreecommitdiff
path: root/fts3_tokenizer.h
blob: 708ea38523ac647459439b93c19c33f1e26097af (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
/************** Begin file fts3_tokenizer.h **********************************/
/*
** 2006 July 10
**
** The author disclaims copyright to this source code.
**
*************************************************************************
** Defines the interface to tokenizers used by fulltext-search.  There
** are three basic components:
**
** sqlite3_tokenizer_module is a singleton defining the tokenizer
** interface functions.  This is essentially the class structure for
** tokenizers.
**
** sqlite3_tokenizer is used to define a particular tokenizer, perhaps
** including customization information defined at creation time.
**
** sqlite3_tokenizer_cursor is generated by a tokenizer to generate
** tokens from a particular input.
*/
#ifndef _FTS3_TOKENIZER_H_
#define _FTS3_TOKENIZER_H_

/* TODO(shess) Only used for SQLITE_OK and SQLITE_DONE at this time.
** If tokenizers are to be allowed to call sqlite3_*() functions, then
** we will need a way to register the API consistently.
*/

/*
** Structures used by the tokenizer interface. When a new tokenizer
** implementation is registered, the caller provides a pointer to
** an sqlite3_tokenizer_module containing pointers to the callback
** functions that make up an implementation.
**
** When an fts3 table is created, it passes any arguments passed to
** the tokenizer clause of the CREATE VIRTUAL TABLE statement to the
** sqlite3_tokenizer_module.xCreate() function of the requested tokenizer
** implementation. The xCreate() function in turn returns an 
** sqlite3_tokenizer structure representing the specific tokenizer to
** be used for the fts3 table (customized by the tokenizer clause arguments).
**
** To tokenize an input buffer, the sqlite3_tokenizer_module.xOpen()
** method is called. It returns an sqlite3_tokenizer_cursor object
** that may be used to tokenize a specific input buffer based on
** the tokenization rules supplied by a specific sqlite3_tokenizer
** object.
*/
typedef struct sqlite3_tokenizer_module sqlite3_tokenizer_module;
typedef struct sqlite3_tokenizer sqlite3_tokenizer;
typedef struct sqlite3_tokenizer_cursor sqlite3_tokenizer_cursor;

struct sqlite3_tokenizer_module {

  /*
  ** Structure version. Should always be set to 0 or 1.
  */
  int iVersion;

  /*
  ** Create a new tokenizer. The values in the argv[] array are the
  ** arguments passed to the "tokenizer" clause of the CREATE VIRTUAL
  ** TABLE statement that created the fts3 table. For example, if
  ** the following SQL is executed:
  **
  **   CREATE .. USING fts3( ... , tokenizer <tokenizer-name> arg1 arg2)
  **
  ** then argc is set to 2, and the argv[] array contains pointers
  ** to the strings "arg1" and "arg2".
  **
  ** This method should return either SQLITE_OK (0), or an SQLite error 
  ** code. If SQLITE_OK is returned, then *ppTokenizer should be set
  ** to point at the newly created tokenizer structure. The generic
  ** sqlite3_tokenizer.pModule variable should not be initialised by
  ** this callback. The caller will do so.
  */
  int (*xCreate)(
    int argc,                           /* Size of argv array */
    const char *const*argv,             /* Tokenizer argument strings */
    sqlite3_tokenizer **ppTokenizer     /* OUT: Created tokenizer */
  );

  /*
  ** Destroy an existing tokenizer. The fts3 module calls this method
  ** exactly once for each successful call to xCreate().
  */
  int (*xDestroy)(sqlite3_tokenizer *pTokenizer);

  /*
  ** Create a tokenizer cursor to tokenize an input buffer. The caller
  ** is responsible for ensuring that the input buffer remains valid
  ** until the cursor is closed (using the xClose() method). 
  */
  int (*xOpen)(
    sqlite3_tokenizer *pTokenizer,       /* Tokenizer object */
    const char *pInput, int nBytes,      /* Input buffer */
    sqlite3_tokenizer_cursor **ppCursor  /* OUT: Created tokenizer cursor */
  );

  /*
  ** Destroy an existing tokenizer cursor. The fts3 module calls this 
  ** method exactly once for each successful call to xOpen().
  */
  int (*xClose)(sqlite3_tokenizer_cursor *pCursor);

  /*
  ** Retrieve the next token from the tokenizer cursor pCursor. This
  ** method should either return SQLITE_OK and set the values of the
  ** "OUT" variables identified below, or SQLITE_DONE to indicate that
  ** the end of the buffer has been reached, or an SQLite error code.
  **
  ** *ppToken should be set to point at a buffer containing the 
  ** normalized version of the token (i.e. after any case-folding and/or
  ** stemming has been performed). *pnBytes should be set to the length
  ** of this buffer in bytes. The input text that generated the token is
  ** identified by the byte offsets returned in *piStartOffset and
  ** *piEndOffset. *piStartOffset should be set to the index of the first
  ** byte of the token in the input buffer. *piEndOffset should be set
  ** to the index of the first byte just past the end of the token in
  ** the input buffer.
  **
  ** The buffer *ppToken is set to point at is managed by the tokenizer
  ** implementation. It is only required to be valid until the next call
  ** to xNext() or xClose(). 
  */
  /* TODO(shess) current implementation requires pInput to be
  ** nul-terminated.  This should either be fixed, or pInput/nBytes
  ** should be converted to zInput.
  */
  int (*xNext)(
    sqlite3_tokenizer_cursor *pCursor,   /* Tokenizer cursor */
    const char **ppToken, int *pnBytes,  /* OUT: Normalized text for token */
    int *piStartOffset,  /* OUT: Byte offset of token in input buffer */
    int *piEndOffset,    /* OUT: Byte offset of end of token in input buffer */
    int *piPosition      /* OUT: Number of tokens returned before this one */
  );

  /***********************************************************************
  ** Methods below this point are only available if iVersion>=1.
  */

  /* 
  ** Configure the language id of a tokenizer cursor.
  */
  int (*xLanguageid)(sqlite3_tokenizer_cursor *pCsr, int iLangid);
};

struct sqlite3_tokenizer {
  const sqlite3_tokenizer_module *pModule;  /* The module for this tokenizer */
  /* Tokenizer implementations will typically add additional fields */
};

struct sqlite3_tokenizer_cursor {
  sqlite3_tokenizer *pTokenizer;       /* Tokenizer for this cursor. */
  /* Tokenizer implementations will typically add additional fields */
};

int fts3_global_term_cnt(int iTerm, int iCol);
int fts3_term_cnt(int iTerm, int iCol);


#endif /* _FTS3_TOKENIZER_H_ */

/************** End of fts3_tokenizer.h **************************************/