/ Check-in [0ba67b64]
Login
SQLite training in Houston TX on 2019-11-05 (details)
Part of the 2019 Tcl Conference

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:Modify the fts3tokenize table implementation so that it does not use the SQL function fts3_tokenizer. The user may have installed an authorizer callback that prohibits this.
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: 0ba67b64de258883e4c43db09e131bb67083855e
User & Date: dan 2013-05-07 12:16:48
Context
2013-05-08
14:20
Fix the wholenumber virtual table so that it returns higher costs for unconstrained usage. Cherrypick from [ceff8955020cd13]. check-in: 7227b615 user: drh tags: trunk
2013-05-07
17:49
Make sure the ORDER BY collating sequences are compatible with the comparison collations before using the merge algorithm for compound SELECT statements. Candidate fix for ticket [6709574d2a8d8]. Closed-Leaf check-in: fc3630cd user: drh tags: tkt-6709574
12:16
Modify the fts3tokenize table implementation so that it does not use the SQL function fts3_tokenizer. The user may have installed an authorizer callback that prohibits this. check-in: 0ba67b64 user: dan tags: trunk
2013-05-06
21:01
In the command-line shell, output the elements of the schema in the order that they appear in the sqlite_master table. This is necessary in some cases to make the schema parsable. check-in: e5b3cd74 user: drh tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to ext/fts3/fts3.c.

3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
....
3643
3644
3645
3646
3647
3648
3649



3650
3651

3652
3653
3654
3655
3656
3657
3658
  rc = sqlite3Fts3InitTerm(db);
  if( rc!=SQLITE_OK ) return rc;
#endif

  rc = sqlite3Fts3InitAux(db);
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3Fts3InitTok(db);
  if( rc!=SQLITE_OK ) return rc;

  sqlite3Fts3SimpleTokenizerModule(&pSimple);
  sqlite3Fts3PorterTokenizerModule(&pPorter);

  /* Allocate and initialize the hash-table used to store tokenizers. */
  pHash = sqlite3_malloc(sizeof(Fts3Hash));
  if( !pHash ){
    rc = SQLITE_NOMEM;
................................................................................
        db, "fts3", &fts3Module, (void *)pHash, hashDestroy
    );
    if( rc==SQLITE_OK ){
      rc = sqlite3_create_module_v2(
          db, "fts4", &fts3Module, (void *)pHash, 0
      );
    }



    return rc;
  }


  /* An error has occurred. Delete the hash table and return the error code. */
  assert( rc!=SQLITE_OK );
  if( pHash ){
    sqlite3Fts3HashClear(pHash);
    sqlite3_free(pHash);
  }







<
<
<







 







>
>
>


>







3587
3588
3589
3590
3591
3592
3593



3594
3595
3596
3597
3598
3599
3600
....
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
  rc = sqlite3Fts3InitTerm(db);
  if( rc!=SQLITE_OK ) return rc;
#endif

  rc = sqlite3Fts3InitAux(db);
  if( rc!=SQLITE_OK ) return rc;




  sqlite3Fts3SimpleTokenizerModule(&pSimple);
  sqlite3Fts3PorterTokenizerModule(&pPorter);

  /* Allocate and initialize the hash-table used to store tokenizers. */
  pHash = sqlite3_malloc(sizeof(Fts3Hash));
  if( !pHash ){
    rc = SQLITE_NOMEM;
................................................................................
        db, "fts3", &fts3Module, (void *)pHash, hashDestroy
    );
    if( rc==SQLITE_OK ){
      rc = sqlite3_create_module_v2(
          db, "fts4", &fts3Module, (void *)pHash, 0
      );
    }
    if( rc==SQLITE_OK ){
      rc = sqlite3Fts3InitTok(db, (void *)pHash);
    }
    return rc;
  }


  /* An error has occurred. Delete the hash table and return the error code. */
  assert( rc!=SQLITE_OK );
  if( pHash ){
    sqlite3Fts3HashClear(pHash);
    sqlite3_free(pHash);
  }

Changes to ext/fts3/fts3Int.h.

546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
int sqlite3Fts3MsrIncrNext(
    Fts3Table *, Fts3MultiSegReader *, sqlite3_int64 *, char **, int *);
int sqlite3Fts3EvalPhrasePoslist(Fts3Cursor *, Fts3Expr *, int iCol, char **); 
int sqlite3Fts3MsrOvfl(Fts3Cursor *, Fts3MultiSegReader *, int *);
int sqlite3Fts3MsrIncrRestart(Fts3MultiSegReader *pCsr);

/* fts3_tokenize_vtab.c */
int sqlite3Fts3InitTok(sqlite3*);

/* fts3_unicode2.c (functions generated by parsing unicode text files) */
#ifdef SQLITE_ENABLE_FTS4_UNICODE61
int sqlite3FtsUnicodeFold(int, int);
int sqlite3FtsUnicodeIsalnum(int);
int sqlite3FtsUnicodeIsdiacritic(int);
#endif

#endif /* !SQLITE_CORE || SQLITE_ENABLE_FTS3 */
#endif /* _FTSINT_H */







|










546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
int sqlite3Fts3MsrIncrNext(
    Fts3Table *, Fts3MultiSegReader *, sqlite3_int64 *, char **, int *);
int sqlite3Fts3EvalPhrasePoslist(Fts3Cursor *, Fts3Expr *, int iCol, char **); 
int sqlite3Fts3MsrOvfl(Fts3Cursor *, Fts3MultiSegReader *, int *);
int sqlite3Fts3MsrIncrRestart(Fts3MultiSegReader *pCsr);

/* fts3_tokenize_vtab.c */
int sqlite3Fts3InitTok(sqlite3*, Fts3Hash *);

/* fts3_unicode2.c (functions generated by parsing unicode text files) */
#ifdef SQLITE_ENABLE_FTS4_UNICODE61
int sqlite3FtsUnicodeFold(int, int);
int sqlite3FtsUnicodeIsalnum(int);
int sqlite3FtsUnicodeIsdiacritic(int);
#endif

#endif /* !SQLITE_CORE || SQLITE_ENABLE_FTS3 */
#endif /* _FTSINT_H */

Changes to ext/fts3/fts3_tokenize_vtab.c.

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
...
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
...
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
...
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
...
453
454
455
456
457
458
459
460
461
462
463
464
};

/*
** Query FTS for the tokenizer implementation named zName.
*/
static int fts3tokQueryTokenizer(
  sqlite3 *db,

  const char *zName,
  const sqlite3_tokenizer_module **pp

){
  int rc;
  sqlite3_stmt *pStmt;
  const char *zSql = "SELECT fts3_tokenizer(?)";


  *pp = 0;
  rc = sqlite3_prepare_v2(db, zSql, -1, &pStmt, 0);
  if( rc!=SQLITE_OK ){

    return rc;
  }

  sqlite3_bind_text(pStmt, 1, zName, -1, SQLITE_STATIC);
  if( SQLITE_ROW==sqlite3_step(pStmt) ){
    if( sqlite3_column_type(pStmt, 0)==SQLITE_BLOB ){
      memcpy((void*)pp, sqlite3_column_blob(pStmt, 0), sizeof(*pp));
    }
  }

  return sqlite3_finalize(pStmt);
}

/*
** The second argument, argv[], is an array of pointers to nul-terminated
** strings. This function makes a copy of the array and strings into a 
** single block of memory. It then dequotes any of the strings that appear
** to be quoted.
................................................................................
**   argv[0]: module name
**   argv[1]: database name 
**   argv[2]: table name
**   argv[3]: first argument (tokenizer name)
*/
static int fts3tokConnectMethod(
  sqlite3 *db,                    /* Database connection */
  void *pUnused,                  /* Unused */
  int argc,                       /* Number of elements in argv array */
  const char * const *argv,       /* xCreate/xConnect argument array */
  sqlite3_vtab **ppVtab,          /* OUT: New sqlite3_vtab object */
  char **pzErr                    /* OUT: sqlite3_malloc'd error message */
){
  Fts3tokTable *pTab;
  const sqlite3_tokenizer_module *pMod = 0;
  sqlite3_tokenizer *pTok = 0;
  int rc;
  char **azDequote = 0;
  int nDequote;
  UNUSED_PARAMETER(pUnused);

  rc = sqlite3_declare_vtab(db, FTS3_TOK_SCHEMA);
  if( rc!=SQLITE_OK ) return rc;

  nDequote = argc-3;
  rc = fts3tokDequoteArray(nDequote, &argv[3], &azDequote);

................................................................................
  if( rc==SQLITE_OK ){
    const char *zModule;
    if( nDequote<1 ){
      zModule = "simple";
    }else{
      zModule = azDequote[0];
    }
    rc = fts3tokQueryTokenizer(db, zModule, &pMod);
  }

  if( rc!=SQLITE_OK ){
    *pzErr = sqlite3_mprintf("%s", sqlite3_errmsg(db));
  }else if( pMod==0 ){
    rc = SQLITE_ERROR;
  }else{
    const char * const *azArg = (const char * const *)&azDequote[1];
    rc = pMod->xCreate((nDequote>1 ? nDequote-1 : 0), azArg, &pTok);
  }

  if( rc==SQLITE_OK ){
    pTab = (Fts3tokTable *)sqlite3_malloc(sizeof(Fts3tokTable));
    if( pTab==0 ){
................................................................................
  return SQLITE_OK;
}

/*
** Register the fts3tok module with database connection db. Return SQLITE_OK
** if successful or an error code if sqlite3_create_module() fails.
*/
int sqlite3Fts3InitTok(sqlite3 *db){
  static const sqlite3_module fts3tok_module = {
     0,                           /* iVersion      */
     fts3tokConnectMethod,        /* xCreate       */
     fts3tokConnectMethod,        /* xConnect      */
     fts3tokBestIndexMethod,      /* xBestIndex    */
     fts3tokDisconnectMethod,     /* xDisconnect   */
     fts3tokDisconnectMethod,     /* xDestroy      */
................................................................................
     0,                           /* xRename       */
     0,                           /* xSavepoint    */
     0,                           /* xRelease      */
     0                            /* xRollbackTo   */
  };
  int rc;                         /* Return code */

  rc = sqlite3_create_module(db, "fts3tokenize", &fts3tok_module, 0);
  return rc;
}

#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */







>

|
>

<
<
|
>

<
|
|
>
|


<
|
|
<
<
<
<
<







 







|











<







 







|


|
<
<
|
<







 







|







 







|




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
...
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172

173
174
175
176
177
178
179
...
180
181
182
183
184
185
186
187
188
189
190


191

192
193
194
195
196
197
198
...
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
...
444
445
446
447
448
449
450
451
452
453
454
455
};

/*
** Query FTS for the tokenizer implementation named zName.
*/
static int fts3tokQueryTokenizer(
  sqlite3 *db,
  Fts3Hash *pHash,
  const char *zName,
  const sqlite3_tokenizer_module **pp,
  char **pzErr
){


  sqlite3_tokenizer_module *p;
  int nName = strlen(zName);


  p = (sqlite3_tokenizer_module *)sqlite3Fts3HashFind(pHash, zName, nName+1);
  if( !p ){
    *pzErr = sqlite3_mprintf("unknown tokenizer: %s", zName);
    return SQLITE_ERROR;
  }


  *pp = p;
  return SQLITE_OK;





}

/*
** The second argument, argv[], is an array of pointers to nul-terminated
** strings. This function makes a copy of the array and strings into a 
** single block of memory. It then dequotes any of the strings that appear
** to be quoted.
................................................................................
**   argv[0]: module name
**   argv[1]: database name 
**   argv[2]: table name
**   argv[3]: first argument (tokenizer name)
*/
static int fts3tokConnectMethod(
  sqlite3 *db,                    /* Database connection */
  void *pHash,                    /* Hash table of tokenizers */
  int argc,                       /* Number of elements in argv array */
  const char * const *argv,       /* xCreate/xConnect argument array */
  sqlite3_vtab **ppVtab,          /* OUT: New sqlite3_vtab object */
  char **pzErr                    /* OUT: sqlite3_malloc'd error message */
){
  Fts3tokTable *pTab;
  const sqlite3_tokenizer_module *pMod = 0;
  sqlite3_tokenizer *pTok = 0;
  int rc;
  char **azDequote = 0;
  int nDequote;


  rc = sqlite3_declare_vtab(db, FTS3_TOK_SCHEMA);
  if( rc!=SQLITE_OK ) return rc;

  nDequote = argc-3;
  rc = fts3tokDequoteArray(nDequote, &argv[3], &azDequote);

................................................................................
  if( rc==SQLITE_OK ){
    const char *zModule;
    if( nDequote<1 ){
      zModule = "simple";
    }else{
      zModule = azDequote[0];
    }
    rc = fts3tokQueryTokenizer(db, (Fts3Hash *)pHash, zModule, &pMod, pzErr);
  }

  assert( (rc==SQLITE_OK)==(pMod!=0) );


  if( rc==SQLITE_OK ){

    const char * const *azArg = (const char * const *)&azDequote[1];
    rc = pMod->xCreate((nDequote>1 ? nDequote-1 : 0), azArg, &pTok);
  }

  if( rc==SQLITE_OK ){
    pTab = (Fts3tokTable *)sqlite3_malloc(sizeof(Fts3tokTable));
    if( pTab==0 ){
................................................................................
  return SQLITE_OK;
}

/*
** Register the fts3tok module with database connection db. Return SQLITE_OK
** if successful or an error code if sqlite3_create_module() fails.
*/
int sqlite3Fts3InitTok(sqlite3 *db, Fts3Hash *pHash){
  static const sqlite3_module fts3tok_module = {
     0,                           /* iVersion      */
     fts3tokConnectMethod,        /* xCreate       */
     fts3tokConnectMethod,        /* xConnect      */
     fts3tokBestIndexMethod,      /* xBestIndex    */
     fts3tokDisconnectMethod,     /* xDisconnect   */
     fts3tokDisconnectMethod,     /* xDestroy      */
................................................................................
     0,                           /* xRename       */
     0,                           /* xSavepoint    */
     0,                           /* xRelease      */
     0                            /* xRollbackTo   */
  };
  int rc;                         /* Return code */

  rc = sqlite3_create_module(db, "fts3tokenize", &fts3tok_module, (void*)pHash);
  return rc;
}

#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */

Changes to test/fts3tok1.test.

102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
#-------------------------------------------------------------------------
# Error cases.
#
do_catchsql_test 2.0 {
  CREATE VIRTUAL TABLE tX USING fts3tokenize(nosuchtokenizer);
} {1 {unknown tokenizer: nosuchtokenizer}}

proc fts3_tokenizer {args} { return 1 }
db function fts3_tokenizer -argcount 1 fts3_tokenizer
do_catchsql_test 2.1 {
  CREATE VIRTUAL TABLE tX USING fts3tokenize(simple);
} {1 {vtable constructor failed: tX}}

db close
sqlite3 db test.db
do_catchsql_test 2.2 {
  CREATE VIRTUAL TABLE t4 USING fts3tokenize;
  SELECT * FROM t4;
} {1 {SQL logic error or missing database}}


finish_test









<
<

<
<
<
<
<
<








102
103
104
105
106
107
108


109






110
111
112
113
114
115
116
117
#-------------------------------------------------------------------------
# Error cases.
#
do_catchsql_test 2.0 {
  CREATE VIRTUAL TABLE tX USING fts3tokenize(nosuchtokenizer);
} {1 {unknown tokenizer: nosuchtokenizer}}



do_catchsql_test 2.1 {






  CREATE VIRTUAL TABLE t4 USING fts3tokenize;
  SELECT * FROM t4;
} {1 {SQL logic error or missing database}}


finish_test