/ Check-in [fbc53da8]
Login

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

Overview
Comment:Out-of-memory cleanup in tokenizers. Handle NULL return from malloc/calloc/realloc appropriately, and use sizeof(var) instead of sizeof(type) to make certain that we don't get a mismatch between them as the code rots. (CVS 3693)
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1:fbc53da8c645935c74e49af2ab2cf447dc72ba4e
User & Date: shess 2007-03-16 18:30:54
Context
2007-03-17
07:22
Combine the two very similar definitions of (crashsql) in the test scripts. (CVS 3694) check-in: eaf434d5 user: danielk1977 tags: trunk
2007-03-16
18:30
Out-of-memory cleanup in tokenizers. Handle NULL return from malloc/calloc/realloc appropriately, and use sizeof(var) instead of sizeof(type) to make certain that we don't get a mismatch between them as the code rots. (CVS 3693) check-in: fbc53da8 user: shess tags: trunk
2007-03-15
15:35
Save an if(...) clause made redundant by (3691). (CVS 3692) check-in: 8153edf8 user: danielk1977 tags: trunk
Changes
Hide Diffs Unified Diffs Show Whitespace Changes Patch

Changes to ext/fts1/fts1_porter.c.

66
67
68
69
70
71
72
73


74
75
76
77
78
79
80
..
92
93
94
95
96
97
98
99


100
101
102
103
104
105
106
...
601
602
603
604
605
606
607

608
609
610
611
612
613
614
** Create a new tokenizer instance.
*/
static int porterCreate(
  int argc, const char * const *argv,
  sqlite3_tokenizer **ppTokenizer
){
  porter_tokenizer *t;
  t = (porter_tokenizer *) calloc(sizeof(porter_tokenizer), 1);


  *ppTokenizer = &t->base;
  return SQLITE_OK;
}

/*
** Destroy a tokenizer
*/
................................................................................
static int porterOpen(
  sqlite3_tokenizer *pTokenizer,         /* The tokenizer */
  const char *zInput, int nInput,        /* String to be tokenized */
  sqlite3_tokenizer_cursor **ppCursor    /* OUT: Tokenization cursor */
){
  porter_tokenizer_cursor *c;

  c = (porter_tokenizer_cursor *) malloc(sizeof(porter_tokenizer_cursor));


  c->zInput = zInput;
  if( zInput==0 ){
    c->nInput = 0;
  }else if( nInput<0 ){
    c->nInput = (int)strlen(zInput);
  }else{
    c->nInput = nInput;
................................................................................
    }

    if( c->iOffset>iStartOffset ){
      int n = c->iOffset-iStartOffset;
      if( n>c->nAllocated ){
        c->nAllocated = n+20;
        c->zToken = realloc(c->zToken, c->nAllocated);

      }
      porter_stemmer(&z[iStartOffset], n, c->zToken, pnBytes);
      *pzToken = c->zToken;
      *piStartOffset = iStartOffset;
      *piEndOffset = c->iOffset;
      *piPosition = c->iToken++;
      return SQLITE_OK;







|
>
>







 







|
>
>







 







>







66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
..
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
...
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
** Create a new tokenizer instance.
*/
static int porterCreate(
  int argc, const char * const *argv,
  sqlite3_tokenizer **ppTokenizer
){
  porter_tokenizer *t;
  t = (porter_tokenizer *) calloc(sizeof(*t), 1);
  if( t==NULL ) return SQLITE_NOMEM;

  *ppTokenizer = &t->base;
  return SQLITE_OK;
}

/*
** Destroy a tokenizer
*/
................................................................................
static int porterOpen(
  sqlite3_tokenizer *pTokenizer,         /* The tokenizer */
  const char *zInput, int nInput,        /* String to be tokenized */
  sqlite3_tokenizer_cursor **ppCursor    /* OUT: Tokenization cursor */
){
  porter_tokenizer_cursor *c;

  c = (porter_tokenizer_cursor *) malloc(sizeof(*c));
  if( c==NULL ) return SQLITE_NOMEM;

  c->zInput = zInput;
  if( zInput==0 ){
    c->nInput = 0;
  }else if( nInput<0 ){
    c->nInput = (int)strlen(zInput);
  }else{
    c->nInput = nInput;
................................................................................
    }

    if( c->iOffset>iStartOffset ){
      int n = c->iOffset-iStartOffset;
      if( n>c->nAllocated ){
        c->nAllocated = n+20;
        c->zToken = realloc(c->zToken, c->nAllocated);
        if( c->zToken==NULL ) return SQLITE_NOMEM;
      }
      porter_stemmer(&z[iStartOffset], n, c->zToken, pnBytes);
      *pzToken = c->zToken;
      *piStartOffset = iStartOffset;
      *piEndOffset = c->iOffset;
      *piPosition = c->iToken++;
      return SQLITE_OK;

Changes to ext/fts1/fts1_tokenizer1.c.

57
58
59
60
61
62
63
64


65
66
67
68
69
70
71
...
107
108
109
110
111
112
113
114


115
116
117
118
119
120
121
...
171
172
173
174
175
176
177

178
179
180
181
182
183
184
*/
static int simpleCreate(
  int argc, const char * const *argv,
  sqlite3_tokenizer **ppTokenizer
){
  simple_tokenizer *t;

  t = (simple_tokenizer *) calloc(sizeof(simple_tokenizer), 1);


  /* TODO(shess) Delimiters need to remain the same from run to run,
  ** else we need to reindex.  One solution would be a meta-table to
  ** track such information in the database, then we'd only want this
  ** information on the initial create.
  */
  if( argc>1 ){
    int i, n = strlen(argv[1]);
................................................................................
static int simpleOpen(
  sqlite3_tokenizer *pTokenizer,         /* The tokenizer */
  const char *pInput, int nBytes,        /* String to be tokenized */
  sqlite3_tokenizer_cursor **ppCursor    /* OUT: Tokenization cursor */
){
  simple_tokenizer_cursor *c;

  c = (simple_tokenizer_cursor *) malloc(sizeof(simple_tokenizer_cursor));


  c->pInput = pInput;
  if( pInput==0 ){
    c->nBytes = 0;
  }else if( nBytes<0 ){
    c->nBytes = (int)strlen(pInput);
  }else{
    c->nBytes = nBytes;
................................................................................
    }

    if( c->iOffset>iStartOffset ){
      int i, n = c->iOffset-iStartOffset;
      if( n>c->nTokenAllocated ){
        c->nTokenAllocated = n+20;
        c->pToken = realloc(c->pToken, c->nTokenAllocated);

      }
      for(i=0; i<n; i++){
        /* TODO(shess) This needs expansion to handle UTF-8
        ** case-insensitivity.
        */
        unsigned char ch = p[iStartOffset+i];
        c->pToken[i] = ch<0x80 ? tolower(ch) : ch;







|
>
>







 







|
>
>







 







>







57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
...
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
...
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
*/
static int simpleCreate(
  int argc, const char * const *argv,
  sqlite3_tokenizer **ppTokenizer
){
  simple_tokenizer *t;

  t = (simple_tokenizer *) calloc(sizeof(*t), 1);
  if( t==NULL ) return SQLITE_NOMEM;

  /* TODO(shess) Delimiters need to remain the same from run to run,
  ** else we need to reindex.  One solution would be a meta-table to
  ** track such information in the database, then we'd only want this
  ** information on the initial create.
  */
  if( argc>1 ){
    int i, n = strlen(argv[1]);
................................................................................
static int simpleOpen(
  sqlite3_tokenizer *pTokenizer,         /* The tokenizer */
  const char *pInput, int nBytes,        /* String to be tokenized */
  sqlite3_tokenizer_cursor **ppCursor    /* OUT: Tokenization cursor */
){
  simple_tokenizer_cursor *c;

  c = (simple_tokenizer_cursor *) malloc(sizeof(*c));
  if( c==NULL ) return SQLITE_NOMEM;

  c->pInput = pInput;
  if( pInput==0 ){
    c->nBytes = 0;
  }else if( nBytes<0 ){
    c->nBytes = (int)strlen(pInput);
  }else{
    c->nBytes = nBytes;
................................................................................
    }

    if( c->iOffset>iStartOffset ){
      int i, n = c->iOffset-iStartOffset;
      if( n>c->nTokenAllocated ){
        c->nTokenAllocated = n+20;
        c->pToken = realloc(c->pToken, c->nTokenAllocated);
        if( c->pToken==NULL ) return SQLITE_NOMEM;
      }
      for(i=0; i<n; i++){
        /* TODO(shess) This needs expansion to handle UTF-8
        ** case-insensitivity.
        */
        unsigned char ch = p[iStartOffset+i];
        c->pToken[i] = ch<0x80 ? tolower(ch) : ch;

Changes to ext/fts2/fts2_porter.c.

66
67
68
69
70
71
72
73


74
75
76
77
78
79
80
..
92
93
94
95
96
97
98
99


100
101
102
103
104
105
106
...
601
602
603
604
605
606
607

608
609
610
611
612
613
614
** Create a new tokenizer instance.
*/
static int porterCreate(
  int argc, const char * const *argv,
  sqlite3_tokenizer **ppTokenizer
){
  porter_tokenizer *t;
  t = (porter_tokenizer *) calloc(sizeof(porter_tokenizer), 1);


  *ppTokenizer = &t->base;
  return SQLITE_OK;
}

/*
** Destroy a tokenizer
*/
................................................................................
static int porterOpen(
  sqlite3_tokenizer *pTokenizer,         /* The tokenizer */
  const char *zInput, int nInput,        /* String to be tokenized */
  sqlite3_tokenizer_cursor **ppCursor    /* OUT: Tokenization cursor */
){
  porter_tokenizer_cursor *c;

  c = (porter_tokenizer_cursor *) malloc(sizeof(porter_tokenizer_cursor));


  c->zInput = zInput;
  if( zInput==0 ){
    c->nInput = 0;
  }else if( nInput<0 ){
    c->nInput = (int)strlen(zInput);
  }else{
    c->nInput = nInput;
................................................................................
    }

    if( c->iOffset>iStartOffset ){
      int n = c->iOffset-iStartOffset;
      if( n>c->nAllocated ){
        c->nAllocated = n+20;
        c->zToken = realloc(c->zToken, c->nAllocated);

      }
      porter_stemmer(&z[iStartOffset], n, c->zToken, pnBytes);
      *pzToken = c->zToken;
      *piStartOffset = iStartOffset;
      *piEndOffset = c->iOffset;
      *piPosition = c->iToken++;
      return SQLITE_OK;







|
>
>







 







|
>
>







 







>







66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
..
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
...
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
** Create a new tokenizer instance.
*/
static int porterCreate(
  int argc, const char * const *argv,
  sqlite3_tokenizer **ppTokenizer
){
  porter_tokenizer *t;
  t = (porter_tokenizer *) calloc(sizeof(*t), 1);
  if( t==NULL ) return SQLITE_NOMEM;

  *ppTokenizer = &t->base;
  return SQLITE_OK;
}

/*
** Destroy a tokenizer
*/
................................................................................
static int porterOpen(
  sqlite3_tokenizer *pTokenizer,         /* The tokenizer */
  const char *zInput, int nInput,        /* String to be tokenized */
  sqlite3_tokenizer_cursor **ppCursor    /* OUT: Tokenization cursor */
){
  porter_tokenizer_cursor *c;

  c = (porter_tokenizer_cursor *) malloc(sizeof(*c));
  if( c==NULL ) return SQLITE_NOMEM;

  c->zInput = zInput;
  if( zInput==0 ){
    c->nInput = 0;
  }else if( nInput<0 ){
    c->nInput = (int)strlen(zInput);
  }else{
    c->nInput = nInput;
................................................................................
    }

    if( c->iOffset>iStartOffset ){
      int n = c->iOffset-iStartOffset;
      if( n>c->nAllocated ){
        c->nAllocated = n+20;
        c->zToken = realloc(c->zToken, c->nAllocated);
        if( c->zToken==NULL ) return SQLITE_NOMEM;
      }
      porter_stemmer(&z[iStartOffset], n, c->zToken, pnBytes);
      *pzToken = c->zToken;
      *piStartOffset = iStartOffset;
      *piEndOffset = c->iOffset;
      *piPosition = c->iToken++;
      return SQLITE_OK;

Changes to ext/fts2/fts2_tokenizer1.c.

57
58
59
60
61
62
63
64


65
66
67
68
69
70
71
...
107
108
109
110
111
112
113
114


115
116
117
118
119
120
121
...
171
172
173
174
175
176
177

178
179
180
181
182
183
184
*/
static int simpleCreate(
  int argc, const char * const *argv,
  sqlite3_tokenizer **ppTokenizer
){
  simple_tokenizer *t;

  t = (simple_tokenizer *) calloc(sizeof(simple_tokenizer), 1);


  /* TODO(shess) Delimiters need to remain the same from run to run,
  ** else we need to reindex.  One solution would be a meta-table to
  ** track such information in the database, then we'd only want this
  ** information on the initial create.
  */
  if( argc>1 ){
    int i, n = strlen(argv[1]);
................................................................................
static int simpleOpen(
  sqlite3_tokenizer *pTokenizer,         /* The tokenizer */
  const char *pInput, int nBytes,        /* String to be tokenized */
  sqlite3_tokenizer_cursor **ppCursor    /* OUT: Tokenization cursor */
){
  simple_tokenizer_cursor *c;

  c = (simple_tokenizer_cursor *) malloc(sizeof(simple_tokenizer_cursor));


  c->pInput = pInput;
  if( pInput==0 ){
    c->nBytes = 0;
  }else if( nBytes<0 ){
    c->nBytes = (int)strlen(pInput);
  }else{
    c->nBytes = nBytes;
................................................................................
    }

    if( c->iOffset>iStartOffset ){
      int i, n = c->iOffset-iStartOffset;
      if( n>c->nTokenAllocated ){
        c->nTokenAllocated = n+20;
        c->pToken = realloc(c->pToken, c->nTokenAllocated);

      }
      for(i=0; i<n; i++){
        /* TODO(shess) This needs expansion to handle UTF-8
        ** case-insensitivity.
        */
        unsigned char ch = p[iStartOffset+i];
        c->pToken[i] = ch<0x80 ? tolower(ch) : ch;







|
>
>







 







|
>
>







 







>







57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
...
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
...
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
*/
static int simpleCreate(
  int argc, const char * const *argv,
  sqlite3_tokenizer **ppTokenizer
){
  simple_tokenizer *t;

  t = (simple_tokenizer *) calloc(sizeof(*t), 1);
  if( t==NULL ) return SQLITE_NOMEM;

  /* TODO(shess) Delimiters need to remain the same from run to run,
  ** else we need to reindex.  One solution would be a meta-table to
  ** track such information in the database, then we'd only want this
  ** information on the initial create.
  */
  if( argc>1 ){
    int i, n = strlen(argv[1]);
................................................................................
static int simpleOpen(
  sqlite3_tokenizer *pTokenizer,         /* The tokenizer */
  const char *pInput, int nBytes,        /* String to be tokenized */
  sqlite3_tokenizer_cursor **ppCursor    /* OUT: Tokenization cursor */
){
  simple_tokenizer_cursor *c;

  c = (simple_tokenizer_cursor *) malloc(sizeof(*c));
  if( c==NULL ) return SQLITE_NOMEM;

  c->pInput = pInput;
  if( pInput==0 ){
    c->nBytes = 0;
  }else if( nBytes<0 ){
    c->nBytes = (int)strlen(pInput);
  }else{
    c->nBytes = nBytes;
................................................................................
    }

    if( c->iOffset>iStartOffset ){
      int i, n = c->iOffset-iStartOffset;
      if( n>c->nTokenAllocated ){
        c->nTokenAllocated = n+20;
        c->pToken = realloc(c->pToken, c->nTokenAllocated);
        if( c->pToken==NULL ) return SQLITE_NOMEM;
      }
      for(i=0; i<n; i++){
        /* TODO(shess) This needs expansion to handle UTF-8
        ** case-insensitivity.
        */
        unsigned char ch = p[iStartOffset+i];
        c->pToken[i] = ch<0x80 ? tolower(ch) : ch;