SQLite

Check-in [bc3f7900d5]
Login

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

Overview
Comment:Handle parser stack overflow when parsing fts5 query expressions. Fix some compiler warnings in fts5 code.
Downloads: Tarball | ZIP archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: bc3f7900d5a06829d123814a5ac7b951bcfc1560
User & Date: dan 2016-02-11 17:01:32.344
Context
2016-02-11
18:08
Avoid a buffer overread when reading a corrupt fts5 structure record. (check-in: facbc424e5 user: dan tags: trunk)
17:01
Handle parser stack overflow when parsing fts5 query expressions. Fix some compiler warnings in fts5 code. (check-in: bc3f7900d5 user: dan tags: trunk)
15:37
Add JSON1 and FTS5 to the set of extensions subject to close compiler warning analysis. Fix some warnings in each. More (harmless) warnings still exist in FTS5. (check-in: cfe2eb88b5 user: drh tags: trunk)
Changes
Unified Diff Ignore Whitespace Patch
Changes to ext/fts5/fts5Int.h.
81
82
83
84
85
86
87




88
89
90
91
92
93
94
#endif

/* Mark a function parameter as unused, to suppress nuisance compiler
** warnings. */
#ifndef UNUSED_PARAM
# define UNUSED_PARAM(X)  (void)(X)
#endif





typedef struct Fts5Global Fts5Global;
typedef struct Fts5Colset Fts5Colset;

/* If a NEAR() clump or phrase may only match a specific set of columns, 
** then an object of the following type is used to record the set of columns.
** Each entry in the aiCol[] array is a column that may be matched.







>
>
>
>







81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
#endif

/* Mark a function parameter as unused, to suppress nuisance compiler
** warnings. */
#ifndef UNUSED_PARAM
# define UNUSED_PARAM(X)  (void)(X)
#endif

#ifndef UNUSED_PARAM2
# define UNUSED_PARAM2(X, Y)  (void)(X), (void)(Y)
#endif

typedef struct Fts5Global Fts5Global;
typedef struct Fts5Colset Fts5Colset;

/* If a NEAR() clump or phrase may only match a specific set of columns, 
** then an object of the following type is used to record the set of columns.
** Each entry in the aiCol[] array is a column that may be matched.
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
Fts5PoslistPopulator *sqlite3Fts5ExprClearPoslists(Fts5Expr*, int);
int sqlite3Fts5ExprPopulatePoslists(
    Fts5Config*, Fts5Expr*, Fts5PoslistPopulator*, int, const char*, int
);
void sqlite3Fts5ExprCheckPoslists(Fts5Expr*, i64);
void sqlite3Fts5ExprClearEof(Fts5Expr*);

int sqlite3Fts5ExprClonePhrase(Fts5Config*, Fts5Expr*, int, Fts5Expr**);

int sqlite3Fts5ExprPhraseCollist(Fts5Expr *, int, const u8 **, int *);

/*******************************************
** The fts5_expr.c API above this point is used by the other hand-written
** C code in this module. The interfaces below this point are called by
** the parser code in fts5parse.y.  */







|







676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
Fts5PoslistPopulator *sqlite3Fts5ExprClearPoslists(Fts5Expr*, int);
int sqlite3Fts5ExprPopulatePoslists(
    Fts5Config*, Fts5Expr*, Fts5PoslistPopulator*, int, const char*, int
);
void sqlite3Fts5ExprCheckPoslists(Fts5Expr*, i64);
void sqlite3Fts5ExprClearEof(Fts5Expr*);

int sqlite3Fts5ExprClonePhrase(Fts5Expr*, int, Fts5Expr**);

int sqlite3Fts5ExprPhraseCollist(Fts5Expr *, int, const u8 **, int *);

/*******************************************
** The fts5_expr.c API above this point is used by the other hand-written
** C code in this module. The interfaces below this point are called by
** the parser code in fts5parse.y.  */
Changes to ext/fts5/fts5_aux.c.
153
154
155
156
157
158
159


160
161
162
163
164
165
166
  int nToken,                     /* Size of token in bytes */
  int iStartOff,                  /* Start offset of token */
  int iEndOff                     /* End offset of token */
){
  HighlightContext *p = (HighlightContext*)pContext;
  int rc = SQLITE_OK;
  int iPos;



  if( tflags & FTS5_TOKEN_COLOCATED ) return SQLITE_OK;
  iPos = p->iPos++;

  if( p->iRangeEnd>0 ){
    if( iPos<p->iRangeStart || iPos>p->iRangeEnd ) return SQLITE_OK;
    if( p->iRangeStart && iPos==p->iRangeStart ) p->iOff = iStartOff;







>
>







153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
  int nToken,                     /* Size of token in bytes */
  int iStartOff,                  /* Start offset of token */
  int iEndOff                     /* End offset of token */
){
  HighlightContext *p = (HighlightContext*)pContext;
  int rc = SQLITE_OK;
  int iPos;

  UNUSED_PARAM2(pToken, nToken);

  if( tflags & FTS5_TOKEN_COLOCATED ) return SQLITE_OK;
  iPos = p->iPos++;

  if( p->iRangeEnd>0 ){
    if( iPos<p->iRangeStart || iPos>p->iRangeEnd ) return SQLITE_OK;
    if( p->iRangeStart && iPos==p->iRangeStart ) p->iOff = iStartOff;
387
388
389
390
391
392
393

394
395
396
397
398
399
400
*/
static int fts5CountCb(
  const Fts5ExtensionApi *pApi, 
  Fts5Context *pFts,
  void *pUserData                 /* Pointer to sqlite3_int64 variable */
){
  sqlite3_int64 *pn = (sqlite3_int64*)pUserData;

  (*pn)++;
  return SQLITE_OK;
}

/*
** Set *ppData to point to the Fts5Bm25Data object for the current query. 
** If the object has not already been allocated, allocate and populate it







>







389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
*/
static int fts5CountCb(
  const Fts5ExtensionApi *pApi, 
  Fts5Context *pFts,
  void *pUserData                 /* Pointer to sqlite3_int64 variable */
){
  sqlite3_int64 *pn = (sqlite3_int64*)pUserData;
  UNUSED_PARAM2(pApi, pFts);
  (*pn)++;
  return SQLITE_OK;
}

/*
** Set *ppData to point to the Fts5Bm25Data object for the current query. 
** If the object has not already been allocated, allocate and populate it
Changes to ext/fts5/fts5_expr.c.
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
}

/*
** Argument pTerm must be a synonym iterator.
*/
static int fts5ExprSynonymList(
  Fts5ExprTerm *pTerm, 
  int bCollist, 
  Fts5Colset *pColset,
  i64 iRowid,
  Fts5Buffer *pBuf,               /* Use this buffer for space if required */
  u8 **pa, int *pn
){
  Fts5PoslistReader aStatic[4];
  Fts5PoslistReader *aIter = aStatic;
  int nIter = 0;







<
<







316
317
318
319
320
321
322


323
324
325
326
327
328
329
}

/*
** Argument pTerm must be a synonym iterator.
*/
static int fts5ExprSynonymList(
  Fts5ExprTerm *pTerm, 


  i64 iRowid,
  Fts5Buffer *pBuf,               /* Use this buffer for space if required */
  u8 **pa, int *pn
){
  Fts5PoslistReader aStatic[4];
  Fts5PoslistReader *aIter = aStatic;
  int nIter = 0;
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
**
** SQLITE_OK is returned if an error occurs, or an SQLite error code 
** otherwise. It is not considered an error code if the current rowid is 
** not a match.
*/
static int fts5ExprPhraseIsMatch(
  Fts5ExprNode *pNode,            /* Node pPhrase belongs to */
  Fts5Colset *pColset,            /* Restrict matches to these columns */
  Fts5ExprPhrase *pPhrase,        /* Phrase object to initialize */
  int *pbMatch                    /* OUT: Set to true if really a match */
){
  Fts5PoslistWriter writer = {0};
  Fts5PoslistReader aStatic[4];
  Fts5PoslistReader *aIter = aStatic;
  int i;







<







399
400
401
402
403
404
405

406
407
408
409
410
411
412
**
** SQLITE_OK is returned if an error occurs, or an SQLite error code 
** otherwise. It is not considered an error code if the current rowid is 
** not a match.
*/
static int fts5ExprPhraseIsMatch(
  Fts5ExprNode *pNode,            /* Node pPhrase belongs to */

  Fts5ExprPhrase *pPhrase,        /* Phrase object to initialize */
  int *pbMatch                    /* OUT: Set to true if really a match */
){
  Fts5PoslistWriter writer = {0};
  Fts5PoslistReader aStatic[4];
  Fts5PoslistReader *aIter = aStatic;
  int i;
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
  for(i=0; i<pPhrase->nTerm; i++){
    Fts5ExprTerm *pTerm = &pPhrase->aTerm[i];
    int n = 0;
    int bFlag = 0;
    u8 *a = 0;
    if( pTerm->pSynonym ){
      Fts5Buffer buf = {0, 0, 0};
      rc = fts5ExprSynonymList(
          pTerm, 0, pColset, pNode->iRowid, &buf, &a, &n
      );
      if( rc ){
        sqlite3_free(a);
        goto ismatch_out;
      }
      if( a==buf.p ) bFlag = 1;
    }else{
      a = (u8*)pTerm->pIter->pData;







|
<
<







427
428
429
430
431
432
433
434


435
436
437
438
439
440
441
  for(i=0; i<pPhrase->nTerm; i++){
    Fts5ExprTerm *pTerm = &pPhrase->aTerm[i];
    int n = 0;
    int bFlag = 0;
    u8 *a = 0;
    if( pTerm->pSynonym ){
      Fts5Buffer buf = {0, 0, 0};
      rc = fts5ExprSynonymList(pTerm, pNode->iRowid, &buf, &a, &n);


      if( rc ){
        sqlite3_free(a);
        goto ismatch_out;
      }
      if( a==buf.p ) bFlag = 1;
    }else{
      a = (u8*)pTerm->pIter->pData;
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
    /* Check that each phrase in the nearset matches the current row.
    ** Populate the pPhrase->poslist buffers at the same time. If any
    ** phrase is not a match, break out of the loop early.  */
    for(i=0; rc==SQLITE_OK && i<pNear->nPhrase; i++){
      Fts5ExprPhrase *pPhrase = pNear->apPhrase[i];
      if( pPhrase->nTerm>1 || pPhrase->aTerm[0].pSynonym || pNear->pColset ){
        int bMatch = 0;
        rc = fts5ExprPhraseIsMatch(pNode, pNear->pColset, pPhrase, &bMatch);
        if( bMatch==0 ) break;
      }else{
        Fts5IndexIter *pIter = pPhrase->aTerm[0].pIter;
        fts5BufferSet(&rc, &pPhrase->poslist, pIter->nData, pIter->pData);
      }
    }








|







718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
    /* Check that each phrase in the nearset matches the current row.
    ** Populate the pPhrase->poslist buffers at the same time. If any
    ** phrase is not a match, break out of the loop early.  */
    for(i=0; rc==SQLITE_OK && i<pNear->nPhrase; i++){
      Fts5ExprPhrase *pPhrase = pNear->apPhrase[i];
      if( pPhrase->nTerm>1 || pPhrase->aTerm[0].pSynonym || pNear->pColset ){
        int bMatch = 0;
        rc = fts5ExprPhraseIsMatch(pNode, pPhrase, &bMatch);
        if( bMatch==0 ) break;
      }else{
        Fts5IndexIter *pIter = pPhrase->aTerm[0].pIter;
        fts5BufferSet(&rc, &pPhrase->poslist, pIter->nData, pIter->pData);
      }
    }

1471
1472
1473
1474
1475
1476
1477


1478
1479
1480
1481
1482
1483
1484
  int iUnused1,                   /* Start offset of token */
  int iUnused2                    /* End offset of token */
){
  int rc = SQLITE_OK;
  const int SZALLOC = 8;
  TokenCtx *pCtx = (TokenCtx*)pContext;
  Fts5ExprPhrase *pPhrase = pCtx->pPhrase;



  /* If an error has already occurred, this is a no-op */
  if( pCtx->rc!=SQLITE_OK ) return pCtx->rc;

  assert( pPhrase==0 || pPhrase->nTerm>0 );
  if( pPhrase && (tflags & FTS5_TOKEN_COLOCATED) ){
    Fts5ExprTerm *pSyn;







>
>







1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
  int iUnused1,                   /* Start offset of token */
  int iUnused2                    /* End offset of token */
){
  int rc = SQLITE_OK;
  const int SZALLOC = 8;
  TokenCtx *pCtx = (TokenCtx*)pContext;
  Fts5ExprPhrase *pPhrase = pCtx->pPhrase;

  UNUSED_PARAM2(iUnused1, iUnused2);

  /* If an error has already occurred, this is a no-op */
  if( pCtx->rc!=SQLITE_OK ) return pCtx->rc;

  assert( pPhrase==0 || pPhrase->nTerm>0 );
  if( pPhrase && (tflags & FTS5_TOKEN_COLOCATED) ){
    Fts5ExprTerm *pSyn;
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
}

/*
** Create a new FTS5 expression by cloning phrase iPhrase of the
** expression passed as the second argument.
*/
int sqlite3Fts5ExprClonePhrase(
  Fts5Config *pConfig,
  Fts5Expr *pExpr, 
  int iPhrase, 
  Fts5Expr **ppNew
){
  int rc = SQLITE_OK;             /* Return code */
  Fts5ExprPhrase *pOrig;          /* The phrase extracted from pExpr */
  int i;                          /* Used to iterate through phrase terms */

  Fts5Expr *pNew = 0;             /* Expression to return via *ppNew */

  TokenCtx sCtx = {0,0};          /* Context object for fts5ParseTokenize */


  pOrig = pExpr->apExprPhrase[iPhrase];

  pNew = (Fts5Expr*)sqlite3Fts5MallocZero(&rc, sizeof(Fts5Expr));
  if( rc==SQLITE_OK ){
    pNew->apExprPhrase = (Fts5ExprPhrase**)sqlite3Fts5MallocZero(&rc, 
        sizeof(Fts5ExprPhrase*));
  }
  if( rc==SQLITE_OK ){
    pNew->pRoot = (Fts5ExprNode*)sqlite3Fts5MallocZero(&rc, 







<







<

<


<

<







1604
1605
1606
1607
1608
1609
1610

1611
1612
1613
1614
1615
1616
1617

1618

1619
1620

1621

1622
1623
1624
1625
1626
1627
1628
}

/*
** Create a new FTS5 expression by cloning phrase iPhrase of the
** expression passed as the second argument.
*/
int sqlite3Fts5ExprClonePhrase(

  Fts5Expr *pExpr, 
  int iPhrase, 
  Fts5Expr **ppNew
){
  int rc = SQLITE_OK;             /* Return code */
  Fts5ExprPhrase *pOrig;          /* The phrase extracted from pExpr */
  int i;                          /* Used to iterate through phrase terms */

  Fts5Expr *pNew = 0;             /* Expression to return via *ppNew */

  TokenCtx sCtx = {0,0};          /* Context object for fts5ParseTokenize */


  pOrig = pExpr->apExprPhrase[iPhrase];

  pNew = (Fts5Expr*)sqlite3Fts5MallocZero(&rc, sizeof(Fts5Expr));
  if( rc==SQLITE_OK ){
    pNew->apExprPhrase = (Fts5ExprPhrase**)sqlite3Fts5MallocZero(&rc, 
        sizeof(Fts5ExprPhrase*));
  }
  if( rc==SQLITE_OK ){
    pNew->pRoot = (Fts5ExprNode*)sqlite3Fts5MallocZero(&rc, 
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407


2408
2409
2410
2411
2412
2413
2414
}

static int fts5ExprPopulatePoslistsCb(
  void *pCtx,                /* Copy of 2nd argument to xTokenize() */
  int tflags,                /* Mask of FTS5_TOKEN_* flags */
  const char *pToken,        /* Pointer to buffer containing token */
  int nToken,                /* Size of token in bytes */
  int iStart,                /* Byte offset of token within input text */
  int iEnd                   /* Byte offset of end of token within input text */
){
  Fts5ExprCtx *p = (Fts5ExprCtx*)pCtx;
  Fts5Expr *pExpr = p->pExpr;
  int i;



  if( (tflags & FTS5_TOKEN_COLOCATED)==0 ) p->iOff++;
  for(i=0; i<pExpr->nPhrase; i++){
    Fts5ExprTerm *pTerm;
    if( p->aPopulator[i].bOk==0 ) continue;
    for(pTerm=&pExpr->apExprPhrase[i]->aTerm[0]; pTerm; pTerm=pTerm->pSynonym){
      int nTerm = strlen(pTerm->zTerm);







|
|




>
>







2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
}

static int fts5ExprPopulatePoslistsCb(
  void *pCtx,                /* Copy of 2nd argument to xTokenize() */
  int tflags,                /* Mask of FTS5_TOKEN_* flags */
  const char *pToken,        /* Pointer to buffer containing token */
  int nToken,                /* Size of token in bytes */
  int iUnused1,              /* Byte offset of token within input text */
  int iUnused2               /* Byte offset of end of token within input text */
){
  Fts5ExprCtx *p = (Fts5ExprCtx*)pCtx;
  Fts5Expr *pExpr = p->pExpr;
  int i;

  UNUSED_PARAM2(iUnused1, iUnused2);

  if( (tflags & FTS5_TOKEN_COLOCATED)==0 ) p->iOff++;
  for(i=0; i<pExpr->nPhrase; i++){
    Fts5ExprTerm *pTerm;
    if( p->aPopulator[i].bOk==0 ) continue;
    for(pTerm=&pExpr->apExprPhrase[i]->aTerm[0]; pTerm; pTerm=pTerm->pSynonym){
      int nTerm = strlen(pTerm->zTerm);
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
   && pNode->iRowid==pExpr->pRoot->iRowid 
   && pPhrase->poslist.n>0
  ){
    Fts5ExprTerm *pTerm = &pPhrase->aTerm[0];
    if( pTerm->pSynonym ){
      Fts5Buffer *pBuf = (Fts5Buffer*)&pTerm->pSynonym[1];
      rc = fts5ExprSynonymList(
          pTerm, 1, 0, pNode->iRowid, pBuf, (u8**)ppCollist, pnCollist
      );
    }else{
      *ppCollist = pPhrase->aTerm[0].pIter->pData;
      *pnCollist = pPhrase->aTerm[0].pIter->nData;
    }
  }else{
    *ppCollist = 0;
    *pnCollist = 0;
  }

  return rc;
}








|













2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
   && pNode->iRowid==pExpr->pRoot->iRowid 
   && pPhrase->poslist.n>0
  ){
    Fts5ExprTerm *pTerm = &pPhrase->aTerm[0];
    if( pTerm->pSynonym ){
      Fts5Buffer *pBuf = (Fts5Buffer*)&pTerm->pSynonym[1];
      rc = fts5ExprSynonymList(
          pTerm, pNode->iRowid, pBuf, (u8**)ppCollist, pnCollist
      );
    }else{
      *ppCollist = pPhrase->aTerm[0].pIter->pData;
      *pnCollist = pPhrase->aTerm[0].pIter->nData;
    }
  }else{
    *ppCollist = 0;
    *pnCollist = 0;
  }

  return rc;
}

Changes to ext/fts5/fts5_index.c.
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776


1777
1778
1779
1780
1781
1782
1783
** Advance iterator pIter to the next entry.
**
** This version of fts5SegIterNext() is only used by reverse iterators.
*/
static void fts5SegIterNext_Reverse(
  Fts5Index *p,                   /* FTS5 backend object */
  Fts5SegIter *pIter,             /* Iterator to advance */
  int *pbNewTerm                  /* OUT: Set for new term */
){
  assert( pIter->flags & FTS5_SEGITER_REVERSE );
  assert( pIter->pNextLeaf==0 );


  if( pIter->iRowidOffset>0 ){
    u8 *a = pIter->pLeaf->p;
    int iOff;
    i64 iDelta;

    pIter->iRowidOffset--;
    pIter->iLeafOffset = pIter->aRowidOffset[pIter->iRowidOffset];







|



>
>







1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
** Advance iterator pIter to the next entry.
**
** This version of fts5SegIterNext() is only used by reverse iterators.
*/
static void fts5SegIterNext_Reverse(
  Fts5Index *p,                   /* FTS5 backend object */
  Fts5SegIter *pIter,             /* Iterator to advance */
  int *pbUnused                   /* Unused */
){
  assert( pIter->flags & FTS5_SEGITER_REVERSE );
  assert( pIter->pNextLeaf==0 );
  UNUSED_PARAM(pbUnused);

  if( pIter->iRowidOffset>0 ){
    u8 *a = pIter->pLeaf->p;
    int iOff;
    i64 iDelta;

    pIter->iRowidOffset--;
    pIter->iLeafOffset = pIter->aRowidOffset[pIter->iRowidOffset];
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
** pSeg. If there is no such term in the index, the iterator is set to EOF.
**
** If an error occurs, Fts5Index.rc is set to an appropriate error code. If 
** an error has already occurred when this function is called, it is a no-op.
*/
static void fts5SegIterSeekInit(
  Fts5Index *p,                   /* FTS5 backend */
  Fts5Buffer *pBuf,               /* Buffer to use for loading pages */
  const u8 *pTerm, int nTerm,     /* Term to seek to */
  int flags,                      /* Mask of FTS5INDEX_XXX flags */
  Fts5StructureSegment *pSeg,     /* Description of segment */
  Fts5SegIter *pIter              /* Object to populate */
){
  int iPg = 1;
  int bGe = (flags & FTS5INDEX_QUERY_SCAN);







<







2248
2249
2250
2251
2252
2253
2254

2255
2256
2257
2258
2259
2260
2261
** pSeg. If there is no such term in the index, the iterator is set to EOF.
**
** If an error occurs, Fts5Index.rc is set to an appropriate error code. If 
** an error has already occurred when this function is called, it is a no-op.
*/
static void fts5SegIterSeekInit(
  Fts5Index *p,                   /* FTS5 backend */

  const u8 *pTerm, int nTerm,     /* Term to seek to */
  int flags,                      /* Mask of FTS5INDEX_XXX flags */
  Fts5StructureSegment *pSeg,     /* Description of segment */
  Fts5SegIter *pIter              /* Object to populate */
){
  int iPg = 1;
  int bGe = (flags & FTS5INDEX_QUERY_SCAN);
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
  }while( p->rc==SQLITE_OK );
}


/*
** Free the iterator object passed as the second argument.
*/
static void fts5MultiIterFree(Fts5Index *p, Fts5Iter *pIter){
  if( pIter ){
    int i;
    for(i=0; i<pIter->nSeg; i++){
      fts5SegIterClear(&pIter->aSeg[i]);
    }
    fts5StructureRelease(pIter->pStruct);
    fts5BufferFree(&pIter->poslist);







|







2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
  }while( p->rc==SQLITE_OK );
}


/*
** Free the iterator object passed as the second argument.
*/
static void fts5MultiIterFree(Fts5Iter *pIter){
  if( pIter ){
    int i;
    for(i=0; i<pIter->nSeg; i++){
      fts5SegIterClear(&pIter->aSeg[i]);
    }
    fts5StructureRelease(pIter->pStruct);
    fts5BufferFree(&pIter->poslist);
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
** If it does so successfully, 0 is returned. Otherwise 1.
**
** If non-zero is returned, the caller should call fts5MultiIterAdvanced()
** on the iterator instead. That function does the same as this one, except
** that it deals with more complicated cases as well.
*/ 
static int fts5MultiIterAdvanceRowid(
  Fts5Index *p,                   /* FTS5 backend to iterate within */
  Fts5Iter *pIter,                /* Iterator to update aFirst[] array for */
  int iChanged,                   /* Index of sub-iterator just advanced */
  Fts5SegIter **ppFirst
){
  Fts5SegIter *pNew = &pIter->aSeg[iChanged];

  if( pNew->iRowid==pIter->iSwitchRowid







<







2675
2676
2677
2678
2679
2680
2681

2682
2683
2684
2685
2686
2687
2688
** If it does so successfully, 0 is returned. Otherwise 1.
**
** If non-zero is returned, the caller should call fts5MultiIterAdvanced()
** on the iterator instead. That function does the same as this one, except
** that it deals with more complicated cases as well.
*/ 
static int fts5MultiIterAdvanceRowid(

  Fts5Iter *pIter,                /* Iterator to update aFirst[] array for */
  int iChanged,                   /* Index of sub-iterator just advanced */
  Fts5SegIter **ppFirst
){
  Fts5SegIter *pNew = &pIter->aSeg[iChanged];

  if( pNew->iRowid==pIter->iSwitchRowid
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
    if( bUseFrom && pSeg->pDlidx ){
      fts5SegIterNextFrom(p, pSeg, iFrom);
    }else{
      pSeg->xNext(p, pSeg, &bNewTerm);
    }

    if( pSeg->pLeaf==0 || bNewTerm 
     || fts5MultiIterAdvanceRowid(p, pIter, iFirst, &pSeg)
    ){
      fts5MultiIterAdvanced(p, pIter, iFirst, 1);
      fts5MultiIterSetEof(pIter);
      pSeg = &pIter->aSeg[pIter->aFirst[1].iFirst];
      if( pSeg->pLeaf==0 ) return;
    }








|







2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
    if( bUseFrom && pSeg->pDlidx ){
      fts5SegIterNextFrom(p, pSeg, iFrom);
    }else{
      pSeg->xNext(p, pSeg, &bNewTerm);
    }

    if( pSeg->pLeaf==0 || bNewTerm 
     || fts5MultiIterAdvanceRowid(pIter, iFirst, &pSeg)
    ){
      fts5MultiIterAdvanced(p, pIter, iFirst, 1);
      fts5MultiIterSetEof(pIter);
      pSeg = &pIter->aSeg[pIter->aFirst[1].iFirst];
      if( pSeg->pLeaf==0 ) return;
    }

2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803

2804
2805
2806
2807
2808
2809
2810
      int iFirst = pIter->aFirst[1].iFirst;
      Fts5SegIter *pSeg = &pIter->aSeg[iFirst];
      int bNewTerm = 0;

      assert( p->rc==SQLITE_OK );
      pSeg->xNext(p, pSeg, &bNewTerm);
      if( pSeg->pLeaf==0 || bNewTerm 
       || fts5MultiIterAdvanceRowid(p, pIter, iFirst, &pSeg)
      ){
        fts5MultiIterAdvanced(p, pIter, iFirst, 1);
        fts5MultiIterSetEof(pIter);
        *pbNewTerm = 1;
      }else{
        *pbNewTerm = 0;
      }
      fts5AssertMultiIterSetup(p, pIter);

    }while( fts5MultiIterIsEmpty(p, pIter) );
  }
}

static void fts5IterSetOutputs_Noop(Fts5Iter *pIter, Fts5SegIter *pSeg){

}

static Fts5Iter *fts5MultiIterAlloc(
  Fts5Index *p,                   /* FTS5 backend to iterate within */
  int nSeg
){
  Fts5Iter *pNew;







|













|
>







2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
      int iFirst = pIter->aFirst[1].iFirst;
      Fts5SegIter *pSeg = &pIter->aSeg[iFirst];
      int bNewTerm = 0;

      assert( p->rc==SQLITE_OK );
      pSeg->xNext(p, pSeg, &bNewTerm);
      if( pSeg->pLeaf==0 || bNewTerm 
       || fts5MultiIterAdvanceRowid(pIter, iFirst, &pSeg)
      ){
        fts5MultiIterAdvanced(p, pIter, iFirst, 1);
        fts5MultiIterSetEof(pIter);
        *pbNewTerm = 1;
      }else{
        *pbNewTerm = 0;
      }
      fts5AssertMultiIterSetup(p, pIter);

    }while( fts5MultiIterIsEmpty(p, pIter) );
  }
}

static void fts5IterSetOutputs_Noop(Fts5Iter *pUnused1, Fts5SegIter *pUnused2){
  UNUSED_PARAM2(pUnused1, pUnused2);
}

static Fts5Iter *fts5MultiIterAlloc(
  Fts5Index *p,                   /* FTS5 backend to iterate within */
  int nSeg
){
  Fts5Iter *pNew;
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832

2833
2834
2835
2836
2837
2838
2839
    pNew->pIndex = p;
    pNew->xSetOutputs = fts5IterSetOutputs_Noop;
  }
  return pNew;
}

static void fts5PoslistCallback(
  Fts5Index *p, 
  void *pContext, 
  const u8 *pChunk, int nChunk
){

  assert_nc( nChunk>=0 );
  if( nChunk>0 ){
    fts5BufferSafeAppendBlob((Fts5Buffer*)pContext, pChunk, nChunk);
  }
}

typedef struct PoslistCallbackCtx PoslistCallbackCtx;







|



>







2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
    pNew->pIndex = p;
    pNew->xSetOutputs = fts5IterSetOutputs_Noop;
  }
  return pNew;
}

static void fts5PoslistCallback(
  Fts5Index *pUnused, 
  void *pContext, 
  const u8 *pChunk, int nChunk
){
  UNUSED_PARAM(pUnused);
  assert_nc( nChunk>=0 );
  if( nChunk>0 ){
    fts5BufferSafeAppendBlob((Fts5Buffer*)pContext, pChunk, nChunk);
  }
}

typedef struct PoslistCallbackCtx PoslistCallbackCtx;
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870

2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892

2893
2894
2895
2896
2897
2898
2899
  for(i=0; i<pColset->nCol; i++){
    if( pColset->aiCol[i]==iCol ) return 1;
  }
  return 0;
}

static void fts5PoslistOffsetsCallback(
  Fts5Index *p, 
  void *pContext, 
  const u8 *pChunk, int nChunk
){
  PoslistOffsetsCtx *pCtx = (PoslistOffsetsCtx*)pContext;

  assert_nc( nChunk>=0 );
  if( nChunk>0 ){
    int i = 0;
    while( i<nChunk ){
      int iVal;
      i += fts5GetVarint32(&pChunk[i], iVal);
      iVal += pCtx->iRead - 2;
      pCtx->iRead = iVal;
      if( fts5IndexColsetTest(pCtx->pColset, iVal) ){
        fts5BufferSafeAppendVarint(pCtx->pBuf, iVal + 2 - pCtx->iWrite);
        pCtx->iWrite = iVal;
      }
    }
  }
}

static void fts5PoslistFilterCallback(
  Fts5Index *p, 
  void *pContext, 
  const u8 *pChunk, int nChunk
){
  PoslistCallbackCtx *pCtx = (PoslistCallbackCtx*)pContext;

  assert_nc( nChunk>=0 );
  if( nChunk>0 ){
    /* Search through to find the first varint with value 1. This is the
    ** start of the next columns hits. */
    int i = 0;
    int iStart = 0;








|




>

















|




>







2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
  for(i=0; i<pColset->nCol; i++){
    if( pColset->aiCol[i]==iCol ) return 1;
  }
  return 0;
}

static void fts5PoslistOffsetsCallback(
  Fts5Index *pUnused, 
  void *pContext, 
  const u8 *pChunk, int nChunk
){
  PoslistOffsetsCtx *pCtx = (PoslistOffsetsCtx*)pContext;
  UNUSED_PARAM(pUnused);
  assert_nc( nChunk>=0 );
  if( nChunk>0 ){
    int i = 0;
    while( i<nChunk ){
      int iVal;
      i += fts5GetVarint32(&pChunk[i], iVal);
      iVal += pCtx->iRead - 2;
      pCtx->iRead = iVal;
      if( fts5IndexColsetTest(pCtx->pColset, iVal) ){
        fts5BufferSafeAppendVarint(pCtx->pBuf, iVal + 2 - pCtx->iWrite);
        pCtx->iWrite = iVal;
      }
    }
  }
}

static void fts5PoslistFilterCallback(
  Fts5Index *pUnused,
  void *pContext, 
  const u8 *pChunk, int nChunk
){
  PoslistCallbackCtx *pCtx = (PoslistCallbackCtx*)pContext;
  UNUSED_PARAM(pUnused);
  assert_nc( nChunk>=0 );
  if( nChunk>0 ){
    /* Search through to find the first varint with value 1. This is the
    ** start of the next columns hits. */
    int i = 0;
    int iStart = 0;

3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
  int iLevel,                     /* Level to iterate (-1 for all) */
  int nSegment,                   /* Number of segments to merge (iLevel>=0) */
  Fts5Iter **ppOut                /* New object */
){
  int nSeg = 0;                   /* Number of segment-iters in use */
  int iIter = 0;                  /* */
  int iSeg;                       /* Used to iterate through segments */
  Fts5Buffer buf = {0,0,0};       /* Buffer used by fts5SegIterSeekInit() */
  Fts5StructureLevel *pLvl;
  Fts5Iter *pNew;

  assert( (pTerm==0 && nTerm==0) || iLevel<0 );

  /* Allocate space for the new multi-seg-iterator. */
  if( p->rc==SQLITE_OK ){







<







3253
3254
3255
3256
3257
3258
3259

3260
3261
3262
3263
3264
3265
3266
  int iLevel,                     /* Level to iterate (-1 for all) */
  int nSegment,                   /* Number of segments to merge (iLevel>=0) */
  Fts5Iter **ppOut                /* New object */
){
  int nSeg = 0;                   /* Number of segment-iters in use */
  int iIter = 0;                  /* */
  int iSeg;                       /* Used to iterate through segments */

  Fts5StructureLevel *pLvl;
  Fts5Iter *pNew;

  assert( (pTerm==0 && nTerm==0) || iLevel<0 );

  /* Allocate space for the new multi-seg-iterator. */
  if( p->rc==SQLITE_OK ){
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
      for(pLvl=&pStruct->aLevel[0]; pLvl<pEnd; pLvl++){
        for(iSeg=pLvl->nSeg-1; iSeg>=0; iSeg--){
          Fts5StructureSegment *pSeg = &pLvl->aSeg[iSeg];
          Fts5SegIter *pIter = &pNew->aSeg[iIter++];
          if( pTerm==0 ){
            fts5SegIterInit(p, pSeg, pIter);
          }else{
            fts5SegIterSeekInit(p, &buf, pTerm, nTerm, flags, pSeg, pIter);
          }
        }
      }
    }else{
      pLvl = &pStruct->aLevel[iLevel];
      for(iSeg=nSeg-1; iSeg>=0; iSeg--){
        fts5SegIterInit(p, &pLvl->aSeg[iSeg], &pNew->aSeg[iIter++]);







|







3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
      for(pLvl=&pStruct->aLevel[0]; pLvl<pEnd; pLvl++){
        for(iSeg=pLvl->nSeg-1; iSeg>=0; iSeg--){
          Fts5StructureSegment *pSeg = &pLvl->aSeg[iSeg];
          Fts5SegIter *pIter = &pNew->aSeg[iIter++];
          if( pTerm==0 ){
            fts5SegIterInit(p, pSeg, pIter);
          }else{
            fts5SegIterSeekInit(p, pTerm, nTerm, flags, pSeg, pIter);
          }
        }
      }
    }else{
      pLvl = &pStruct->aLevel[iLevel];
      for(iSeg=nSeg-1; iSeg>=0; iSeg--){
        fts5SegIterInit(p, &pLvl->aSeg[iSeg], &pNew->aSeg[iIter++]);
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
      fts5MultiIterNext(p, pNew, 0, 0);
    }else if( pNew->base.bEof==0 ){
      Fts5SegIter *pSeg = &pNew->aSeg[pNew->aFirst[1].iFirst];
      pNew->xSetOutputs(pNew, pSeg);
    }

  }else{
    fts5MultiIterFree(p, pNew);
    *ppOut = 0;
  }
  fts5BufferFree(&buf);

}

/*
** Create an Fts5Iter that iterates through the doclist provided
** as the second argument.
*/
static void fts5MultiIterNew2(







|


<
<







3332
3333
3334
3335
3336
3337
3338
3339
3340
3341


3342
3343
3344
3345
3346
3347
3348
      fts5MultiIterNext(p, pNew, 0, 0);
    }else if( pNew->base.bEof==0 ){
      Fts5SegIter *pSeg = &pNew->aSeg[pNew->aFirst[1].iFirst];
      pNew->xSetOutputs(pNew, pSeg);
    }

  }else{
    fts5MultiIterFree(pNew);
    *ppOut = 0;
  }


}

/*
** Create an Fts5Iter that iterates through the doclist provided
** as the second argument.
*/
static void fts5MultiIterNew2(
3471
3472
3473
3474
3475
3476
3477
3478



3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
  if( p->pHash ){
    sqlite3Fts5HashClear(p->pHash);
    p->nPendingData = 0;
  }
}

/*
** Return the size of the prefix, in bytes, that buffer (nNew/pNew) shares



** with buffer (nOld/pOld).
*/
static int fts5PrefixCompress(
  int nOld, const u8 *pOld,
  int nNew, const u8 *pNew
){
  int i;
  assert( fts5BlobCompare(pOld, nOld, pNew, nNew)<0 );
  for(i=0; i<nOld; i++){
    if( pOld[i]!=pNew[i] ) break;
  }
  return i;
}

static void fts5WriteDlidxClear(







|
>
>
>
|

|
<
<
<

<







3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485



3486

3487
3488
3489
3490
3491
3492
3493
  if( p->pHash ){
    sqlite3Fts5HashClear(p->pHash);
    p->nPendingData = 0;
  }
}

/*
** Return the size of the prefix, in bytes, that buffer 
** (pNew/<length-unknown>) shares with buffer (pOld/nOld).
**
** Buffer (pNew/<length-unknown>) is guaranteed to be greater 
** than buffer (pOld/nOld).
*/
static int fts5PrefixCompress(int nOld, const u8 *pOld, const u8 *pNew){



  int i;

  for(i=0; i<nOld; i++){
    if( pOld[i]!=pNew[i] ) break;
  }
  return i;
}

static void fts5WriteDlidxClear(
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
      ** Usually, the previous term is available in pPage->term. The exception
      ** is if this is the first term written in an incremental-merge step.
      ** In this case the previous term is not available, so just write a
      ** copy of (pTerm/nTerm) into the parent node. This is slightly
      ** inefficient, but still correct.  */
      int n = nTerm;
      if( pPage->term.n ){
        n = 1 + fts5PrefixCompress(pPage->term.n, pPage->term.p, nTerm, pTerm);
      }
      fts5WriteBtreeTerm(p, pWriter, n, pTerm);
      pPage = &pWriter->writer;
    }
  }else{
    nPrefix = fts5PrefixCompress(pPage->term.n, pPage->term.p, nTerm, pTerm);
    fts5BufferAppendVarint(&p->rc, &pPage->buf, nPrefix);
  }

  /* Append the number of bytes of new data, then the term data itself
  ** to the page. */
  fts5BufferAppendVarint(&p->rc, &pPage->buf, nTerm - nPrefix);
  fts5BufferAppendBlob(&p->rc, &pPage->buf, nTerm - nPrefix, &pTerm[nPrefix]);







|





|







3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
      ** Usually, the previous term is available in pPage->term. The exception
      ** is if this is the first term written in an incremental-merge step.
      ** In this case the previous term is not available, so just write a
      ** copy of (pTerm/nTerm) into the parent node. This is slightly
      ** inefficient, but still correct.  */
      int n = nTerm;
      if( pPage->term.n ){
        n = 1 + fts5PrefixCompress(pPage->term.n, pPage->term.p, pTerm);
      }
      fts5WriteBtreeTerm(p, pWriter, n, pTerm);
      pPage = &pWriter->writer;
    }
  }else{
    nPrefix = fts5PrefixCompress(pPage->term.n, pPage->term.p, pTerm);
    fts5BufferAppendVarint(&p->rc, &pPage->buf, nPrefix);
  }

  /* Append the number of bytes of new data, then the term data itself
  ** to the page. */
  fts5BufferAppendVarint(&p->rc, &pPage->buf, nTerm - nPrefix);
  fts5BufferAppendBlob(&p->rc, &pPage->buf, nTerm - nPrefix, &pTerm[nPrefix]);
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
    }
  }else{
    assert( pSeg->pgnoLast>0 );
    fts5TrimSegments(p, pIter);
    pLvl->nMerge = nInput;
  }

  fts5MultiIterFree(p, pIter);
  fts5BufferFree(&term);
  if( pnRem ) *pnRem -= writer.nLeafWritten;
}

/*
** Do up to nPg pages of automerge work on the index.
*/







|







4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
    }
  }else{
    assert( pSeg->pgnoLast>0 );
    fts5TrimSegments(p, pIter);
    pLvl->nMerge = nInput;
  }

  fts5MultiIterFree(pIter);
  fts5BufferFree(&term);
  if( pnRem ) *pnRem -= writer.nLeafWritten;
}

/*
** Do up to nPg pages of automerge work on the index.
*/
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546

4547
4548
4549
4550
4551
4552
4553

  return fts5IndexReturn(p);
}

static void fts5AppendRowid(
  Fts5Index *p,
  i64 iDelta,
  Fts5Iter *pMulti,
  Fts5Buffer *pBuf
){

  fts5BufferAppendVarint(&p->rc, pBuf, iDelta);
}

static void fts5AppendPoslist(
  Fts5Index *p,
  i64 iDelta,
  Fts5Iter *pMulti,







|


>







4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554

  return fts5IndexReturn(p);
}

static void fts5AppendRowid(
  Fts5Index *p,
  i64 iDelta,
  Fts5Iter *pUnused,
  Fts5Buffer *pBuf
){
  UNUSED_PARAM(pUnused);
  fts5BufferAppendVarint(&p->rc, pBuf, iDelta);
}

static void fts5AppendPoslist(
  Fts5Index *p,
  i64 iDelta,
  Fts5Iter *pMulti,
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896

    for(i=0; i<nBuf; i++){
      if( p->rc==SQLITE_OK ){
        xMerge(p, &doclist, &aBuf[i]);
      }
      fts5BufferFree(&aBuf[i]);
    }
    fts5MultiIterFree(p, p1);

    pData = fts5IdxMalloc(p, sizeof(Fts5Data) + doclist.n);
    if( pData ){
      pData->p = (u8*)&pData[1];
      pData->nn = pData->szLeaf = doclist.n;
      memcpy(pData->p, doclist.p, doclist.n);
      fts5MultiIterNew2(p, pData, bDesc, ppIter);







|







4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897

    for(i=0; i<nBuf; i++){
      if( p->rc==SQLITE_OK ){
        xMerge(p, &doclist, &aBuf[i]);
      }
      fts5BufferFree(&aBuf[i]);
    }
    fts5MultiIterFree(p1);

    pData = fts5IdxMalloc(p, sizeof(Fts5Data) + doclist.n);
    if( pData ){
      pData->p = (u8*)&pData[1];
      pData->nn = pData->szLeaf = doclist.n;
      memcpy(pData->p, doclist.p, doclist.n);
      fts5MultiIterNew2(p, pData, bDesc, ppIter);
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
        fts5StructureRelease(pStruct);
      }
    }else{
      /* Scan multiple terms in the main index */
      int bDesc = (flags & FTS5INDEX_QUERY_DESC)!=0;
      buf.p[0] = FTS5_MAIN_PREFIX;
      fts5SetupPrefixIter(p, bDesc, buf.p, nToken+1, pColset, &pRet);
      assert( pRet->pColset==0 );
      fts5IterSetOutputCb(&p->rc, pRet);
      if( p->rc==SQLITE_OK ){
        Fts5SegIter *pSeg = &pRet->aSeg[pRet->aFirst[1].iFirst];
        if( pSeg->pLeaf ) pRet->xSetOutputs(pRet, pSeg);
      }
    }








|







5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
        fts5StructureRelease(pStruct);
      }
    }else{
      /* Scan multiple terms in the main index */
      int bDesc = (flags & FTS5INDEX_QUERY_DESC)!=0;
      buf.p[0] = FTS5_MAIN_PREFIX;
      fts5SetupPrefixIter(p, bDesc, buf.p, nToken+1, pColset, &pRet);
      assert( p->rc!=SQLITE_OK || pRet->pColset==0 );
      fts5IterSetOutputCb(&p->rc, pRet);
      if( p->rc==SQLITE_OK ){
        Fts5SegIter *pSeg = &pRet->aSeg[pRet->aFirst[1].iFirst];
        if( pSeg->pLeaf ) pRet->xSetOutputs(pRet, pSeg);
      }
    }

5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
/*
** Close an iterator opened by an earlier call to sqlite3Fts5IndexQuery().
*/
void sqlite3Fts5IterClose(Fts5IndexIter *pIndexIter){
  if( pIndexIter ){
    Fts5Iter *pIter = (Fts5Iter*)pIndexIter;
    Fts5Index *pIndex = pIter->pIndex;
    fts5MultiIterFree(pIter->pIndex, pIter);
    fts5CloseReader(pIndex);
  }
}

/*
** Read and decode the "averages" record from the database. 
**







|







5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
/*
** Close an iterator opened by an earlier call to sqlite3Fts5IndexQuery().
*/
void sqlite3Fts5IterClose(Fts5IndexIter *pIndexIter){
  if( pIndexIter ){
    Fts5Iter *pIter = (Fts5Iter*)pIndexIter;
    Fts5Index *pIndex = pIter->pIndex;
    fts5MultiIterFree(pIter);
    fts5CloseReader(pIndex);
  }
}

/*
** Read and decode the "averages" record from the database. 
**
5805
5806
5807
5808
5809
5810
5811
5812
5813
5814
5815
5816
5817
5818
5819
        int iTokOff = FTS5_POS2OFFSET(iPos);
        cksum2 ^= sqlite3Fts5IndexEntryCksum(iRowid, iCol, iTokOff, -1, z, n);
      }
    }
  }
  fts5TestTerm(p, &term, 0, 0, cksum2, &cksum3);

  fts5MultiIterFree(p, pIter);
  if( p->rc==SQLITE_OK && cksum!=cksum2 ) p->rc = FTS5_CORRUPT;

  fts5StructureRelease(pStruct);
#ifdef SQLITE_DEBUG
  fts5BufferFree(&term);
#endif
  fts5BufferFree(&poslist);







|







5806
5807
5808
5809
5810
5811
5812
5813
5814
5815
5816
5817
5818
5819
5820
        int iTokOff = FTS5_POS2OFFSET(iPos);
        cksum2 ^= sqlite3Fts5IndexEntryCksum(iRowid, iCol, iTokOff, -1, z, n);
      }
    }
  }
  fts5TestTerm(p, &term, 0, 0, cksum2, &cksum3);

  fts5MultiIterFree(pIter);
  if( p->rc==SQLITE_OK && cksum!=cksum2 ) p->rc = FTS5_CORRUPT;

  fts5StructureRelease(pStruct);
#ifdef SQLITE_DEBUG
  fts5BufferFree(&term);
#endif
  fts5BufferFree(&poslist);
6042
6043
6044
6045
6046
6047
6048

6049
6050
6051
6052
6053
6054
6055
  u8 *a = 0;
  Fts5Buffer s;                   /* Build up text to return here */
  int rc = SQLITE_OK;             /* Return code */
  int nSpace = 0;
  int eDetailNone = (sqlite3_user_data(pCtx)!=0);

  assert( nArg==2 );

  memset(&s, 0, sizeof(Fts5Buffer));
  iRowid = sqlite3_value_int64(apVal[0]);

  /* Make a copy of the second argument (a blob) in aBlob[]. The aBlob[]
  ** copy is followed by FTS5_DATA_ZERO_PADDING 0x00 bytes, which prevents
  ** buffer overreads even if the record is corrupt.  */
  n = sqlite3_value_bytes(apVal[1]);







>







6043
6044
6045
6046
6047
6048
6049
6050
6051
6052
6053
6054
6055
6056
6057
  u8 *a = 0;
  Fts5Buffer s;                   /* Build up text to return here */
  int rc = SQLITE_OK;             /* Return code */
  int nSpace = 0;
  int eDetailNone = (sqlite3_user_data(pCtx)!=0);

  assert( nArg==2 );
  UNUSED_PARAM(nArg);
  memset(&s, 0, sizeof(Fts5Buffer));
  iRowid = sqlite3_value_int64(apVal[0]);

  /* Make a copy of the second argument (a blob) in aBlob[]. The aBlob[]
  ** copy is followed by FTS5_DATA_ZERO_PADDING 0x00 bytes, which prevents
  ** buffer overreads even if the record is corrupt.  */
  n = sqlite3_value_bytes(apVal[1]);
Changes to ext/fts5/fts5_main.c.
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115



1116
1117
1118
1119
1120
1121
1122
**   1. Full-text search using a MATCH operator.
**   2. A by-rowid lookup.
**   3. A full-table scan.
*/
static int fts5FilterMethod(
  sqlite3_vtab_cursor *pCursor,   /* The cursor used for this query */
  int idxNum,                     /* Strategy index */
  const char *idxStr,             /* Unused */
  int nVal,                       /* Number of elements in apVal */
  sqlite3_value **apVal           /* Arguments for the indexing scheme */
){
  Fts5Table *pTab = (Fts5Table*)(pCursor->pVtab);
  Fts5Config *pConfig = pTab->pConfig;
  Fts5Cursor *pCsr = (Fts5Cursor*)pCursor;
  int rc = SQLITE_OK;             /* Error code */
  int iVal = 0;                   /* Counter for apVal[] */
  int bDesc;                      /* True if ORDER BY [rank|rowid] DESC */
  int bOrderByRank;               /* True if ORDER BY rank */
  sqlite3_value *pMatch = 0;      /* <tbl> MATCH ? expression (or NULL) */
  sqlite3_value *pRank = 0;       /* rank MATCH ? expression (or NULL) */
  sqlite3_value *pRowidEq = 0;    /* rowid = ? expression (or NULL) */
  sqlite3_value *pRowidLe = 0;    /* rowid <= ? expression (or NULL) */
  sqlite3_value *pRowidGe = 0;    /* rowid >= ? expression (or NULL) */
  char **pzErrmsg = pConfig->pzErrmsg;




  if( pCsr->ePlan ){
    fts5FreeCursorComponents(pCsr);
    memset(&pCsr->ePlan, 0, sizeof(Fts5Cursor) - ((u8*)&pCsr->ePlan-(u8*)pCsr));
  }

  assert( pCsr->pStmt==0 );







|
















>
>
>







1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
**   1. Full-text search using a MATCH operator.
**   2. A by-rowid lookup.
**   3. A full-table scan.
*/
static int fts5FilterMethod(
  sqlite3_vtab_cursor *pCursor,   /* The cursor used for this query */
  int idxNum,                     /* Strategy index */
  const char *zUnused,            /* Unused */
  int nVal,                       /* Number of elements in apVal */
  sqlite3_value **apVal           /* Arguments for the indexing scheme */
){
  Fts5Table *pTab = (Fts5Table*)(pCursor->pVtab);
  Fts5Config *pConfig = pTab->pConfig;
  Fts5Cursor *pCsr = (Fts5Cursor*)pCursor;
  int rc = SQLITE_OK;             /* Error code */
  int iVal = 0;                   /* Counter for apVal[] */
  int bDesc;                      /* True if ORDER BY [rank|rowid] DESC */
  int bOrderByRank;               /* True if ORDER BY rank */
  sqlite3_value *pMatch = 0;      /* <tbl> MATCH ? expression (or NULL) */
  sqlite3_value *pRank = 0;       /* rank MATCH ? expression (or NULL) */
  sqlite3_value *pRowidEq = 0;    /* rowid = ? expression (or NULL) */
  sqlite3_value *pRowidLe = 0;    /* rowid <= ? expression (or NULL) */
  sqlite3_value *pRowidGe = 0;    /* rowid >= ? expression (or NULL) */
  char **pzErrmsg = pConfig->pzErrmsg;

  UNUSED_PARAM(zUnused);
  UNUSED_PARAM(nVal);

  if( pCsr->ePlan ){
    fts5FreeCursorComponents(pCsr);
    memset(&pCsr->ePlan, 0, sizeof(Fts5Cursor) - ((u8*)&pCsr->ePlan-(u8*)pCsr));
  }

  assert( pCsr->pStmt==0 );
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
    }
  }
  return rc;
}

static int fts5SpecialDelete(
  Fts5Table *pTab, 
  sqlite3_value **apVal, 
  sqlite3_int64 *piRowid
){
  int rc = SQLITE_OK;
  int eType1 = sqlite3_value_type(apVal[1]);
  if( eType1==SQLITE_INTEGER ){
    sqlite3_int64 iDel = sqlite3_value_int64(apVal[1]);
    rc = sqlite3Fts5StorageDelete(pTab->pStorage, iDel, &apVal[2]);
  }







|
<







1397
1398
1399
1400
1401
1402
1403
1404

1405
1406
1407
1408
1409
1410
1411
    }
  }
  return rc;
}

static int fts5SpecialDelete(
  Fts5Table *pTab, 
  sqlite3_value **apVal

){
  int rc = SQLITE_OK;
  int eType1 = sqlite3_value_type(apVal[1]);
  if( eType1==SQLITE_INTEGER ){
    sqlite3_int64 iDel = sqlite3_value_int64(apVal[1]);
    rc = sqlite3Fts5StorageDelete(pTab->pStorage, iDel, &apVal[2]);
  }
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
   && sqlite3_value_type(apVal[2+pConfig->nCol])!=SQLITE_NULL 
  ){
    /* A "special" INSERT op. These are handled separately. */
    const char *z = (const char*)sqlite3_value_text(apVal[2+pConfig->nCol]);
    if( pConfig->eContent!=FTS5_CONTENT_NORMAL 
      && 0==sqlite3_stricmp("delete", z) 
    ){
      rc = fts5SpecialDelete(pTab, apVal, pRowid);
    }else{
      rc = fts5SpecialInsert(pTab, z, apVal[2 + pConfig->nCol + 1]);
    }
  }else{
    /* A regular INSERT, UPDATE or DELETE statement. The trick here is that
    ** any conflict on the rowid value must be detected before any 
    ** modifications are made to the database file. There are 4 cases:







|







1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
   && sqlite3_value_type(apVal[2+pConfig->nCol])!=SQLITE_NULL 
  ){
    /* A "special" INSERT op. These are handled separately. */
    const char *z = (const char*)sqlite3_value_text(apVal[2+pConfig->nCol]);
    if( pConfig->eContent!=FTS5_CONTENT_NORMAL 
      && 0==sqlite3_stricmp("delete", z) 
    ){
      rc = fts5SpecialDelete(pTab, apVal);
    }else{
      rc = fts5SpecialInsert(pTab, z, apVal[2 + pConfig->nCol + 1]);
    }
  }else{
    /* A regular INSERT, UPDATE or DELETE statement. The trick here is that
    ** any conflict on the rowid value must be detected before any 
    ** modifications are made to the database file. There are 4 cases:
1572
1573
1574
1575
1576
1577
1578

1579
1580
1581
1582
1583
1584
1585
1586
1587
1588

1589
1590
1591
1592
1593
1594
1595
  return rc;
}

/*
** Implementation of xBegin() method. 
*/
static int fts5BeginMethod(sqlite3_vtab *pVtab){

  fts5CheckTransactionState((Fts5Table*)pVtab, FTS5_BEGIN, 0);
  return SQLITE_OK;
}

/*
** Implementation of xCommit() method. This is a no-op. The contents of
** the pending-terms hash-table have already been flushed into the database
** by fts5SyncMethod().
*/
static int fts5CommitMethod(sqlite3_vtab *pVtab){

  fts5CheckTransactionState((Fts5Table*)pVtab, FTS5_COMMIT, 0);
  return SQLITE_OK;
}

/*
** Implementation of xRollback(). Discard the contents of the pending-terms
** hash-table. Any changes made to the database are reverted by SQLite.







>










>







1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
  return rc;
}

/*
** Implementation of xBegin() method. 
*/
static int fts5BeginMethod(sqlite3_vtab *pVtab){
  UNUSED_PARAM(pVtab);  /* Call below is a no-op for NDEBUG builds */
  fts5CheckTransactionState((Fts5Table*)pVtab, FTS5_BEGIN, 0);
  return SQLITE_OK;
}

/*
** Implementation of xCommit() method. This is a no-op. The contents of
** the pending-terms hash-table have already been flushed into the database
** by fts5SyncMethod().
*/
static int fts5CommitMethod(sqlite3_vtab *pVtab){
  UNUSED_PARAM(pVtab);  /* Call below is a no-op for NDEBUG builds */
  fts5CheckTransactionState((Fts5Table*)pVtab, FTS5_COMMIT, 0);
  return SQLITE_OK;
}

/*
** Implementation of xRollback(). Discard the contents of the pending-terms
** hash-table. Any changes made to the database are reverted by SQLite.
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847


1848
1849
1850
1851
1852
1853
1854
static sqlite3_int64 fts5ApiRowid(Fts5Context *pCtx){
  return fts5CursorRowid((Fts5Cursor*)pCtx);
}

static int fts5ColumnSizeCb(
  void *pContext,                 /* Pointer to int */
  int tflags,
  const char *pToken,             /* Buffer containing token */
  int nToken,                     /* Size of token in bytes */
  int iStart,                     /* Start offset of token */
  int iEnd                        /* End offset of token */
){
  int *pCnt = (int*)pContext;


  if( (tflags & FTS5_TOKEN_COLOCATED)==0 ){
    (*pCnt)++;
  }
  return SQLITE_OK;
}

static int fts5ApiColumnSize(Fts5Context *pCtx, int iCol, int *pnToken){







|
|
|
|


>
>







1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
static sqlite3_int64 fts5ApiRowid(Fts5Context *pCtx){
  return fts5CursorRowid((Fts5Cursor*)pCtx);
}

static int fts5ColumnSizeCb(
  void *pContext,                 /* Pointer to int */
  int tflags,
  const char *pUnused,            /* Buffer containing token */
  int nUnused,                    /* Size of token in bytes */
  int iUnused1,                   /* Start offset of token */
  int iUnused2                    /* End offset of token */
){
  int *pCnt = (int*)pContext;
  UNUSED_PARAM2(pUnused, nUnused);
  UNUSED_PARAM2(iUnused1, iUnused2);
  if( (tflags & FTS5_TOKEN_COLOCATED)==0 ){
    (*pCnt)++;
  }
  return SQLITE_OK;
}

static int fts5ApiColumnSize(Fts5Context *pCtx, int iCol, int *pnToken){
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966

1967
1968
1969
1970
1971
1972
1973
    }
  }

  return pRet;
}

static void fts5ApiPhraseNext(
  Fts5Context *pCtx, 
  Fts5PhraseIter *pIter, 
  int *piCol, int *piOff
){

  if( pIter->a>=pIter->b ){
    *piCol = -1;
    *piOff = -1;
  }else{
    int iVal;
    pIter->a += fts5GetVarint32(pIter->a, iVal);
    if( iVal==1 ){







|



>







1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
    }
  }

  return pRet;
}

static void fts5ApiPhraseNext(
  Fts5Context *pUnused, 
  Fts5PhraseIter *pIter, 
  int *piCol, int *piOff
){
  UNUSED_PARAM(pUnused);
  if( pIter->a>=pIter->b ){
    *piCol = -1;
    *piOff = -1;
  }else{
    int iVal;
    pIter->a += fts5GetVarint32(pIter->a, iVal);
    if( iVal==1 ){
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
  Fts5Cursor *pCsr = (Fts5Cursor*)pCtx;
  Fts5Table *pTab = (Fts5Table*)(pCsr->base.pVtab);
  int rc;
  Fts5Cursor *pNew = 0;

  rc = fts5OpenMethod(pCsr->base.pVtab, (sqlite3_vtab_cursor**)&pNew);
  if( rc==SQLITE_OK ){
    Fts5Config *pConf = pTab->pConfig;
    pNew->ePlan = FTS5_PLAN_MATCH;
    pNew->iFirstRowid = SMALLEST_INT64;
    pNew->iLastRowid = LARGEST_INT64;
    pNew->base.pVtab = (sqlite3_vtab*)pTab;
    rc = sqlite3Fts5ExprClonePhrase(pConf, pCsr->pExpr, iPhrase, &pNew->pExpr);
  }

  if( rc==SQLITE_OK ){
    for(rc = fts5CursorFirst(pTab, pNew, 0);
        rc==SQLITE_OK && CsrFlagTest(pNew, FTS5CSR_EOF)==0;
        rc = fts5NextMethod((sqlite3_vtab_cursor*)pNew)
    ){







<




|







2118
2119
2120
2121
2122
2123
2124

2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
  Fts5Cursor *pCsr = (Fts5Cursor*)pCtx;
  Fts5Table *pTab = (Fts5Table*)(pCsr->base.pVtab);
  int rc;
  Fts5Cursor *pNew = 0;

  rc = fts5OpenMethod(pCsr->base.pVtab, (sqlite3_vtab_cursor**)&pNew);
  if( rc==SQLITE_OK ){

    pNew->ePlan = FTS5_PLAN_MATCH;
    pNew->iFirstRowid = SMALLEST_INT64;
    pNew->iLastRowid = LARGEST_INT64;
    pNew->base.pVtab = (sqlite3_vtab*)pTab;
    rc = sqlite3Fts5ExprClonePhrase(pCsr->pExpr, iPhrase, &pNew->pExpr);
  }

  if( rc==SQLITE_OK ){
    for(rc = fts5CursorFirst(pTab, pNew, 0);
        rc==SQLITE_OK && CsrFlagTest(pNew, FTS5CSR_EOF)==0;
        rc = fts5NextMethod((sqlite3_vtab_cursor*)pNew)
    ){
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343

2344
2345
2346
2347
2348
2349
2350

/*
** This routine implements the xFindFunction method for the FTS3
** virtual table.
*/
static int fts5FindFunctionMethod(
  sqlite3_vtab *pVtab,            /* Virtual table handle */
  int nArg,                       /* Number of SQL function arguments */
  const char *zName,              /* Name of SQL function */
  void (**pxFunc)(sqlite3_context*,int,sqlite3_value**), /* OUT: Result */
  void **ppArg                    /* OUT: User data for *pxFunc */
){
  Fts5Table *pTab = (Fts5Table*)pVtab;
  Fts5Auxiliary *pAux;


  pAux = fts5FindAuxiliary(pTab, zName);
  if( pAux ){
    *pxFunc = fts5ApiCallback;
    *ppArg = (void*)pAux;
    return 1;
  }








|







>







2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357

/*
** This routine implements the xFindFunction method for the FTS3
** virtual table.
*/
static int fts5FindFunctionMethod(
  sqlite3_vtab *pVtab,            /* Virtual table handle */
  int nUnused,                    /* Number of SQL function arguments */
  const char *zName,              /* Name of SQL function */
  void (**pxFunc)(sqlite3_context*,int,sqlite3_value**), /* OUT: Result */
  void **ppArg                    /* OUT: User data for *pxFunc */
){
  Fts5Table *pTab = (Fts5Table*)pVtab;
  Fts5Auxiliary *pAux;

  UNUSED_PARAM(nUnused);
  pAux = fts5FindAuxiliary(pTab, zName);
  if( pAux ){
    *pxFunc = fts5ApiCallback;
    *ppArg = (void*)pAux;
    return 1;
  }

2366
2367
2368
2369
2370
2371
2372

2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384

2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396

2397
2398
2399
2400
2401
2402
2403
/*
** The xSavepoint() method.
**
** Flush the contents of the pending-terms table to disk.
*/
static int fts5SavepointMethod(sqlite3_vtab *pVtab, int iSavepoint){
  Fts5Table *pTab = (Fts5Table*)pVtab;

  fts5CheckTransactionState(pTab, FTS5_SAVEPOINT, iSavepoint);
  fts5TripCursors(pTab);
  return sqlite3Fts5StorageSync(pTab->pStorage, 0);
}

/*
** The xRelease() method.
**
** This is a no-op.
*/
static int fts5ReleaseMethod(sqlite3_vtab *pVtab, int iSavepoint){
  Fts5Table *pTab = (Fts5Table*)pVtab;

  fts5CheckTransactionState(pTab, FTS5_RELEASE, iSavepoint);
  fts5TripCursors(pTab);
  return sqlite3Fts5StorageSync(pTab->pStorage, 0);
}

/*
** The xRollbackTo() method.
**
** Discard the contents of the pending terms table.
*/
static int fts5RollbackToMethod(sqlite3_vtab *pVtab, int iSavepoint){
  Fts5Table *pTab = (Fts5Table*)pVtab;

  fts5CheckTransactionState(pTab, FTS5_ROLLBACKTO, iSavepoint);
  fts5TripCursors(pTab);
  return sqlite3Fts5StorageRollback(pTab->pStorage);
}

/*
** Register a new auxiliary function with global context pGlobal.







>












>












>







2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
/*
** The xSavepoint() method.
**
** Flush the contents of the pending-terms table to disk.
*/
static int fts5SavepointMethod(sqlite3_vtab *pVtab, int iSavepoint){
  Fts5Table *pTab = (Fts5Table*)pVtab;
  UNUSED_PARAM(iSavepoint);  /* Call below is a no-op for NDEBUG builds */
  fts5CheckTransactionState(pTab, FTS5_SAVEPOINT, iSavepoint);
  fts5TripCursors(pTab);
  return sqlite3Fts5StorageSync(pTab->pStorage, 0);
}

/*
** The xRelease() method.
**
** This is a no-op.
*/
static int fts5ReleaseMethod(sqlite3_vtab *pVtab, int iSavepoint){
  Fts5Table *pTab = (Fts5Table*)pVtab;
  UNUSED_PARAM(iSavepoint);  /* Call below is a no-op for NDEBUG builds */
  fts5CheckTransactionState(pTab, FTS5_RELEASE, iSavepoint);
  fts5TripCursors(pTab);
  return sqlite3Fts5StorageSync(pTab->pStorage, 0);
}

/*
** The xRollbackTo() method.
**
** Discard the contents of the pending terms table.
*/
static int fts5RollbackToMethod(sqlite3_vtab *pVtab, int iSavepoint){
  Fts5Table *pTab = (Fts5Table*)pVtab;
  UNUSED_PARAM(iSavepoint);  /* Call below is a no-op for NDEBUG builds */
  fts5CheckTransactionState(pTab, FTS5_ROLLBACKTO, iSavepoint);
  fts5TripCursors(pTab);
  return sqlite3Fts5StorageRollback(pTab->pStorage);
}

/*
** Register a new auxiliary function with global context pGlobal.
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579

2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594

2595
2596
2597
2598
2599
2600
2601

  sqlite3_free(pGlobal);
}

static void fts5Fts5Func(
  sqlite3_context *pCtx,          /* Function call context */
  int nArg,                       /* Number of args */
  sqlite3_value **apVal           /* Function arguments */
){
  Fts5Global *pGlobal = (Fts5Global*)sqlite3_user_data(pCtx);
  char buf[8];

  assert( nArg==0 );
  assert( sizeof(buf)>=sizeof(pGlobal) );
  memcpy(buf, (void*)&pGlobal, sizeof(pGlobal));
  sqlite3_result_blob(pCtx, buf, sizeof(pGlobal), SQLITE_TRANSIENT);
}

/*
** Implementation of fts5_source_id() function.
*/
static void fts5SourceIdFunc(
  sqlite3_context *pCtx,          /* Function call context */
  int nArg,                       /* Number of args */
  sqlite3_value **apVal           /* Function arguments */
){
  assert( nArg==0 );

  sqlite3_result_text(pCtx, "--FTS5-SOURCE-ID--", -1, SQLITE_TRANSIENT);
}

static int fts5Init(sqlite3 *db){
  static const sqlite3_module fts5Mod = {
    /* iVersion      */ 2,
    /* xCreate       */ fts5CreateMethod,







|



>












|


>







2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613

  sqlite3_free(pGlobal);
}

static void fts5Fts5Func(
  sqlite3_context *pCtx,          /* Function call context */
  int nArg,                       /* Number of args */
  sqlite3_value **apUnused        /* Function arguments */
){
  Fts5Global *pGlobal = (Fts5Global*)sqlite3_user_data(pCtx);
  char buf[8];
  UNUSED_PARAM2(nArg, apUnused);
  assert( nArg==0 );
  assert( sizeof(buf)>=sizeof(pGlobal) );
  memcpy(buf, (void*)&pGlobal, sizeof(pGlobal));
  sqlite3_result_blob(pCtx, buf, sizeof(pGlobal), SQLITE_TRANSIENT);
}

/*
** Implementation of fts5_source_id() function.
*/
static void fts5SourceIdFunc(
  sqlite3_context *pCtx,          /* Function call context */
  int nArg,                       /* Number of args */
  sqlite3_value **apUnused        /* Function arguments */
){
  assert( nArg==0 );
  UNUSED_PARAM2(nArg, apUnused);
  sqlite3_result_text(pCtx, "--FTS5-SOURCE-ID--", -1, SQLITE_TRANSIENT);
}

static int fts5Init(sqlite3 *db){
  static const sqlite3_module fts5Mod = {
    /* iVersion      */ 2,
    /* xCreate       */ fts5CreateMethod,
Changes to ext/fts5/fts5_storage.c.
358
359
360
361
362
363
364
365
366
367
368
369

370
371
372
373
374
375
376
** Tokenization callback used when inserting tokens into the FTS index.
*/
static int fts5StorageInsertCallback(
  void *pContext,                 /* Pointer to Fts5InsertCtx object */
  int tflags,
  const char *pToken,             /* Buffer containing token */
  int nToken,                     /* Size of token in bytes */
  int iStart,                     /* Start offset of token */
  int iEnd                        /* End offset of token */
){
  Fts5InsertCtx *pCtx = (Fts5InsertCtx*)pContext;
  Fts5Index *pIdx = pCtx->pStorage->pIndex;

  if( (tflags & FTS5_TOKEN_COLOCATED)==0 || pCtx->szCol==0 ){
    pCtx->szCol++;
  }
  return sqlite3Fts5IndexWrite(pIdx, pCtx->iCol, pCtx->szCol-1, pToken, nToken);
}

/*







|
|



>







358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
** Tokenization callback used when inserting tokens into the FTS index.
*/
static int fts5StorageInsertCallback(
  void *pContext,                 /* Pointer to Fts5InsertCtx object */
  int tflags,
  const char *pToken,             /* Buffer containing token */
  int nToken,                     /* Size of token in bytes */
  int iUnused1,                   /* Start offset of token */
  int iUnused2                    /* End offset of token */
){
  Fts5InsertCtx *pCtx = (Fts5InsertCtx*)pContext;
  Fts5Index *pIdx = pCtx->pStorage->pIndex;
  UNUSED_PARAM2(iUnused1, iUnused2);
  if( (tflags & FTS5_TOKEN_COLOCATED)==0 || pCtx->szCol==0 ){
    pCtx->szCol++;
  }
  return sqlite3Fts5IndexWrite(pIdx, pCtx->iCol, pCtx->szCol-1, pToken, nToken);
}

/*
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809


810
811
812
813
814
815
816
** Tokenization callback used by integrity check.
*/
static int fts5StorageIntegrityCallback(
  void *pContext,                 /* Pointer to Fts5IntegrityCtx object */
  int tflags,
  const char *pToken,             /* Buffer containing token */
  int nToken,                     /* Size of token in bytes */
  int iStart,                     /* Start offset of token */
  int iEnd                        /* End offset of token */
){
  Fts5IntegrityCtx *pCtx = (Fts5IntegrityCtx*)pContext;
  Fts5Termset *pTermset = pCtx->pTermset;
  int bPresent;
  int ii;
  int rc = SQLITE_OK;
  int iPos;
  int iCol;



  if( (tflags & FTS5_TOKEN_COLOCATED)==0 || pCtx->szCol==0 ){
    pCtx->szCol++;
  }

  switch( pCtx->pConfig->eDetail ){
    case FTS5_DETAIL_FULL:







|
|








>
>







794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
** Tokenization callback used by integrity check.
*/
static int fts5StorageIntegrityCallback(
  void *pContext,                 /* Pointer to Fts5IntegrityCtx object */
  int tflags,
  const char *pToken,             /* Buffer containing token */
  int nToken,                     /* Size of token in bytes */
  int iUnused1,                   /* Start offset of token */
  int iUnused2                    /* End offset of token */
){
  Fts5IntegrityCtx *pCtx = (Fts5IntegrityCtx*)pContext;
  Fts5Termset *pTermset = pCtx->pTermset;
  int bPresent;
  int ii;
  int rc = SQLITE_OK;
  int iPos;
  int iCol;

  UNUSED_PARAM2(iUnused1, iUnused2);

  if( (tflags & FTS5_TOKEN_COLOCATED)==0 || pCtx->szCol==0 ){
    pCtx->szCol++;
  }

  switch( pCtx->pConfig->eDetail ){
    case FTS5_DETAIL_FULL:
Changes to ext/fts5/fts5_tokenize.c.
58
59
60
61
62
63
64
65
66
67
68
69
70

71
72
73
74
75
76
77
  sqlite3_free(p);
}

/*
** Create an "ascii" tokenizer.
*/
static int fts5AsciiCreate(
  void *pCtx, 
  const char **azArg, int nArg,
  Fts5Tokenizer **ppOut
){
  int rc = SQLITE_OK;
  AsciiTokenizer *p = 0;

  if( nArg%2 ){
    rc = SQLITE_ERROR;
  }else{
    p = sqlite3_malloc(sizeof(AsciiTokenizer));
    if( p==0 ){
      rc = SQLITE_NOMEM;
    }else{







|





>







58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
  sqlite3_free(p);
}

/*
** Create an "ascii" tokenizer.
*/
static int fts5AsciiCreate(
  void *pUnused, 
  const char **azArg, int nArg,
  Fts5Tokenizer **ppOut
){
  int rc = SQLITE_OK;
  AsciiTokenizer *p = 0;
  UNUSED_PARAM(pUnused);
  if( nArg%2 ){
    rc = SQLITE_ERROR;
  }else{
    p = sqlite3_malloc(sizeof(AsciiTokenizer));
    if( p==0 ){
      rc = SQLITE_NOMEM;
    }else{
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

/*
** Tokenize some text using the ascii tokenizer.
*/
static int fts5AsciiTokenize(
  Fts5Tokenizer *pTokenizer,
  void *pCtx,
  int flags,
  const char *pText, int nText,
  int (*xToken)(void*, int, const char*, int nToken, int iStart, int iEnd)
){
  AsciiTokenizer *p = (AsciiTokenizer*)pTokenizer;
  int rc = SQLITE_OK;
  int ie;
  int is = 0;

  char aFold[64];
  int nFold = sizeof(aFold);
  char *pFold = aFold;
  unsigned char *a = p->aTokenChar;



  while( is<nText && rc==SQLITE_OK ){
    int nByte;

    /* Skip any leading divider characters. */
    while( is<nText && ((pText[is]&0x80)==0 && a[(int)pText[is]]==0) ){
      is++;







|












>
>







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

/*
** Tokenize some text using the ascii tokenizer.
*/
static int fts5AsciiTokenize(
  Fts5Tokenizer *pTokenizer,
  void *pCtx,
  int iUnused,
  const char *pText, int nText,
  int (*xToken)(void*, int, const char*, int nToken, int iStart, int iEnd)
){
  AsciiTokenizer *p = (AsciiTokenizer*)pTokenizer;
  int rc = SQLITE_OK;
  int ie;
  int is = 0;

  char aFold[64];
  int nFold = sizeof(aFold);
  char *pFold = aFold;
  unsigned char *a = p->aTokenChar;

  UNUSED_PARAM(iUnused);

  while( is<nText && rc==SQLITE_OK ){
    int nByte;

    /* Skip any leading divider characters. */
    while( is<nText && ((pText[is]&0x80)==0 && a[(int)pText[is]]==0) ){
      is++;
319
320
321
322
323
324
325
326
327
328
329
330
331


332
333
334
335
336
337
338
  return;
}

/*
** Create a "unicode61" tokenizer.
*/
static int fts5UnicodeCreate(
  void *pCtx, 
  const char **azArg, int nArg,
  Fts5Tokenizer **ppOut
){
  int rc = SQLITE_OK;             /* Return code */
  Unicode61Tokenizer *p = 0;      /* New tokenizer object */ 



  if( nArg%2 ){
    rc = SQLITE_ERROR;
  }else{
    p = (Unicode61Tokenizer*)sqlite3_malloc(sizeof(Unicode61Tokenizer));
    if( p ){
      int i;







|





>
>







322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
  return;
}

/*
** Create a "unicode61" tokenizer.
*/
static int fts5UnicodeCreate(
  void *pUnused, 
  const char **azArg, int nArg,
  Fts5Tokenizer **ppOut
){
  int rc = SQLITE_OK;             /* Return code */
  Unicode61Tokenizer *p = 0;      /* New tokenizer object */ 

  UNUSED_PARAM(pUnused);

  if( nArg%2 ){
    rc = SQLITE_ERROR;
  }else{
    p = (Unicode61Tokenizer*)sqlite3_malloc(sizeof(Unicode61Tokenizer));
    if( p ){
      int i;
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403


404
405
406
407
408
409
410
  assert( (sqlite3Fts5UnicodeIsalnum(iCode) & 0xFFFFFFFE)==0 );
  return sqlite3Fts5UnicodeIsalnum(iCode) ^ fts5UnicodeIsException(p, iCode);
}

static int fts5UnicodeTokenize(
  Fts5Tokenizer *pTokenizer,
  void *pCtx,
  int flags,
  const char *pText, int nText,
  int (*xToken)(void*, int, const char*, int nToken, int iStart, int iEnd)
){
  Unicode61Tokenizer *p = (Unicode61Tokenizer*)pTokenizer;
  int rc = SQLITE_OK;
  unsigned char *a = p->aTokenChar;

  unsigned char *zTerm = (unsigned char*)&pText[nText];
  unsigned char *zCsr = (unsigned char *)pText;

  /* Output buffer */
  char *aFold = p->aFold;
  int nFold = p->nFold;
  const char *pEnd = &aFold[nFold-6];



  /* Each iteration of this loop gobbles up a contiguous run of separators,
  ** then the next token.  */
  while( rc==SQLITE_OK ){
    int iCode;                    /* non-ASCII codepoint read from input */
    char *zOut = aFold;
    int is;







|














>
>







387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
  assert( (sqlite3Fts5UnicodeIsalnum(iCode) & 0xFFFFFFFE)==0 );
  return sqlite3Fts5UnicodeIsalnum(iCode) ^ fts5UnicodeIsException(p, iCode);
}

static int fts5UnicodeTokenize(
  Fts5Tokenizer *pTokenizer,
  void *pCtx,
  int iUnused,
  const char *pText, int nText,
  int (*xToken)(void*, int, const char*, int nToken, int iStart, int iEnd)
){
  Unicode61Tokenizer *p = (Unicode61Tokenizer*)pTokenizer;
  int rc = SQLITE_OK;
  unsigned char *a = p->aTokenChar;

  unsigned char *zTerm = (unsigned char*)&pText[nText];
  unsigned char *zCsr = (unsigned char *)pText;

  /* Output buffer */
  char *aFold = p->aFold;
  int nFold = p->nFold;
  const char *pEnd = &aFold[nFold-6];

  UNUSED_PARAM(iUnused);

  /* Each iteration of this loop gobbles up a contiguous run of separators,
  ** then the next token.  */
  while( rc==SQLITE_OK ){
    int iCode;                    /* non-ASCII codepoint read from input */
    char *zOut = aFold;
    int is;
Changes to ext/fts5/fts5_vocab.c.
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248


249
250
251
252
253
254
255
  return fts5VocabInitVtab(db, pAux, argc, argv, ppVtab, pzErr);
}

/* 
** Implementation of the xBestIndex method.
*/
static int fts5VocabBestIndexMethod(
  sqlite3_vtab *pVTab, 
  sqlite3_index_info *pInfo
){
  int i;
  int iTermEq = -1;
  int iTermGe = -1;
  int iTermLe = -1;
  int idxNum = 0;
  int nArg = 0;



  for(i=0; i<pInfo->nConstraint; i++){
    struct sqlite3_index_constraint *p = &pInfo->aConstraint[i];
    if( p->usable==0 ) continue;
    if( p->iColumn==0 ){          /* term column */
      if( p->op==SQLITE_INDEX_CONSTRAINT_EQ ) iTermEq = i;
      if( p->op==SQLITE_INDEX_CONSTRAINT_LE ) iTermLe = i;







|








>
>







233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
  return fts5VocabInitVtab(db, pAux, argc, argv, ppVtab, pzErr);
}

/* 
** Implementation of the xBestIndex method.
*/
static int fts5VocabBestIndexMethod(
  sqlite3_vtab *pUnused,
  sqlite3_index_info *pInfo
){
  int i;
  int iTermEq = -1;
  int iTermGe = -1;
  int iTermLe = -1;
  int idxNum = 0;
  int nArg = 0;

  UNUSED_PARAM(pUnused);

  for(i=0; i<pInfo->nConstraint; i++){
    struct sqlite3_index_constraint *p = &pInfo->aConstraint[i];
    if( p->usable==0 ) continue;
    if( p->iColumn==0 ){          /* term column */
      if( p->op==SQLITE_INDEX_CONSTRAINT_EQ ) iTermEq = i;
      if( p->op==SQLITE_INDEX_CONSTRAINT_LE ) iTermLe = i;
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505


506
507
508
509
510
511
512

/*
** This is the xFilter implementation for the virtual table.
*/
static int fts5VocabFilterMethod(
  sqlite3_vtab_cursor *pCursor,   /* The cursor used for this query */
  int idxNum,                     /* Strategy index */
  const char *idxStr,             /* Unused */
  int nVal,                       /* Number of elements in apVal */
  sqlite3_value **apVal           /* Arguments for the indexing scheme */
){
  Fts5VocabCursor *pCsr = (Fts5VocabCursor*)pCursor;
  int rc = SQLITE_OK;

  int iVal = 0;
  int f = FTS5INDEX_QUERY_SCAN;
  const char *zTerm = 0;
  int nTerm = 0;

  sqlite3_value *pEq = 0;
  sqlite3_value *pGe = 0;
  sqlite3_value *pLe = 0;



  fts5VocabResetCursor(pCsr);
  if( idxNum & FTS5_VOCAB_TERM_EQ ) pEq = apVal[iVal++];
  if( idxNum & FTS5_VOCAB_TERM_GE ) pGe = apVal[iVal++];
  if( idxNum & FTS5_VOCAB_TERM_LE ) pLe = apVal[iVal++];

  if( pEq ){







|
|













>
>







486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516

/*
** This is the xFilter implementation for the virtual table.
*/
static int fts5VocabFilterMethod(
  sqlite3_vtab_cursor *pCursor,   /* The cursor used for this query */
  int idxNum,                     /* Strategy index */
  const char *zUnused,            /* Unused */
  int nUnused,                    /* Number of elements in apVal */
  sqlite3_value **apVal           /* Arguments for the indexing scheme */
){
  Fts5VocabCursor *pCsr = (Fts5VocabCursor*)pCursor;
  int rc = SQLITE_OK;

  int iVal = 0;
  int f = FTS5INDEX_QUERY_SCAN;
  const char *zTerm = 0;
  int nTerm = 0;

  sqlite3_value *pEq = 0;
  sqlite3_value *pGe = 0;
  sqlite3_value *pLe = 0;

  UNUSED_PARAM2(zUnused, nUnused);

  fts5VocabResetCursor(pCsr);
  if( idxNum & FTS5_VOCAB_TERM_EQ ) pEq = apVal[iVal++];
  if( idxNum & FTS5_VOCAB_TERM_GE ) pGe = apVal[iVal++];
  if( idxNum & FTS5_VOCAB_TERM_LE ) pLe = apVal[iVal++];

  if( pEq ){
Changes to ext/fts5/fts5parse.y.
29
30
31
32
33
34
35
36

37
38
39
40
41
42
43
//
%syntax_error {
  sqlite3Fts5ParseError(
    pParse, "fts5: syntax error near \"%.*s\"",TOKEN.n,TOKEN.p
  );
}
%stack_overflow {
  assert( 0 );

}

// The name of the generated procedure that implements the parser
// is as follows:
%name sqlite3Fts5Parser

// The following text is included near the beginning of the C source







|
>







29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
//
%syntax_error {
  sqlite3Fts5ParseError(
    pParse, "fts5: syntax error near \"%.*s\"",TOKEN.n,TOKEN.p
  );
}
%stack_overflow {
  UNUSED_PARAM(yypMinor); /* Silence some compiler warnings */
  sqlite3Fts5ParseError(pParse, "fts5: parser stack overflow");
}

// The name of the generated procedure that implements the parser
// is as follows:
%name sqlite3Fts5Parser

// The following text is included near the beginning of the C source
Changes to ext/fts5/test/fts5simple.test.
394
395
396
397
398
399
400










401
402
403
}

#-------------------------------------------------------------------------
do_execsql_test 18.1 {
  CREATE VIRTUAL TABLE x4 USING fts5(x);
  SELECT rowid FROM x4('""');
}











finish_test








>
>
>
>
>
>
>
>
>
>



394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
}

#-------------------------------------------------------------------------
do_execsql_test 18.1 {
  CREATE VIRTUAL TABLE x4 USING fts5(x);
  SELECT rowid FROM x4('""');
}

#-------------------------------------------------------------------------
reset_db
do_execsql_test 19.1 {
  CREATE VIRTUAL TABLE x1 USING fts5(a,b,c);
}

do_catchsql_test 19.2 {
  SELECT * FROM x1 WHERE x1 MATCH 'c0 AND (c1 AND (c2 AND (c3 AND (c4 AND (c5 AND (c6 AND (c7 AND (c8 AND (c9 AND (c10 AND (c11 AND (c12 AND (c13 AND (c14 AND (c15 AND (c16 AND (c17 AND (c18 AND (c19 AND (c20 AND (c21 AND (c22 AND (c23 AND (c24 AND (c25 AND (c26 AND (c27 AND (c28 AND (c29 AND (c30 AND (c31 AND (c32 AND (c33 AND (c34 AND (c35 AND (c36 AND (c37 AND (c38 AND (c39 AND (c40 AND (c41 AND (c42 AND (c43 AND (c44 AND (c45 AND (c46 AND (c47 AND (c48 AND (c49 AND (c50 AND (c51 AND (c52 AND (c53 AND (c54 AND (c55 AND (c56 AND (c57 AND (c58 AND (c59 AND (c60 AND (c61 AND (c62 AND (c63 AND (c64 AND (c65 AND (c66 AND (c67 AND (c68 AND (c69 AND (c70 AND (c71 AND (c72 AND (c73 AND (c74 AND (c75 AND (c76 AND (c77 AND (c78 AND (c79 AND (c80 AND (c81 AND (c82 AND (c83 AND (c84 AND (c85 AND (c86 AND (c87 AND (c88 AND (c89 AND (c90 AND (c91 AND (c92 AND (c93 AND (c94 AND (c95 AND (c96 AND (c97 AND (c98 AND (c99 AND (c100 AND (c101 AND (c102 AND (c103 AND (c104 AND (c105 AND (c106 AND (c107 AND (c108 AND (c109 AND (c110 AND (c111 AND (c112 AND (c113 AND (c114 AND (c115 AND (c116 AND (c117 AND (c118 AND (c119 AND (c120 AND (c121 AND (c122 AND (c123 AND (c124 AND (c125 AND (c126 AND (c127 AND (c128 AND (c129 AND (c130 AND (c131 AND (c132 AND (c133 AND (c134 AND (c135 AND (c136 AND (c137 AND (c138 AND (c139 AND (c140 AND (c141 AND (c142 AND (c143 AND (c144 AND (c145 AND (c146 AND (c147 AND (c148 AND (c149 AND (c150 AND (c151 AND (c152 AND (c153 AND (c154 AND (c155 AND (c156 AND (c157 AND (c158 AND (c159 AND (c160 AND (c161 AND (c162 AND (c163 AND (c164 AND (c165 AND (c166 AND (c167 AND (c168 AND (c169 AND (c170 AND (c171 AND (c172 AND (c173 AND (c174 AND (c175 AND (c176 AND (c177 AND (c178 AND (c179 AND (c180 AND (c181 AND (c182 AND (c183 AND (c184 AND (c185 AND (c186 AND (c187 AND (c188 AND (c189 AND (c190 AND (c191 AND (c192 AND (c193 AND (c194 AND (c195 AND (c196 AND (c197 AND (c198 AND (c199 AND c200)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))';
} {1 {fts5: parser stack overflow}}

finish_test