/ Check-in [9b19b847]
Login

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

Overview
Comment:Fix all known instances of 'repeated the' style typos in comments. No changes to code.
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1:9b19b847533f944f289d93dcdba29c0d67bf251c
User & Date: mistachkin 2012-08-25 10:01:29
Context
2012-08-27
14:39
Fix an incorrect assert in the virtual table logic - it could fire following an I/O error after sqlite3_close_v2() was added. check-in: 4ccc18e9 user: drh tags: trunk
2012-08-25
10:01
Fix all known instances of 'repeated the' style typos in comments. No changes to code. check-in: 9b19b847 user: mistachkin tags: trunk
02:11
Fix a harmless compiler warning. check-in: 929b5184 user: drh tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to doc/lemon.html.

477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
     of the shift.  No parsing conflict is reported.
<li> If the precedence of the token it be shifted is less than the
     precedence of the rule to reduce, then resolve in favor of the
     reduce action.  No parsing conflict is reported.
<li> If the precedences are the same and the shift token is
     right-associative, then resolve in favor of the shift.
     No parsing conflict is reported.
<li> If the precedences are the same the the shift token is
     left-associative, then resolve in favor of the reduce.
     No parsing conflict is reported.
<li> Otherwise, resolve the conflict by doing the shift and
     report the parsing conflict.
</ul>
Reduce-reduce conflicts are resolved this way:
<ul>







|







477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
     of the shift.  No parsing conflict is reported.
<li> If the precedence of the token it be shifted is less than the
     precedence of the rule to reduce, then resolve in favor of the
     reduce action.  No parsing conflict is reported.
<li> If the precedences are the same and the shift token is
     right-associative, then resolve in favor of the shift.
     No parsing conflict is reported.
<li> If the precedences are the same the shift token is
     left-associative, then resolve in favor of the reduce.
     No parsing conflict is reported.
<li> Otherwise, resolve the conflict by doing the shift and
     report the parsing conflict.
</ul>
Reduce-reduce conflicts are resolved this way:
<ul>

Changes to doc/pager-invariants.txt.

40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
     being deleted, truncated, or zeroed.
 
 (6) If a master journal file is used, then all writes to the database file
     are synced prior to the master journal being deleted.
 
 *** Definition: Two databases (or the same database at two points it time)
     are said to be "logically equivalent" if they give the same answer to
     all queries.  Note in particular the the content of freelist leaf
     pages can be changed arbitarily without effecting the logical equivalence
     of the database.
 
 (7) At any time, if any subset, including the empty set and the total set,
     of the unsynced changes to a rollback journal are removed and the 
     journal is rolled back, the resulting database file will be logical
     equivalent to the database file at the beginning of the transaction.







|







40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
     being deleted, truncated, or zeroed.
 
 (6) If a master journal file is used, then all writes to the database file
     are synced prior to the master journal being deleted.
 
 *** Definition: Two databases (or the same database at two points it time)
     are said to be "logically equivalent" if they give the same answer to
     all queries.  Note in particular the content of freelist leaf
     pages can be changed arbitarily without effecting the logical equivalence
     of the database.
 
 (7) At any time, if any subset, including the empty set and the total set,
     of the unsynced changes to a rollback journal are removed and the 
     journal is rolled back, the resulting database file will be logical
     equivalent to the database file at the beginning of the transaction.

Changes to ext/fts2/fts2.c.

5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
*/
/* TODO(shess) This "solution" is not satisfactory.  Really, there
** should be check-in function for all statement handles which
** arranges to call sqlite3_reset().  This most likely will require
** modification to control flow all over the place, though, so for now
** just punt.
**
** Note the the current system assumes that segment merges will run to
** completion, which is why this particular probably hasn't arisen in
** this case.  Probably a brittle assumption.
*/
static int leavesReaderReset(LeavesReader *pReader){
  return sqlite3_reset(pReader->pStmt);
}








|







5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
*/
/* TODO(shess) This "solution" is not satisfactory.  Really, there
** should be check-in function for all statement handles which
** arranges to call sqlite3_reset().  This most likely will require
** modification to control flow all over the place, though, so for now
** just punt.
**
** Note the current system assumes that segment merges will run to
** completion, which is why this particular probably hasn't arisen in
** this case.  Probably a brittle assumption.
*/
static int leavesReaderReset(LeavesReader *pReader){
  return sqlite3_reset(pReader->pStmt);
}

Changes to ext/fts3/fts3_write.c.

2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
....
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
  assert( iIndex>=0 && iIndex<p->nIndex );

  rc = sqlite3Fts3SegReaderCursor(p, iLangid, iIndex, iLevel, 0, 0, 1, 0, &csr);
  if( rc!=SQLITE_OK || csr.nSegment==0 ) goto finished;

  if( iLevel==FTS3_SEGCURSOR_ALL ){
    /* This call is to merge all segments in the database to a single
    ** segment. The level of the new segment is equal to the the numerically 
    ** greatest segment level currently present in the database for this
    ** index. The idx of the new segment is always 0.  */
    if( csr.nSegment==1 ){
      rc = SQLITE_DONE;
      goto finished;
    }
    rc = fts3SegmentMaxLevel(p, iLangid, iIndex, &iNewLevel);
................................................................................
        memcpy(&pBlk->a[pBlk->n], &zTerm[nPrefix], nSuffix);
        pBlk->n += nSuffix;

        memcpy(pNode->key.a, zTerm, nTerm);
        pNode->key.n = nTerm;
      }
    }else{
      /* Otherwise, flush the the current node of layer iLayer to disk.
      ** Then allocate a new, empty sibling node. The key will be written
      ** into the parent of this node. */
      rc = fts3WriteSegment(p, pNode->iBlock, pNode->block.a, pNode->block.n);

      assert( pNode->block.nAlloc>=p->nNodeSize );
      pNode->block.a[0] = (char)iLayer;
      pNode->block.n = 1 + sqlite3Fts3PutVarint(&pNode->block.a[1], iPtr+1);







|







 







|







2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
....
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
  assert( iIndex>=0 && iIndex<p->nIndex );

  rc = sqlite3Fts3SegReaderCursor(p, iLangid, iIndex, iLevel, 0, 0, 1, 0, &csr);
  if( rc!=SQLITE_OK || csr.nSegment==0 ) goto finished;

  if( iLevel==FTS3_SEGCURSOR_ALL ){
    /* This call is to merge all segments in the database to a single
    ** segment. The level of the new segment is equal to the numerically
    ** greatest segment level currently present in the database for this
    ** index. The idx of the new segment is always 0.  */
    if( csr.nSegment==1 ){
      rc = SQLITE_DONE;
      goto finished;
    }
    rc = fts3SegmentMaxLevel(p, iLangid, iIndex, &iNewLevel);
................................................................................
        memcpy(&pBlk->a[pBlk->n], &zTerm[nPrefix], nSuffix);
        pBlk->n += nSuffix;

        memcpy(pNode->key.a, zTerm, nTerm);
        pNode->key.n = nTerm;
      }
    }else{
      /* Otherwise, flush the current node of layer iLayer to disk.
      ** Then allocate a new, empty sibling node. The key will be written
      ** into the parent of this node. */
      rc = fts3WriteSegment(p, pNode->iBlock, pNode->block.a, pNode->block.n);

      assert( pNode->block.nAlloc>=p->nNodeSize );
      pNode->block.a[0] = (char)iLayer;
      pNode->block.n = 1 + sqlite3Fts3PutVarint(&pNode->block.a[1], iPtr+1);

Changes to src/btree.c.

6079
6080
6081
6082
6083
6084
6085
6086
6087
6088
6089
6090
6091
6092
6093
  szCell = (u16*)&apCell[nMaxCells];
  aSpace1 = (u8*)&szCell[nMaxCells];
  assert( EIGHT_BYTE_ALIGNMENT(aSpace1) );

  /*
  ** Load pointers to all cells on sibling pages and the divider cells
  ** into the local apCell[] array.  Make copies of the divider cells
  ** into space obtained from aSpace1[] and remove the the divider Cells
  ** from pParent.
  **
  ** If the siblings are on leaf pages, then the child pointers of the
  ** divider cells are stripped from the cells before they are copied
  ** into aSpace1[].  In this way, all cells in apCell[] are without
  ** child pointers.  If siblings are not leaves, then all cell in
  ** apCell[] include child pointers.  Either way, all cells in apCell[]







|







6079
6080
6081
6082
6083
6084
6085
6086
6087
6088
6089
6090
6091
6092
6093
  szCell = (u16*)&apCell[nMaxCells];
  aSpace1 = (u8*)&szCell[nMaxCells];
  assert( EIGHT_BYTE_ALIGNMENT(aSpace1) );

  /*
  ** Load pointers to all cells on sibling pages and the divider cells
  ** into the local apCell[] array.  Make copies of the divider cells
  ** into space obtained from aSpace1[] and remove the divider cells
  ** from pParent.
  **
  ** If the siblings are on leaf pages, then the child pointers of the
  ** divider cells are stripped from the cells before they are copied
  ** into aSpace1[].  In this way, all cells in apCell[] are without
  ** child pointers.  If siblings are not leaves, then all cell in
  ** apCell[] include child pointers.  Either way, all cells in apCell[]

Changes to src/build.c.

2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
    */
    assert( pName1 && pName2 );
    iDb = sqlite3TwoPartName(pParse, pName1, pName2, &pName);
    if( iDb<0 ) goto exit_create_index;
    assert( pName && pName->z );

#ifndef SQLITE_OMIT_TEMPDB
    /* If the index name was unqualified, check if the the table
    ** is a temp table. If so, set the database to 1. Do not do this
    ** if initialising a database schema.
    */
    if( !db->init.busy ){
      pTab = sqlite3SrcListLookup(pParse, pTblName);
      if( pName2->n==0 && pTab && pTab->pSchema==db->aDb[1].pSchema ){
        iDb = 1;







|







2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
    */
    assert( pName1 && pName2 );
    iDb = sqlite3TwoPartName(pParse, pName1, pName2, &pName);
    if( iDb<0 ) goto exit_create_index;
    assert( pName && pName->z );

#ifndef SQLITE_OMIT_TEMPDB
    /* If the index name was unqualified, check if the table
    ** is a temp table. If so, set the database to 1. Do not do this
    ** if initialising a database schema.
    */
    if( !db->init.busy ){
      pTab = sqlite3SrcListLookup(pParse, pTblName);
      if( pName2->n==0 && pTab && pTab->pSchema==db->aDb[1].pSchema ){
        iDb = 1;

Changes to src/insert.c.

1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
        sqlite3HaltConstraint(
          pParse, onError, "PRIMARY KEY must be unique", P4_STATIC);
        break;
      }
      case OE_Replace: {
        /* If there are DELETE triggers on this table and the
        ** recursive-triggers flag is set, call GenerateRowDelete() to
        ** remove the conflicting row from the the table. This will fire
        ** the triggers and remove both the table and index b-tree entries.
        **
        ** Otherwise, if there are no triggers or the recursive-triggers
        ** flag is not set, but the table has one or more indexes, call 
        ** GenerateRowIndexDelete(). This removes the index b-tree entries 
        ** only. The table b-tree entry will be replaced by the new entry 
        ** when it is inserted.  







|







1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
        sqlite3HaltConstraint(
          pParse, onError, "PRIMARY KEY must be unique", P4_STATIC);
        break;
      }
      case OE_Replace: {
        /* If there are DELETE triggers on this table and the
        ** recursive-triggers flag is set, call GenerateRowDelete() to
        ** remove the conflicting row from the table. This will fire
        ** the triggers and remove both the table and index b-tree entries.
        **
        ** Otherwise, if there are no triggers or the recursive-triggers
        ** flag is not set, but the table has one or more indexes, call 
        ** GenerateRowIndexDelete(). This removes the index b-tree entries 
        ** only. The table b-tree entry will be replaced by the new entry 
        ** when it is inserted.  

Changes to src/os_unix.c.

1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
....
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
** set. It logs a message using sqlite3_log() containing the current value of
** errno and, if possible, the human-readable equivalent from strerror() or
** strerror_r().
**
** The first argument passed to the macro should be the error code that
** will be returned to SQLite (e.g. SQLITE_IOERR_DELETE, SQLITE_CANTOPEN). 
** The two subsequent arguments should be the name of the OS function that
** failed (e.g. "unlink", "open") and the the associated file-system path,
** if any.
*/
#define unixLogError(a,b,c)     unixLogErrorAtLine(a,b,c,__LINE__)
static int unixLogErrorAtLine(
  int errcode,                    /* SQLite error code */
  const char *zFunc,              /* Name of OS function that failed */
  const char *zPath,              /* File path associated with error */
................................................................................
  */ 
#if SQLITE_THREADSAFE && defined(HAVE_STRERROR_R)
  char aErr[80];
  memset(aErr, 0, sizeof(aErr));
  zErr = aErr;

  /* If STRERROR_R_CHAR_P (set by autoconf scripts) or __USE_GNU is defined,
  ** assume that the system provides the the GNU version of strerror_r() that 
  ** returns a pointer to a buffer containing the error message. That pointer 
  ** may point to aErr[], or it may point to some static storage somewhere. 
  ** Otherwise, assume that the system provides the POSIX version of 
  ** strerror_r(), which always writes an error message into aErr[].
  **
  ** If the code incorrectly assumes that it is the POSIX version that is
  ** available, the error message will often be an empty string. Not a







|







 







|







1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
....
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
** set. It logs a message using sqlite3_log() containing the current value of
** errno and, if possible, the human-readable equivalent from strerror() or
** strerror_r().
**
** The first argument passed to the macro should be the error code that
** will be returned to SQLite (e.g. SQLITE_IOERR_DELETE, SQLITE_CANTOPEN). 
** The two subsequent arguments should be the name of the OS function that
** failed (e.g. "unlink", "open") and the associated file-system path,
** if any.
*/
#define unixLogError(a,b,c)     unixLogErrorAtLine(a,b,c,__LINE__)
static int unixLogErrorAtLine(
  int errcode,                    /* SQLite error code */
  const char *zFunc,              /* Name of OS function that failed */
  const char *zPath,              /* File path associated with error */
................................................................................
  */ 
#if SQLITE_THREADSAFE && defined(HAVE_STRERROR_R)
  char aErr[80];
  memset(aErr, 0, sizeof(aErr));
  zErr = aErr;

  /* If STRERROR_R_CHAR_P (set by autoconf scripts) or __USE_GNU is defined,
  ** assume that the system provides the GNU version of strerror_r() that
  ** returns a pointer to a buffer containing the error message. That pointer 
  ** may point to aErr[], or it may point to some static storage somewhere. 
  ** Otherwise, assume that the system provides the POSIX version of 
  ** strerror_r(), which always writes an error message into aErr[].
  **
  ** If the code incorrectly assumes that it is the POSIX version that is
  ** available, the error message will often be an empty string. Not a

Changes to src/os_win.c.

1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
** It logs a message using sqlite3_log() containing the current value of
** error code and, if possible, the human-readable equivalent from 
** FormatMessage.
**
** The first argument passed to the macro should be the error code that
** will be returned to SQLite (e.g. SQLITE_IOERR_DELETE, SQLITE_CANTOPEN). 
** The two subsequent arguments should be the name of the OS function that
** failed and the the associated file-system path, if any.
*/
#define winLogError(a,b,c,d)   winLogErrorAtLine(a,b,c,d,__LINE__)
static int winLogErrorAtLine(
  int errcode,                    /* SQLite error code */
  DWORD lastErrno,                /* Win32 last error */
  const char *zFunc,              /* Name of OS function that failed */
  const char *zPath,              /* File path associated with error */







|







1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
** It logs a message using sqlite3_log() containing the current value of
** error code and, if possible, the human-readable equivalent from 
** FormatMessage.
**
** The first argument passed to the macro should be the error code that
** will be returned to SQLite (e.g. SQLITE_IOERR_DELETE, SQLITE_CANTOPEN). 
** The two subsequent arguments should be the name of the OS function that
** failed and the associated file-system path, if any.
*/
#define winLogError(a,b,c,d)   winLogErrorAtLine(a,b,c,d,__LINE__)
static int winLogErrorAtLine(
  int errcode,                    /* SQLite error code */
  DWORD lastErrno,                /* Win32 last error */
  const char *zFunc,              /* Name of OS function that failed */
  const char *zPath,              /* File path associated with error */

Changes to src/pager.c.

71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
....
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
**     being deleted, truncated, or zeroed.
** 
** (6) If a master journal file is used, then all writes to the database file
**     are synced prior to the master journal being deleted.
** 
** Definition: Two databases (or the same database at two points it time)
** are said to be "logically equivalent" if they give the same answer to
** all queries.  Note in particular the the content of freelist leaf
** pages can be changed arbitarily without effecting the logical equivalence
** of the database.
** 
** (7) At any time, if any subset, including the empty set and the total set,
**     of the unsynced changes to a rollback journal are removed and the 
**     journal is rolled back, the resulting database file will be logical
**     equivalent to the database file at the beginning of the transaction.
................................................................................
/*
** Sync the journal. In other words, make sure all the pages that have
** been written to the journal have actually reached the surface of the
** disk and can be restored in the event of a hot-journal rollback.
**
** If the Pager.noSync flag is set, then this function is a no-op.
** Otherwise, the actions required depend on the journal-mode and the 
** device characteristics of the the file-system, as follows:
**
**   * If the journal file is an in-memory journal file, no action need
**     be taken.
**
**   * Otherwise, if the device does not support the SAFE_APPEND property,
**     then the nRec field of the most recently written journal header
**     is updated to contain the number of journal records that have







|







 







|







71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
....
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
**     being deleted, truncated, or zeroed.
** 
** (6) If a master journal file is used, then all writes to the database file
**     are synced prior to the master journal being deleted.
** 
** Definition: Two databases (or the same database at two points it time)
** are said to be "logically equivalent" if they give the same answer to
** all queries.  Note in particular the content of freelist leaf
** pages can be changed arbitarily without effecting the logical equivalence
** of the database.
** 
** (7) At any time, if any subset, including the empty set and the total set,
**     of the unsynced changes to a rollback journal are removed and the 
**     journal is rolled back, the resulting database file will be logical
**     equivalent to the database file at the beginning of the transaction.
................................................................................
/*
** Sync the journal. In other words, make sure all the pages that have
** been written to the journal have actually reached the surface of the
** disk and can be restored in the event of a hot-journal rollback.
**
** If the Pager.noSync flag is set, then this function is a no-op.
** Otherwise, the actions required depend on the journal-mode and the 
** device characteristics of the file-system, as follows:
**
**   * If the journal file is an in-memory journal file, no action need
**     be taken.
**
**   * Otherwise, if the device does not support the SAFE_APPEND property,
**     then the nRec field of the most recently written journal header
**     is updated to contain the number of journal records that have

Changes to src/rowset.c.

436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
    return 1;
  }else{
    return 0;
  }
}

/*
** Check to see if element iRowid was inserted into the the rowset as
** part of any insert batch prior to iBatch.  Return 1 or 0.
**
** If this is the first test of a new batch and if there exist entires
** on pRowSet->pEntry, then sort those entires into the forest at
** pRowSet->pForest so that they can be tested.
*/
int sqlite3RowSetTest(RowSet *pRowSet, u8 iBatch, sqlite3_int64 iRowid){







|







436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
    return 1;
  }else{
    return 0;
  }
}

/*
** Check to see if element iRowid was inserted into the rowset as
** part of any insert batch prior to iBatch.  Return 1 or 0.
**
** If this is the first test of a new batch and if there exist entires
** on pRowSet->pEntry, then sort those entires into the forest at
** pRowSet->pForest so that they can be tested.
*/
int sqlite3RowSetTest(RowSet *pRowSet, u8 iBatch, sqlite3_int64 iRowid){

Changes to src/select.c.

1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
    sqlite3VdbeAddOp3(v, OP_Jump, j2+2, iContinue, j2+2);
    sqlite3VdbeJumpHere(v, j1);
    sqlite3ExprCodeCopy(pParse, pIn->iSdst, regPrev+1, pIn->nSdst);
    sqlite3VdbeAddOp2(v, OP_Integer, 1, regPrev);
  }
  if( pParse->db->mallocFailed ) return 0;

  /* Suppress the the first OFFSET entries if there is an OFFSET clause
  */
  codeOffset(v, p, iContinue);

  switch( pDest->eDest ){
    /* Store the result as data using a unique key.
    */
    case SRT_Table:







|







1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
    sqlite3VdbeAddOp3(v, OP_Jump, j2+2, iContinue, j2+2);
    sqlite3VdbeJumpHere(v, j1);
    sqlite3ExprCodeCopy(pParse, pIn->iSdst, regPrev+1, pIn->nSdst);
    sqlite3VdbeAddOp2(v, OP_Integer, 1, regPrev);
  }
  if( pParse->db->mallocFailed ) return 0;

  /* Suppress the first OFFSET entries if there is an OFFSET clause
  */
  codeOffset(v, p, iContinue);

  switch( pDest->eDest ){
    /* Store the result as data using a unique key.
    */
    case SRT_Table:

Changes to src/sqlite.h.in.

508
509
510
511
512
513
514
515
516
517
518
519
520
521
522

/* Reserved:                         0x00F00000 */

/*
** CAPI3REF: Device Characteristics
**
** The xDeviceCharacteristics method of the [sqlite3_io_methods]
** object returns an integer which is a vector of the these
** bit values expressing I/O characteristics of the mass storage
** device that holds the file that the [sqlite3_io_methods]
** refers to.
**
** The SQLITE_IOCAP_ATOMIC property means that all writes of
** any size are atomic.  The SQLITE_IOCAP_ATOMICnnn values
** mean that writes of blocks that are nnn bytes in size and







|







508
509
510
511
512
513
514
515
516
517
518
519
520
521
522

/* Reserved:                         0x00F00000 */

/*
** CAPI3REF: Device Characteristics
**
** The xDeviceCharacteristics method of the [sqlite3_io_methods]
** object returns an integer which is a vector of these
** bit values expressing I/O characteristics of the mass storage
** device that holds the file that the [sqlite3_io_methods]
** refers to.
**
** The SQLITE_IOCAP_ATOMIC property means that all writes of
** any size are atomic.  The SQLITE_IOCAP_ATOMICnnn values
** mean that writes of blocks that are nnn bytes in size and

Changes to src/sqliteInt.h.

1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
    sqlite3_index_info *pVtabIdx;  /* Virtual table index to use */
  } u;
};

/*
** For each nested loop in a WHERE clause implementation, the WhereInfo
** structure contains a single instance of this structure.  This structure
** is intended to be private the the where.c module and should not be
** access or modified by other modules.
**
** The pIdxInfo field is used to help pick the best index on a
** virtual table.  The pIdxInfo pointer contains indexing
** information for the i-th table in the FROM clause before reordering.
** All the pIdxInfo pointers are freed by whereInfoFree() in where.c.
** All other information in the i-th WhereLevel object for the i-th table







|







1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
    sqlite3_index_info *pVtabIdx;  /* Virtual table index to use */
  } u;
};

/*
** For each nested loop in a WHERE clause implementation, the WhereInfo
** structure contains a single instance of this structure.  This structure
** is intended to be private to the where.c module and should not be
** access or modified by other modules.
**
** The pIdxInfo field is used to help pick the best index on a
** virtual table.  The pIdxInfo pointer contains indexing
** information for the i-th table in the FROM clause before reordering.
** All the pIdxInfo pointers are freed by whereInfoFree() in where.c.
** All other information in the i-th WhereLevel object for the i-th table

Changes to src/test4.c.

5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
** a legal notice, here is a blessing:
**
**    May you do good and not evil.
**    May you find forgiveness for yourself and forgive others.
**    May you share freely, never taking more than you give.
**
*************************************************************************
** Code for testing the the SQLite library in a multithreaded environment.
*/
#include "sqliteInt.h"
#include "tcl.h"
#if SQLITE_OS_UNIX && SQLITE_THREADSAFE
#include <stdlib.h>
#include <string.h>
#include <pthread.h>







|







5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
** a legal notice, here is a blessing:
**
**    May you do good and not evil.
**    May you find forgiveness for yourself and forgive others.
**    May you share freely, never taking more than you give.
**
*************************************************************************
** Code for testing the SQLite library in a multithreaded environment.
*/
#include "sqliteInt.h"
#include "tcl.h"
#if SQLITE_OS_UNIX && SQLITE_THREADSAFE
#include <stdlib.h>
#include <string.h>
#include <pthread.h>

Changes to src/test_vfstrace.c.

41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
**
** The vfstrace_register() function creates a new "shim" VFS named by
** the zTraceName parameter.  A "shim" VFS is an SQLite backend that does
** not really perform the duties of a true backend, but simply filters or
** interprets VFS calls before passing them off to another VFS which does
** the actual work.  In this case the other VFS - the one that does the
** real work - is identified by the second parameter, zOldVfsName.  If
** the the 2nd parameter is NULL then the default VFS is used.  The common
** case is for the 2nd parameter to be NULL.
**
** The third and fourth parameters are the pointer to the output function
** and the second argument to the output function.  For the SQLite
** command-line shell, when the -vfstrace option is used, these parameters
** are fputs and stderr, respectively.
**







|







41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
**
** The vfstrace_register() function creates a new "shim" VFS named by
** the zTraceName parameter.  A "shim" VFS is an SQLite backend that does
** not really perform the duties of a true backend, but simply filters or
** interprets VFS calls before passing them off to another VFS which does
** the actual work.  In this case the other VFS - the one that does the
** real work - is identified by the second parameter, zOldVfsName.  If
** the 2nd parameter is NULL then the default VFS is used.  The common
** case is for the 2nd parameter to be NULL.
**
** The third and fourth parameters are the pointer to the output function
** and the second argument to the output function.  For the SQLite
** command-line shell, when the -vfstrace option is used, these parameters
** are fputs and stderr, respectively.
**

Changes to src/trigger.c.

107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
    if( pName2->n>0 ){
      sqlite3ErrorMsg(pParse, "temporary trigger may not have qualified name");
      goto trigger_cleanup;
    }
    iDb = 1;
    pName = pName1;
  }else{
    /* Figure out the db that the the trigger will be created in */
    iDb = sqlite3TwoPartName(pParse, pName1, pName2, &pName);
    if( iDb<0 ){
      goto trigger_cleanup;
    }
  }
  if( !pTableName || db->mallocFailed ){
    goto trigger_cleanup;







|







107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
    if( pName2->n>0 ){
      sqlite3ErrorMsg(pParse, "temporary trigger may not have qualified name");
      goto trigger_cleanup;
    }
    iDb = 1;
    pName = pName1;
  }else{
    /* Figure out the db that the trigger will be created in */
    iDb = sqlite3TwoPartName(pParse, pName1, pName2, &pName);
    if( iDb<0 ){
      goto trigger_cleanup;
    }
  }
  if( !pTableName || db->mallocFailed ){
    goto trigger_cleanup;

Changes to src/vdbeaux.c.

770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
    pOp->p4.z = sqlite3DbStrNDup(p->db, zP4, n);
    pOp->p4type = P4_DYNAMIC;
  }
}

#ifndef NDEBUG
/*
** Change the comment on the the most recently coded instruction.  Or
** insert a No-op and add the comment to that new instruction.  This
** makes the code easier to read during debugging.  None of this happens
** in a production build.
*/
static void vdbeVComment(Vdbe *p, const char *zFormat, va_list ap){
  assert( p->nOp>0 || p->aOp==0 );
  assert( p->aOp==0 || p->aOp[p->nOp-1].zComment==0 || p->db->mallocFailed );







|







770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
    pOp->p4.z = sqlite3DbStrNDup(p->db, zP4, n);
    pOp->p4type = P4_DYNAMIC;
  }
}

#ifndef NDEBUG
/*
** Change the comment on the most recently coded instruction.  Or
** insert a No-op and add the comment to that new instruction.  This
** makes the code easier to read during debugging.  None of this happens
** in a production build.
*/
static void vdbeVComment(Vdbe *p, const char *zFormat, va_list ap){
  assert( p->nOp>0 || p->aOp==0 );
  assert( p->aOp==0 || p->aOp[p->nOp-1].zComment==0 || p->db->mallocFailed );

Changes to src/wal.c.

146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
** last frame in the wal before frame M for page P in the WAL, or return
** NULL if there are no frames for page P in the WAL prior to M.
**
** The wal-index consists of a header region, followed by an one or
** more index blocks.  
**
** The wal-index header contains the total number of frames within the WAL
** in the the mxFrame field.  
**
** Each index block except for the first contains information on 
** HASHTABLE_NPAGE frames. The first index block contains information on
** HASHTABLE_NPAGE_ONE frames. The values of HASHTABLE_NPAGE_ONE and 
** HASHTABLE_NPAGE are selected so that together the wal-index header and
** first index block are the same size as all other index blocks in the
** wal-index.







|







146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
** last frame in the wal before frame M for page P in the WAL, or return
** NULL if there are no frames for page P in the WAL prior to M.
**
** The wal-index consists of a header region, followed by an one or
** more index blocks.  
**
** The wal-index header contains the total number of frames within the WAL
** in the mxFrame field.
**
** Each index block except for the first contains information on 
** HASHTABLE_NPAGE frames. The first index block contains information on
** HASHTABLE_NPAGE_ONE frames. The values of HASHTABLE_NPAGE_ONE and 
** HASHTABLE_NPAGE are selected so that together the wal-index header and
** first index block are the same size as all other index blocks in the
** wal-index.

Changes to test/crash.test.

115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
do_test crash-1.11 {
  catchsql {
    SELECT * FROM abc;
  }
} {0 {}}

#--------------------------------------------------------------------------
# The following tests test recovery when both the database file and the the
# journal file contain corrupt data. This can happen after pages are
# written to the database file before a transaction is committed due to
# cache-pressure.
#
# crash-2.1: Insert 18 pages of data into the database.
# crash-2.2: Check the database file size looks ok.
# crash-2.3: Delete 15 or so pages (with a 10 page page-cache), then crash.







|







115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
do_test crash-1.11 {
  catchsql {
    SELECT * FROM abc;
  }
} {0 {}}

#--------------------------------------------------------------------------
# The following tests test recovery when both the database file and the
# journal file contain corrupt data. This can happen after pages are
# written to the database file before a transaction is committed due to
# cache-pressure.
#
# crash-2.1: Insert 18 pages of data into the database.
# crash-2.2: Check the database file size looks ok.
# crash-2.3: Delete 15 or so pages (with a 10 page page-cache), then crash.

Changes to test/journal1.test.

37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
    INSERT INTO t1 SELECT a+2, a||b FROM t1;
    INSERT INTO t1 SELECT a+4, a||b FROM t1;
    SELECT count(*) FROM t1;
  }
} 8

# Make changes to the database and save the journal file.
# Then delete the database.  Replace the the journal file
# and try to create a new database with the same name.  The
# old journal should not attempt to rollback into the new
# database.
#
do_test journal1-1.2 {
  execsql {
    BEGIN;







|







37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
    INSERT INTO t1 SELECT a+2, a||b FROM t1;
    INSERT INTO t1 SELECT a+4, a||b FROM t1;
    SELECT count(*) FROM t1;
  }
} 8

# Make changes to the database and save the journal file.
# Then delete the database.  Replace the journal file
# and try to create a new database with the same name.  The
# old journal should not attempt to rollback into the new
# database.
#
do_test journal1-1.2 {
  execsql {
    BEGIN;

Changes to test/rowid.test.

653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
do_test rowid-11.4 {
  execsql {SELECT rowid, a FROM t5 WHERE rowid<='abc'}
} {1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8}

# Test the automatic generation of rowids when the table already contains
# a rowid with the maximum value.
#
# Once the the maximum rowid is taken, rowids are normally chosen at
# random.  By by reseting the random number generator, we can cause
# the rowid guessing loop to collide with prior rowids, and test the
# loop out to its limit of 100 iterations.  After 100 collisions, the
# rowid guesser gives up and reports SQLITE_FULL.
#
do_test rowid-12.1 {
  execsql {







|







653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
do_test rowid-11.4 {
  execsql {SELECT rowid, a FROM t5 WHERE rowid<='abc'}
} {1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8}

# Test the automatic generation of rowids when the table already contains
# a rowid with the maximum value.
#
# Once the maximum rowid is taken, rowids are normally chosen at
# random.  By by reseting the random number generator, we can cause
# the rowid guessing loop to collide with prior rowids, and test the
# loop out to its limit of 100 iterations.  After 100 collisions, the
# rowid guesser gives up and reports SQLITE_FULL.
#
do_test rowid-12.1 {
  execsql {

Changes to test/wal2.test.

82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
# and a reader ([db2]). For each of the 8 integer fields in the wal-index
# header (6 fields and 2 checksum values), do the following:
#
#   1. Modify the database using the writer.
#
#   2. Attempt to read the database using the reader. Before the reader
#      has a chance to snapshot the wal-index header, increment one
#      of the the integer fields (so that the reader ends up with a corrupted
#      header).
#
#   3. Check that the reader recovers the wal-index and reads the correct
#      database content.
#
do_test wal2-1.0 {
  proc tvfs_cb {method filename args} { 







|







82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
# and a reader ([db2]). For each of the 8 integer fields in the wal-index
# header (6 fields and 2 checksum values), do the following:
#
#   1. Modify the database using the writer.
#
#   2. Attempt to read the database using the reader. Before the reader
#      has a chance to snapshot the wal-index header, increment one
#      of the integer fields (so that the reader ends up with a corrupted
#      header).
#
#   3. Check that the reader recovers the wal-index and reads the correct
#      database content.
#
do_test wal2-1.0 {
  proc tvfs_cb {method filename args} {