System.Data.SQLite
Check-in [9166ff7f4f]
Not logged in

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

Overview
Comment:Update core SQLite library to the latest 3.7.13 release candidate.
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: 9166ff7f4fa1e566be6dc403551d32eaab7305fa
User & Date: mistachkin 2012-06-09 00:13:05
Context
2012-06-14
12:27
Update version history docs with the URL for SQLite 3.7.13. check-in: 0e89820825 user: mistachkin tags: trunk
2012-06-12
08:52
Experimental changes to support the prototype sqlite3_close_v2 interface. check-in: 8b0d3259cb user: mistachkin tags: deferred-close
2012-06-09
00:13
Update core SQLite library to the latest 3.7.13 release candidate. check-in: 9166ff7f4f user: mistachkin tags: trunk
2012-06-07
07:44
Fix hyperlink in version docs. check-in: 3d66db6fa8 user: mistachkin tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to SQLite.Interop/src/core/sqlite3.c.

662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
....
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022

5023
5024
5025
5026
5027
5028
5029
.....
70015
70016
70017
70018
70019
70020
70021
70022
70023
70024
70025
70026
70027
70028
70029
......
115646
115647
115648
115649
115650
115651
115652
115653
115654
115655
115656
115657
115658
115659
115660
115661
115662
......
125934
125935
125936
125937
125938
125939
125940
125941
125942
125943
125944
125945
125946
125947
125948
125949
125950
125951
......
125974
125975
125976
125977
125978
125979
125980
125981
125982
125983
125984
125985
125986
125987

125988
125989
125990
125991



125992
125993
125994
125995
125996
125997
125998
125999
126000
126001
126002
126003
126004
126005
126006
126007
126008
126009
126010
126011
126012
126013
126014




126015
126016
126017
126018
126019
126020
126021
126022
......
126192
126193
126194
126195
126196
126197
126198
126199
126200
126201
126202
126203
126204
126205
126206
126207
126208
126209
......
126586
126587
126588
126589
126590
126591
126592

126593
126594
126595
126596
126597
126598
126599
126600
......
126620
126621
126622
126623
126624
126625
126626
126627
126628
126629
126630
126631
126632
126633
126634
......
128031
128032
128033
128034
128035
128036
128037
128038
128039
128040
128041
128042
128043
128044
128045

128046
128047
128048
128049
128050
128051
128052
......
133436
133437
133438
133439
133440
133441
133442


133443
133444
133445
133446
133447
133448
133449
133450
133451
133452
133453
133454

133455


















































































































133456
133457
133458
133459
133460
133461
133462
133463
133464


133465
133466
133467
133468
133469
133470
133471
133472
133473
133474
133475
133476
133477
133478
133479
133480
133481






133482
133483
133484
133485
133486
133487
133488
133489


133490
133491
133492
133493
133494
133495
133496
133497
133498
133499
133500
133501
133502
133503
133504
......
133543
133544
133545
133546
133547
133548
133549
133550
133551
133552
133553
133554
133555
133556
133557

133558
133559
133560
133561
133562
133563
133564
133565
133566
133567
133568
133569
133570
133571
133572
133573
133574
133575
133576
133577
......
133583
133584
133585
133586
133587
133588
133589
133590
133591
133592
133593
133594
133595
133596
133597
133598
133599
133600
133601
133602
133603
133604
133605
133606
133607
**
** See also: [sqlite3_libversion()],
** [sqlite3_libversion_number()], [sqlite3_sourceid()],
** [sqlite_version()] and [sqlite_source_id()].
*/
#define SQLITE_VERSION        "3.7.13"
#define SQLITE_VERSION_NUMBER 3007013
#define SQLITE_SOURCE_ID      "2012-06-07 07:24:04 506008f000ba4af0b35da023b8c52f7a3f5033bd"

/*
** CAPI3REF: Run-Time Library Version Numbers
** KEYWORDS: sqlite3_version, sqlite3_sourceid
**
** These interfaces provide the same information as the [SQLITE_VERSION],
** [SQLITE_VERSION_NUMBER], and [SQLITE_SOURCE_ID] C preprocessor macros
................................................................................

/*
** CAPI3REF: Name Of The Folder Holding Database Files
**
** ^(If this global variable is made to point to a string which is
** the name of a folder (a.k.a. directory), then all database files
** specified with a relative pathname and created or accessed by
** SQLite when using a built-in [sqlite3_vfs | VFS] will be assumed
** to be relative to that directory.)^ ^If this variable is a NULL
** pointer, then SQLite assumes that all database files specified
** with a relative pathname are relative to the current directory
** for the process.

**
** Changing the value of this variable while a database connection is
** open can result in a corrupt database.
**
** It is not safe to read or modify this variable in more than one
** thread at a time.  It is not safe to read or modify this variable
** if a [database connection] is being used at the same time in a separate
................................................................................
  u.bl.pC = p->apCsr[pOp->p1];
  assert( u.bl.pC->isSorter==0 );
  assert( u.bl.pC->isTable || pOp->opcode!=OP_RowData );
  assert( u.bl.pC->isIndex || pOp->opcode==OP_RowData );
  assert( u.bl.pC!=0 );
  assert( u.bl.pC->nullRow==0 );
  assert( u.bl.pC->pseudoTableReg==0 );
  assert( !u.bl.pC->isSorter );
  assert( u.bl.pC->pCursor!=0 );
  u.bl.pCrsr = u.bl.pC->pCursor;
  assert( sqlite3BtreeCursorIsValid(u.bl.pCrsr) );

  /* The OP_RowKey and OP_RowData opcodes always follow OP_NotExists or
  ** OP_Rewind/Op_Next with no intervening instructions that might invalidate
  ** the cursor.  Hence the following sqlite3VdbeCursorMoveto() call is always
................................................................................
          zModeType = "cache";
        }
        if( nOpt==4 && memcmp("mode", zOpt, 4)==0 ){
          static struct OpenMode aOpenMode[] = {
            { "ro",  SQLITE_OPEN_READONLY },
            { "rw",  SQLITE_OPEN_READWRITE }, 
            { "rwc", SQLITE_OPEN_READWRITE | SQLITE_OPEN_CREATE },
            { "memory",
                    SQLITE_OPEN_READWRITE | SQLITE_OPEN_CREATE 
                       | SQLITE_OPEN_MEMORY },
            { 0, 0 }
          };

          mask = SQLITE_OPEN_READONLY | SQLITE_OPEN_READWRITE
                   | SQLITE_OPEN_CREATE | SQLITE_OPEN_MEMORY;
          aMode = aOpenMode;
          limit = mask & flags;
................................................................................

/* #include <tcl.h> */
/* #include <string.h> */

/*
** Implementation of a special SQL scalar function for testing tokenizers 
** designed to be used in concert with the Tcl testing framework. This
** function must be called with two arguments:
**
**   SELECT <function-name>(<key-name>, <input-string>);
**   SELECT <function-name>(<key-name>, <pointer>);
**
** where <function-name> is the name passed as the second argument
** to the sqlite3Fts3InitHashTable() function (e.g. 'fts3_tokenizer')
** concatenated with the string '_test' (e.g. 'fts3_tokenizer_test').
**
** The return value is a string that may be interpreted as a Tcl
** list. For each token in the <input-string>, three elements are
................................................................................
  const char *zErr = 0;

  const char *zName;
  int nName;
  const char *zInput;
  int nInput;

  const char *zArg = 0;

  const char *zToken;
  int nToken;
  int iStart;
  int iEnd;
  int iPos;


  Tcl_Obj *pRet;

  assert( argc==2 || argc==3 );




  nName = sqlite3_value_bytes(argv[0]);
  zName = (const char *)sqlite3_value_text(argv[0]);
  nInput = sqlite3_value_bytes(argv[argc-1]);
  zInput = (const char *)sqlite3_value_text(argv[argc-1]);

  if( argc==3 ){
    zArg = (const char *)sqlite3_value_text(argv[1]);
  }

  pHash = (Fts3Hash *)sqlite3_user_data(context);
  p = (sqlite3_tokenizer_module *)sqlite3Fts3HashFind(pHash, zName, nName+1);

  if( !p ){
    char *zErr = sqlite3_mprintf("unknown tokenizer: %s", zName);
    sqlite3_result_error(context, zErr, -1);
    sqlite3_free(zErr);
    return;
  }

  pRet = Tcl_NewObj();
  Tcl_IncrRefCount(pRet);





  if( SQLITE_OK!=p->xCreate(zArg ? 1 : 0, &zArg, &pTokenizer) ){
    zErr = "error in xCreate()";
    goto finish;
  }
  pTokenizer->pModule = p;
  if( sqlite3Fts3OpenTokenizer(pTokenizer, 0, zInput, nInput, &pCsr) ){
    zErr = "error in xOpen()";
    goto finish;
................................................................................
    rc = sqlite3_create_function(db, zName, 1, any, p, scalarFunc, 0, 0);
  }
  if( SQLITE_OK==rc ){
    rc = sqlite3_create_function(db, zName, 2, any, p, scalarFunc, 0, 0);
  }
#ifdef SQLITE_TEST
  if( SQLITE_OK==rc ){
    rc = sqlite3_create_function(db, zTest, 2, any, p, testFunc, 0, 0);
  }
  if( SQLITE_OK==rc ){
    rc = sqlite3_create_function(db, zTest, 3, any, p, testFunc, 0, 0);
  }
  if( SQLITE_OK==rc ){
    rc = sqlite3_create_function(db, zTest2, 0, any, pdb, intTestFunc, 0, 0);
  }
#endif

#ifdef SQLITE_TEST
................................................................................
**
**   fts3SegReaderNext()
**   fts3SegReaderFirstDocid()
**   fts3SegReaderNextDocid()
*/
struct Fts3SegReader {
  int iIdx;                       /* Index within level, or 0x7FFFFFFF for PT */

  int bLookup;                    /* True for a lookup only */

  sqlite3_int64 iStartBlock;      /* Rowid of first leaf block to traverse */
  sqlite3_int64 iLeafEndBlock;    /* Rowid of final leaf block to traverse */
  sqlite3_int64 iEndBlock;        /* Rowid of final block in segment (or 0) */
  sqlite3_int64 iCurrentBlock;    /* Current leaf block (or 0) */

  char *aNode;                    /* Pointer to node data (or NULL) */
................................................................................
  */
  char *pOffsetList;
  int nOffsetList;                /* For descending pending seg-readers only */
  sqlite3_int64 iDocid;
};

#define fts3SegReaderIsPending(p) ((p)->ppNextElem!=0)
#define fts3SegReaderIsRootOnly(p) ((p)->aNode==(char *)&(p)[1])

/*
** An instance of this structure is used to create a segment b-tree in the
** database. The internal details of this type are only accessed by the
** following functions:
**
**   fts3SegWriterAdd()
................................................................................

  pReader = (Fts3SegReader *)sqlite3_malloc(sizeof(Fts3SegReader) + nExtra);
  if( !pReader ){
    return SQLITE_NOMEM;
  }
  memset(pReader, 0, sizeof(Fts3SegReader));
  pReader->iIdx = iAge;
  pReader->bLookup = bLookup;
  pReader->iStartBlock = iStartLeaf;
  pReader->iLeafEndBlock = iEndLeaf;
  pReader->iEndBlock = iEndBlock;

  if( nExtra ){
    /* The entire segment is stored in the root node. */
    pReader->aNode = (char *)&pReader[1];

    pReader->nNode = nRoot;
    memcpy(pReader->aNode, zRoot, nRoot);
    memset(&pReader->aNode[nRoot], 0, FTS3_NODE_PADDING);
  }else{
    pReader->iCurrentBlock = iStartLeaf-1;
  }
  *ppReader = pReader;
................................................................................

typedef struct unicode_tokenizer unicode_tokenizer;
typedef struct unicode_cursor unicode_cursor;

struct unicode_tokenizer {
  sqlite3_tokenizer base;
  int bRemoveDiacritic;


};

struct unicode_cursor {
  sqlite3_tokenizer_cursor base;
  const unsigned char *aInput;    /* Input text being tokenized */
  int nInput;                     /* Size of aInput[] in bytes */
  int iOff;                       /* Current offset within aInput[] */
  int iToken;                     /* Index of next token to be returned */
  char *zToken;                   /* storage for current token */
  int nAlloc;                     /* space allocated at zToken */
};


/*


















































































































** Create a new tokenizer instance.
*/
static int unicodeCreate(
  int nArg,                       /* Size of array argv[] */
  const char * const *azArg,      /* Tokenizer creation arguments */
  sqlite3_tokenizer **pp          /* OUT: New tokenizer handle */
){
  unicode_tokenizer *pNew;        /* New tokenizer object */
  int i;


  pNew = (unicode_tokenizer *) sqlite3_malloc(sizeof(unicode_tokenizer));
  if( pNew==NULL ){
    return SQLITE_NOMEM;
  }
  memset(pNew, 0, sizeof(unicode_tokenizer));
  pNew->bRemoveDiacritic = 1;

  for(i=0; i<nArg; i++){
    const char *z = azArg[i];
    int n = strlen(z);

    if( n==19 && memcmp("remove_diacritics=1", z, 19)==0 ){
      pNew->bRemoveDiacritic = 1;
    }
    else if( n==19 && memcmp("remove_diacritics=0", z, 19)==0 ){
      pNew->bRemoveDiacritic = 0;
    }






    else{
      /* Unrecognized argument */
      return SQLITE_ERROR;
    }
  }

  *pp = &pNew->base;
  return SQLITE_OK;


}

/*
** Destroy a tokenizer allocated by unicodeCreate().
*/
static int unicodeDestroy(sqlite3_tokenizer *pTokenizer){
  sqlite3_free(pTokenizer);
  return SQLITE_OK;
}

/*
** Prepare to begin tokenizing a particular string.  The input
** string to be tokenized is pInput[0..nBytes-1].  A cursor
** used to incrementally tokenize this string is returned in 
** *ppCursor.
................................................................................
}

/*
** Extract the next token from a tokenization cursor.  The cursor must
** have been opened by a prior call to simpleOpen().
*/
static int unicodeNext(
  sqlite3_tokenizer_cursor *p,    /* Cursor returned by simpleOpen */
  const char **paToken,           /* OUT: Token text */
  int *pnToken,                   /* OUT: Number of bytes at *paToken */
  int *piStart,                   /* OUT: Starting offset of token */
  int *piEnd,                     /* OUT: Ending offset of token */
  int *piPos                      /* OUT: Position integer of token */
){
  unicode_cursor *pCsr = (unicode_cursor *)p;

  int iCode;
  char *zOut;
  const unsigned char *z = &pCsr->aInput[pCsr->iOff];
  const unsigned char *zStart = z;
  const unsigned char *zEnd;
  const unsigned char *zTerm = &pCsr->aInput[pCsr->nInput];

  /* Scan past any delimiter characters before the start of the next token.
  ** Return SQLITE_DONE early if this takes us all the way to the end of 
  ** the input.  */
  while( z<zTerm ){
    READ_UTF8(z, zTerm, iCode);
    if( sqlite3FtsUnicodeIsalnum(iCode) ) break;
    zStart = z;
  }
  if( zStart>=zTerm ) return SQLITE_DONE;

  zOut = pCsr->zToken;
  do {
    int iOut;
................................................................................
      zOut = &zNew[zOut - pCsr->zToken];
      pCsr->zToken = zNew;
      pCsr->nAlloc += 64;
    }

    /* Write the folded case of the last character read to the output */
    zEnd = z;
    iOut = sqlite3FtsUnicodeFold(iCode, 
        ((unicode_tokenizer *)pCsr->base.pTokenizer)->bRemoveDiacritic
    );
    if( iOut ){
      WRITE_UTF8(zOut, iOut);
    }

    /* If the cursor is not at EOF, read the next character */
    if( z>=zTerm ) break;
    READ_UTF8(z, zTerm, iCode);
  }while( sqlite3FtsUnicodeIsalnum(iCode) 
       || sqlite3FtsUnicodeIsdiacritic(iCode)
  );

  /* Set the output variables and return. */
  pCsr->iOff = (z - pCsr->aInput);
  *paToken = pCsr->zToken;
  *pnToken = zOut - pCsr->zToken;







|







 







|



|
>







 







<







 







<
<
|







 







|

<
|







 







|






>



|
>
>
>






<
<
<
<













>
>
>
>
|







 







|
<
<
<







 







>
|







 







|







 







|







>







 







>
>












>

>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>









>
>

<
|
<



|









>
>
>
>
>
>


|



<
|
>
>
|
<
<
<
<
|
<
|







 







|






|
>












|







 







|
<
<







|







662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
....
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
.....
70016
70017
70018
70019
70020
70021
70022

70023
70024
70025
70026
70027
70028
70029
......
115646
115647
115648
115649
115650
115651
115652


115653
115654
115655
115656
115657
115658
115659
115660
......
125932
125933
125934
125935
125936
125937
125938
125939
125940

125941
125942
125943
125944
125945
125946
125947
125948
......
125971
125972
125973
125974
125975
125976
125977
125978
125979
125980
125981
125982
125983
125984
125985
125986
125987
125988
125989
125990
125991
125992
125993
125994
125995
125996
125997
125998




125999
126000
126001
126002
126003
126004
126005
126006
126007
126008
126009
126010
126011
126012
126013
126014
126015
126016
126017
126018
126019
126020
126021
126022
126023
......
126193
126194
126195
126196
126197
126198
126199
126200



126201
126202
126203
126204
126205
126206
126207
......
126584
126585
126586
126587
126588
126589
126590
126591
126592
126593
126594
126595
126596
126597
126598
126599
......
126619
126620
126621
126622
126623
126624
126625
126626
126627
126628
126629
126630
126631
126632
126633
......
128030
128031
128032
128033
128034
128035
128036
128037
128038
128039
128040
128041
128042
128043
128044
128045
128046
128047
128048
128049
128050
128051
128052
......
133436
133437
133438
133439
133440
133441
133442
133443
133444
133445
133446
133447
133448
133449
133450
133451
133452
133453
133454
133455
133456
133457
133458
133459
133460
133461
133462
133463
133464
133465
133466
133467
133468
133469
133470
133471
133472
133473
133474
133475
133476
133477
133478
133479
133480
133481
133482
133483
133484
133485
133486
133487
133488
133489
133490
133491
133492
133493
133494
133495
133496
133497
133498
133499
133500
133501
133502
133503
133504
133505
133506
133507
133508
133509
133510
133511
133512
133513
133514
133515
133516
133517
133518
133519
133520
133521
133522
133523
133524
133525
133526
133527
133528
133529
133530
133531
133532
133533
133534
133535
133536
133537
133538
133539
133540
133541
133542
133543
133544
133545
133546
133547
133548
133549
133550
133551
133552
133553
133554
133555
133556
133557
133558
133559
133560
133561
133562
133563
133564
133565
133566
133567
133568
133569
133570
133571
133572
133573
133574
133575
133576
133577
133578
133579
133580
133581
133582
133583
133584

133585

133586
133587
133588
133589
133590
133591
133592
133593
133594
133595
133596
133597
133598
133599
133600
133601
133602
133603
133604
133605
133606
133607
133608
133609
133610

133611
133612
133613
133614




133615

133616
133617
133618
133619
133620
133621
133622
133623
......
133662
133663
133664
133665
133666
133667
133668
133669
133670
133671
133672
133673
133674
133675
133676
133677
133678
133679
133680
133681
133682
133683
133684
133685
133686
133687
133688
133689
133690
133691
133692
133693
133694
133695
133696
133697
......
133703
133704
133705
133706
133707
133708
133709
133710


133711
133712
133713
133714
133715
133716
133717
133718
133719
133720
133721
133722
133723
133724
133725
**
** See also: [sqlite3_libversion()],
** [sqlite3_libversion_number()], [sqlite3_sourceid()],
** [sqlite_version()] and [sqlite_source_id()].
*/
#define SQLITE_VERSION        "3.7.13"
#define SQLITE_VERSION_NUMBER 3007013
#define SQLITE_SOURCE_ID      "2012-06-08 14:01:53 025227be5495f950c466dfabac140cba69e498be"

/*
** CAPI3REF: Run-Time Library Version Numbers
** KEYWORDS: sqlite3_version, sqlite3_sourceid
**
** These interfaces provide the same information as the [SQLITE_VERSION],
** [SQLITE_VERSION_NUMBER], and [SQLITE_SOURCE_ID] C preprocessor macros
................................................................................

/*
** CAPI3REF: Name Of The Folder Holding Database Files
**
** ^(If this global variable is made to point to a string which is
** the name of a folder (a.k.a. directory), then all database files
** specified with a relative pathname and created or accessed by
** SQLite when using a built-in windows [sqlite3_vfs | VFS] will be assumed
** to be relative to that directory.)^ ^If this variable is a NULL
** pointer, then SQLite assumes that all database files specified
** with a relative pathname are relative to the current directory
** for the process.  Only the windows VFS makes use of this global
** variable; it is ignored by the unix VFS.
**
** Changing the value of this variable while a database connection is
** open can result in a corrupt database.
**
** It is not safe to read or modify this variable in more than one
** thread at a time.  It is not safe to read or modify this variable
** if a [database connection] is being used at the same time in a separate
................................................................................
  u.bl.pC = p->apCsr[pOp->p1];
  assert( u.bl.pC->isSorter==0 );
  assert( u.bl.pC->isTable || pOp->opcode!=OP_RowData );
  assert( u.bl.pC->isIndex || pOp->opcode==OP_RowData );
  assert( u.bl.pC!=0 );
  assert( u.bl.pC->nullRow==0 );
  assert( u.bl.pC->pseudoTableReg==0 );

  assert( u.bl.pC->pCursor!=0 );
  u.bl.pCrsr = u.bl.pC->pCursor;
  assert( sqlite3BtreeCursorIsValid(u.bl.pCrsr) );

  /* The OP_RowKey and OP_RowData opcodes always follow OP_NotExists or
  ** OP_Rewind/Op_Next with no intervening instructions that might invalidate
  ** the cursor.  Hence the following sqlite3VdbeCursorMoveto() call is always
................................................................................
          zModeType = "cache";
        }
        if( nOpt==4 && memcmp("mode", zOpt, 4)==0 ){
          static struct OpenMode aOpenMode[] = {
            { "ro",  SQLITE_OPEN_READONLY },
            { "rw",  SQLITE_OPEN_READWRITE }, 
            { "rwc", SQLITE_OPEN_READWRITE | SQLITE_OPEN_CREATE },


            { "memory", SQLITE_OPEN_MEMORY },
            { 0, 0 }
          };

          mask = SQLITE_OPEN_READONLY | SQLITE_OPEN_READWRITE
                   | SQLITE_OPEN_CREATE | SQLITE_OPEN_MEMORY;
          aMode = aOpenMode;
          limit = mask & flags;
................................................................................

/* #include <tcl.h> */
/* #include <string.h> */

/*
** Implementation of a special SQL scalar function for testing tokenizers 
** designed to be used in concert with the Tcl testing framework. This
** function must be called with two or more arguments:
**

**   SELECT <function-name>(<key-name>, ..., <input-string>);
**
** where <function-name> is the name passed as the second argument
** to the sqlite3Fts3InitHashTable() function (e.g. 'fts3_tokenizer')
** concatenated with the string '_test' (e.g. 'fts3_tokenizer_test').
**
** The return value is a string that may be interpreted as a Tcl
** list. For each token in the <input-string>, three elements are
................................................................................
  const char *zErr = 0;

  const char *zName;
  int nName;
  const char *zInput;
  int nInput;

  const char *azArg[64];

  const char *zToken;
  int nToken;
  int iStart;
  int iEnd;
  int iPos;
  int i;

  Tcl_Obj *pRet;

  if( argc<2 ){
    sqlite3_result_error(context, "insufficient arguments", -1);
    return;
  }

  nName = sqlite3_value_bytes(argv[0]);
  zName = (const char *)sqlite3_value_text(argv[0]);
  nInput = sqlite3_value_bytes(argv[argc-1]);
  zInput = (const char *)sqlite3_value_text(argv[argc-1]);





  pHash = (Fts3Hash *)sqlite3_user_data(context);
  p = (sqlite3_tokenizer_module *)sqlite3Fts3HashFind(pHash, zName, nName+1);

  if( !p ){
    char *zErr = sqlite3_mprintf("unknown tokenizer: %s", zName);
    sqlite3_result_error(context, zErr, -1);
    sqlite3_free(zErr);
    return;
  }

  pRet = Tcl_NewObj();
  Tcl_IncrRefCount(pRet);

  for(i=1; i<argc-1; i++){
    azArg[i-1] = (const char *)sqlite3_value_text(argv[i]);
  }

  if( SQLITE_OK!=p->xCreate(argc-2, azArg, &pTokenizer) ){
    zErr = "error in xCreate()";
    goto finish;
  }
  pTokenizer->pModule = p;
  if( sqlite3Fts3OpenTokenizer(pTokenizer, 0, zInput, nInput, &pCsr) ){
    zErr = "error in xOpen()";
    goto finish;
................................................................................
    rc = sqlite3_create_function(db, zName, 1, any, p, scalarFunc, 0, 0);
  }
  if( SQLITE_OK==rc ){
    rc = sqlite3_create_function(db, zName, 2, any, p, scalarFunc, 0, 0);
  }
#ifdef SQLITE_TEST
  if( SQLITE_OK==rc ){
    rc = sqlite3_create_function(db, zTest, -1, any, p, testFunc, 0, 0);



  }
  if( SQLITE_OK==rc ){
    rc = sqlite3_create_function(db, zTest2, 0, any, pdb, intTestFunc, 0, 0);
  }
#endif

#ifdef SQLITE_TEST
................................................................................
**
**   fts3SegReaderNext()
**   fts3SegReaderFirstDocid()
**   fts3SegReaderNextDocid()
*/
struct Fts3SegReader {
  int iIdx;                       /* Index within level, or 0x7FFFFFFF for PT */
  u8 bLookup;                     /* True for a lookup only */
  u8 rootOnly;                    /* True for a root-only reader */

  sqlite3_int64 iStartBlock;      /* Rowid of first leaf block to traverse */
  sqlite3_int64 iLeafEndBlock;    /* Rowid of final leaf block to traverse */
  sqlite3_int64 iEndBlock;        /* Rowid of final block in segment (or 0) */
  sqlite3_int64 iCurrentBlock;    /* Current leaf block (or 0) */

  char *aNode;                    /* Pointer to node data (or NULL) */
................................................................................
  */
  char *pOffsetList;
  int nOffsetList;                /* For descending pending seg-readers only */
  sqlite3_int64 iDocid;
};

#define fts3SegReaderIsPending(p) ((p)->ppNextElem!=0)
#define fts3SegReaderIsRootOnly(p) ((p)->rootOnly!=0)

/*
** An instance of this structure is used to create a segment b-tree in the
** database. The internal details of this type are only accessed by the
** following functions:
**
**   fts3SegWriterAdd()
................................................................................

  pReader = (Fts3SegReader *)sqlite3_malloc(sizeof(Fts3SegReader) + nExtra);
  if( !pReader ){
    return SQLITE_NOMEM;
  }
  memset(pReader, 0, sizeof(Fts3SegReader));
  pReader->iIdx = iAge;
  pReader->bLookup = bLookup!=0;
  pReader->iStartBlock = iStartLeaf;
  pReader->iLeafEndBlock = iEndLeaf;
  pReader->iEndBlock = iEndBlock;

  if( nExtra ){
    /* The entire segment is stored in the root node. */
    pReader->aNode = (char *)&pReader[1];
    pReader->rootOnly = 1;
    pReader->nNode = nRoot;
    memcpy(pReader->aNode, zRoot, nRoot);
    memset(&pReader->aNode[nRoot], 0, FTS3_NODE_PADDING);
  }else{
    pReader->iCurrentBlock = iStartLeaf-1;
  }
  *ppReader = pReader;
................................................................................

typedef struct unicode_tokenizer unicode_tokenizer;
typedef struct unicode_cursor unicode_cursor;

struct unicode_tokenizer {
  sqlite3_tokenizer base;
  int bRemoveDiacritic;
  int nException;
  int *aiException;
};

struct unicode_cursor {
  sqlite3_tokenizer_cursor base;
  const unsigned char *aInput;    /* Input text being tokenized */
  int nInput;                     /* Size of aInput[] in bytes */
  int iOff;                       /* Current offset within aInput[] */
  int iToken;                     /* Index of next token to be returned */
  char *zToken;                   /* storage for current token */
  int nAlloc;                     /* space allocated at zToken */
};


/*
** Destroy a tokenizer allocated by unicodeCreate().
*/
static int unicodeDestroy(sqlite3_tokenizer *pTokenizer){
  if( pTokenizer ){
    unicode_tokenizer *p = (unicode_tokenizer *)pTokenizer;
    sqlite3_free(p->aiException);
    sqlite3_free(p);
  }
  return SQLITE_OK;
}

/*
** As part of a tokenchars= or separators= option, the CREATE VIRTUAL TABLE
** statement has specified that the tokenizer for this table shall consider
** all characters in string zIn/nIn to be separators (if bAlnum==0) or
** token characters (if bAlnum==1).
**
** For each codepoint in the zIn/nIn string, this function checks if the
** sqlite3FtsUnicodeIsalnum() function already returns the desired result.
** If so, no action is taken. Otherwise, the codepoint is added to the 
** unicode_tokenizer.aiException[] array. For the purposes of tokenization,
** the return value of sqlite3FtsUnicodeIsalnum() is inverted for all
** codepoints in the aiException[] array.
**
** If a standalone diacritic mark (one that sqlite3FtsUnicodeIsdiacritic()
** identifies as a diacritic) occurs in the zIn/nIn string it is ignored.
** It is not possible to change the behaviour of the tokenizer with respect
** to these codepoints.
*/
static int unicodeAddExceptions(
  unicode_tokenizer *p,           /* Tokenizer to add exceptions to */
  int bAlnum,                     /* Replace Isalnum() return value with this */
  const char *zIn,                /* Array of characters to make exceptions */
  int nIn                         /* Length of z in bytes */
){
  const unsigned char *z = (const unsigned char *)zIn;
  const unsigned char *zTerm = &z[nIn];
  int iCode;
  int nEntry = 0;

  assert( bAlnum==0 || bAlnum==1 );

  while( z<zTerm ){
    READ_UTF8(z, zTerm, iCode);
    assert( (sqlite3FtsUnicodeIsalnum(iCode) & 0xFFFFFFFE)==0 );
    if( sqlite3FtsUnicodeIsalnum(iCode)!=bAlnum 
     && sqlite3FtsUnicodeIsdiacritic(iCode)==0 
    ){
      nEntry++;
    }
  }

  if( nEntry ){
    int *aNew;                    /* New aiException[] array */
    int nNew;                     /* Number of valid entries in array aNew[] */

    aNew = sqlite3_realloc(p->aiException, (p->nException+nEntry)*sizeof(int));
    if( aNew==0 ) return SQLITE_NOMEM;
    nNew = p->nException;

    z = (const unsigned char *)zIn;
    while( z<zTerm ){
      READ_UTF8(z, zTerm, iCode);
      if( sqlite3FtsUnicodeIsalnum(iCode)!=bAlnum 
       && sqlite3FtsUnicodeIsdiacritic(iCode)==0
      ){
        int i, j;
        for(i=0; i<nNew && aNew[i]<iCode; i++);
        for(j=nNew; j>i; j--) aNew[j] = aNew[j-1];
        aNew[i] = iCode;
        nNew++;
      }
    }
    p->aiException = aNew;
    p->nException = nNew;
  }

  return SQLITE_OK;
}

/*
** Return true if the p->aiException[] array contains the value iCode.
*/
static int unicodeIsException(unicode_tokenizer *p, int iCode){
  if( p->nException>0 ){
    int *a = p->aiException;
    int iLo = 0;
    int iHi = p->nException-1;

    while( iHi>=iLo ){
      int iTest = (iHi + iLo) / 2;
      if( iCode==a[iTest] ){
        return 1;
      }else if( iCode>a[iTest] ){
        iLo = iTest+1;
      }else{
        iHi = iTest-1;
      }
    }
  }

  return 0;
}

/*
** Return true if, for the purposes of tokenization, codepoint iCode is
** considered a token character (not a separator).
*/
static int unicodeIsAlnum(unicode_tokenizer *p, int iCode){
  assert( (sqlite3FtsUnicodeIsalnum(iCode) & 0xFFFFFFFE)==0 );
  return sqlite3FtsUnicodeIsalnum(iCode) ^ unicodeIsException(p, iCode);
}

/*
** Create a new tokenizer instance.
*/
static int unicodeCreate(
  int nArg,                       /* Size of array argv[] */
  const char * const *azArg,      /* Tokenizer creation arguments */
  sqlite3_tokenizer **pp          /* OUT: New tokenizer handle */
){
  unicode_tokenizer *pNew;        /* New tokenizer object */
  int i;
  int rc = SQLITE_OK;

  pNew = (unicode_tokenizer *) sqlite3_malloc(sizeof(unicode_tokenizer));

  if( pNew==NULL ) return SQLITE_NOMEM;

  memset(pNew, 0, sizeof(unicode_tokenizer));
  pNew->bRemoveDiacritic = 1;

  for(i=0; rc==SQLITE_OK && i<nArg; i++){
    const char *z = azArg[i];
    int n = strlen(z);

    if( n==19 && memcmp("remove_diacritics=1", z, 19)==0 ){
      pNew->bRemoveDiacritic = 1;
    }
    else if( n==19 && memcmp("remove_diacritics=0", z, 19)==0 ){
      pNew->bRemoveDiacritic = 0;
    }
    else if( n>=11 && memcmp("tokenchars=", z, 11)==0 ){
      rc = unicodeAddExceptions(pNew, 1, &z[11], n-11);
    }
    else if( n>=11 && memcmp("separators=", z, 11)==0 ){
      rc = unicodeAddExceptions(pNew, 0, &z[11], n-11);
    }
    else{
      /* Unrecognized argument */
      rc  = SQLITE_ERROR;
    }
  }


  if( rc!=SQLITE_OK ){
    unicodeDestroy((sqlite3_tokenizer *)pNew);
    pNew = 0;
  }




  *pp = (sqlite3_tokenizer *)pNew;

  return rc;
}

/*
** Prepare to begin tokenizing a particular string.  The input
** string to be tokenized is pInput[0..nBytes-1].  A cursor
** used to incrementally tokenize this string is returned in 
** *ppCursor.
................................................................................
}

/*
** Extract the next token from a tokenization cursor.  The cursor must
** have been opened by a prior call to simpleOpen().
*/
static int unicodeNext(
  sqlite3_tokenizer_cursor *pC,   /* Cursor returned by simpleOpen */
  const char **paToken,           /* OUT: Token text */
  int *pnToken,                   /* OUT: Number of bytes at *paToken */
  int *piStart,                   /* OUT: Starting offset of token */
  int *piEnd,                     /* OUT: Ending offset of token */
  int *piPos                      /* OUT: Position integer of token */
){
  unicode_cursor *pCsr = (unicode_cursor *)pC;
  unicode_tokenizer *p = ((unicode_tokenizer *)pCsr->base.pTokenizer);
  int iCode;
  char *zOut;
  const unsigned char *z = &pCsr->aInput[pCsr->iOff];
  const unsigned char *zStart = z;
  const unsigned char *zEnd;
  const unsigned char *zTerm = &pCsr->aInput[pCsr->nInput];

  /* Scan past any delimiter characters before the start of the next token.
  ** Return SQLITE_DONE early if this takes us all the way to the end of 
  ** the input.  */
  while( z<zTerm ){
    READ_UTF8(z, zTerm, iCode);
    if( unicodeIsAlnum(p, iCode) ) break;
    zStart = z;
  }
  if( zStart>=zTerm ) return SQLITE_DONE;

  zOut = pCsr->zToken;
  do {
    int iOut;
................................................................................
      zOut = &zNew[zOut - pCsr->zToken];
      pCsr->zToken = zNew;
      pCsr->nAlloc += 64;
    }

    /* Write the folded case of the last character read to the output */
    zEnd = z;
    iOut = sqlite3FtsUnicodeFold(iCode, p->bRemoveDiacritic);


    if( iOut ){
      WRITE_UTF8(zOut, iOut);
    }

    /* If the cursor is not at EOF, read the next character */
    if( z>=zTerm ) break;
    READ_UTF8(z, zTerm, iCode);
  }while( unicodeIsAlnum(p, iCode) 
       || sqlite3FtsUnicodeIsdiacritic(iCode)
  );

  /* Set the output variables and return. */
  pCsr->iOff = (z - pCsr->aInput);
  *paToken = pCsr->zToken;
  *pnToken = zOut - pCsr->zToken;

Changes to SQLite.Interop/src/core/sqlite3.h.

105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
....
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465

4466
4467
4468
4469
4470
4471
4472
**
** See also: [sqlite3_libversion()],
** [sqlite3_libversion_number()], [sqlite3_sourceid()],
** [sqlite_version()] and [sqlite_source_id()].
*/
#define SQLITE_VERSION        "3.7.13"
#define SQLITE_VERSION_NUMBER 3007013
#define SQLITE_SOURCE_ID      "2012-06-07 07:24:04 506008f000ba4af0b35da023b8c52f7a3f5033bd"

/*
** CAPI3REF: Run-Time Library Version Numbers
** KEYWORDS: sqlite3_version, sqlite3_sourceid
**
** These interfaces provide the same information as the [SQLITE_VERSION],
** [SQLITE_VERSION_NUMBER], and [SQLITE_SOURCE_ID] C preprocessor macros
................................................................................

/*
** CAPI3REF: Name Of The Folder Holding Database Files
**
** ^(If this global variable is made to point to a string which is
** the name of a folder (a.k.a. directory), then all database files
** specified with a relative pathname and created or accessed by
** SQLite when using a built-in [sqlite3_vfs | VFS] will be assumed
** to be relative to that directory.)^ ^If this variable is a NULL
** pointer, then SQLite assumes that all database files specified
** with a relative pathname are relative to the current directory
** for the process.

**
** Changing the value of this variable while a database connection is
** open can result in a corrupt database.
**
** It is not safe to read or modify this variable in more than one
** thread at a time.  It is not safe to read or modify this variable
** if a [database connection] is being used at the same time in a separate







|







 







|



|
>







105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
....
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
**
** See also: [sqlite3_libversion()],
** [sqlite3_libversion_number()], [sqlite3_sourceid()],
** [sqlite_version()] and [sqlite_source_id()].
*/
#define SQLITE_VERSION        "3.7.13"
#define SQLITE_VERSION_NUMBER 3007013
#define SQLITE_SOURCE_ID      "2012-06-08 14:01:53 025227be5495f950c466dfabac140cba69e498be"

/*
** CAPI3REF: Run-Time Library Version Numbers
** KEYWORDS: sqlite3_version, sqlite3_sourceid
**
** These interfaces provide the same information as the [SQLITE_VERSION],
** [SQLITE_VERSION_NUMBER], and [SQLITE_SOURCE_ID] C preprocessor macros
................................................................................

/*
** CAPI3REF: Name Of The Folder Holding Database Files
**
** ^(If this global variable is made to point to a string which is
** the name of a folder (a.k.a. directory), then all database files
** specified with a relative pathname and created or accessed by
** SQLite when using a built-in windows [sqlite3_vfs | VFS] will be assumed
** to be relative to that directory.)^ ^If this variable is a NULL
** pointer, then SQLite assumes that all database files specified
** with a relative pathname are relative to the current directory
** for the process.  Only the windows VFS makes use of this global
** variable; it is ignored by the unix VFS.
**
** Changing the value of this variable while a database connection is
** open can result in a corrupt database.
**
** It is not safe to read or modify this variable in more than one
** thread at a time.  It is not safe to read or modify this variable
** if a [database connection] is being used at the same time in a separate