<li>Support automatic value conversions for columns with a declared type of BIGUINT, INTEGER8, INTEGER16, INTEGER32, INTEGER64, SMALLUINT, TINYSINT, UNSIGNEDINTEGER, UNSIGNEDINTEGER8, UNSIGNEDINTEGER16, UNSIGNEDINTEGER32, UNSIGNEDINTEGER64, INT8, INT16, INT32, INT64, UINT, UINT8, UINT16, UINT32, UINT64, or ULONG.</li>
      <li>Add BindUInt32AsInt64 connection flag to force binding of UInt32 values as Int64 instead. Pursuant to <a href="http://system.data.sqlite.org/index.html/info/c010fa6584">[c010fa6584]</a>.</li>
      <li>Add BindAllAsText connection flag to force binding of all values as text.</li>
      <li>Remove AUTOINCREMENT from the column type name map.&nbsp;<b>** Potentially Incompatible Change **</b></li>
      <li>Avoid throwing overflow exceptions from the SQLite3.GetValue method for integral column types. Partial fix for <a href="http://system.data.sqlite.org/index.html/info/c010fa6584">[c010fa6584]</a>.&nbsp;<b>** Potentially Incompatible Change **</b></li>
      <li>Use the legacy connection closing algorithm when built with the INTEROP_LEGACY_CLOSE compile-time option.</li>
      <li>Support using the directory containing the primary managed-only assembly as the basis for native library pre-loading.</li>

    </ul>
    <p><b>1.0.84.0 - January 9, 2013</b></p>
    <ul>
      <li>Updated to <a href="http://www.sqlite.org/releaselog/3_7_15_2.html">SQLite 3.7.15.2</a>.</li>
      <li>Explicitly dispose of all SQLiteCommand objects managed by the DbDataAdapter class. Fix for <a href="http://system.data.sqlite.org/index.html/info/6434e23a0f">[6434e23a0f]</a>.</li>
      <li>Add Cancel method to the SQLiteConnection class to interrupt a long running query.</li>
      <li>Improve thread safety of the SQLiteLog.LogMessage method.</li>

      <li>Support automatic value conversions for columns with a declared type of BIGUINT, INTEGER8, INTEGER16, INTEGER32, INTEGER64, SMALLUINT, TINYSINT, UNSIGNEDINTEGER, UNSIGNEDINTEGER8, UNSIGNEDINTEGER16, UNSIGNEDINTEGER32, UNSIGNEDINTEGER64, INT8, INT16, INT32, INT64, UINT, UINT8, UINT16, UINT32, UINT64, or ULONG.</li>
      <li>Add BindUInt32AsInt64 connection flag to force binding of UInt32 values as Int64 instead. Pursuant to <a href="http://system.data.sqlite.org/index.html/info/c010fa6584">[c010fa6584]</a>.</li>
      <li>Add BindAllAsText connection flag to force binding of all values as text.</li>
      <li>Remove AUTOINCREMENT from the column type name map.&nbsp;<b>** Potentially Incompatible Change **</b></li>
      <li>Avoid throwing overflow exceptions from the SQLite3.GetValue method for integral column types. Partial fix for <a href="http://system.data.sqlite.org/index.html/info/c010fa6584">[c010fa6584]</a>.&nbsp;<b>** Potentially Incompatible Change **</b></li>
      <li>Use the legacy connection closing algorithm when built with the INTEROP_LEGACY_CLOSE compile-time option.</li>
      <li>Support using the directory containing the primary managed-only assembly as the basis for native library pre-loading.</li>
      <li>Still further enhancements to the build and test automation.</li>
    </ul>
    <p><b>1.0.84.0 - January 9, 2013</b></p>
    <ul>
      <li>Updated to <a href="http://www.sqlite.org/releaselog/3_7_15_2.html">SQLite 3.7.15.2</a>.</li>
      <li>Explicitly dispose of all SQLiteCommand objects managed by the DbDataAdapter class. Fix for <a href="http://system.data.sqlite.org/index.html/info/6434e23a0f">[6434e23a0f]</a>.</li>
      <li>Add Cancel method to the SQLiteConnection class to interrupt a long running query.</li>
      <li>Improve thread safety of the SQLiteLog.LogMessage method.</li>

** Later we learn that _XOPEN_SOURCE is poorly or incorrectly
** implemented on some systems.  So we avoid defining it at all
** if it is already defined or if it is unneeded because we are
** not doing a threadsafe build.  Ticket #2681.
**
** See also ticket #2741.
*/
#if !defined(_XOPEN_SOURCE) && !defined(__DARWIN__) && !defined(__APPLE__) && SQLITE_THREADSAFE

#  define _XOPEN_SOURCE 500  /* Needed to enable pthread recursive mutexes */
#endif

/*
** The TCL headers are only needed when compiling the TCL bindings.
*/
#if defined(SQLITE_TCL) || defined(TCLSH)
................................................................................
**
** See also: [sqlite3_libversion()],
** [sqlite3_libversion_number()], [sqlite3_sourceid()],
** [sqlite_version()] and [sqlite_source_id()].
*/
#define SQLITE_VERSION        "3.7.16"
#define SQLITE_VERSION_NUMBER 3007016
#define SQLITE_SOURCE_ID      "2013-02-15 04:21:01 843e1c543aabab8cd62f28742d5818887d36bcb7"

/*
** 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
................................................................................
SQLITE_PRIVATE Select *sqlite3SelectNew(Parse*,ExprList*,SrcList*,Expr*,ExprList*,
                         Expr*,ExprList*,u16,Expr*,Expr*);
SQLITE_PRIVATE void sqlite3SelectDelete(sqlite3*, Select*);
SQLITE_PRIVATE Table *sqlite3SrcListLookup(Parse*, SrcList*);
SQLITE_PRIVATE int sqlite3IsReadOnly(Parse*, Table*, int);
SQLITE_PRIVATE void sqlite3OpenTable(Parse*, int iCur, int iDb, Table*, int);
#if defined(SQLITE_ENABLE_UPDATE_DELETE_LIMIT) && !defined(SQLITE_OMIT_SUBQUERY)
SQLITE_PRIVATE Expr *sqlite3LimitWhere(Parse *, SrcList *, Expr *, ExprList *, Expr *, Expr *, char *);
#endif
SQLITE_PRIVATE void sqlite3DeleteFrom(Parse*, SrcList*, Expr*);
SQLITE_PRIVATE void sqlite3Update(Parse*, SrcList*, ExprList*, Expr*, int);
SQLITE_PRIVATE WhereInfo *sqlite3WhereBegin(Parse*,SrcList*,Expr*,ExprList*,ExprList*,u16,int);
SQLITE_PRIVATE void sqlite3WhereEnd(WhereInfo*);
SQLITE_PRIVATE int sqlite3ExprCodeGetColumn(Parse*, Table*, int, int, int, u8);
SQLITE_PRIVATE void sqlite3ExprCodeGetColumnOfTable(Vdbe*, Table*, int, int, int);
................................................................................
**     x = sqlite3GetVarint32( A, &B );
**     x = sqlite3PutVarint32( A, B );
**
**     x = getVarint32( A, B );
**     x = putVarint32( A, B );
**
*/
#define getVarint32(A,B)  (u8)((*(A)<(u8)0x80) ? ((B) = (u32)*(A)),1 : sqlite3GetVarint32((A), (u32 *)&(B)))

#define putVarint32(A,B)  (u8)(((u32)(B)<(u32)0x80) ? (*(A) = (unsigned char)(B)),1 : sqlite3PutVarint32((A), (B)))


#define getVarint    sqlite3GetVarint
#define putVarint    sqlite3PutVarint


SQLITE_PRIVATE const char *sqlite3IndexAffinityStr(Vdbe *, Index *);
SQLITE_PRIVATE void sqlite3TableAffinityStr(Vdbe *, Table *);
SQLITE_PRIVATE char sqlite3CompareAffinity(Expr *pExpr, char aff2);
................................................................................
  if( rc2==SQLITE_FULL || rc2==SQLITE_IOERR ){
    pPager->errCode = rc;
    pPager->eState = PAGER_ERROR;
  }
  return rc;
}



/*
** This routine ends a transaction. A transaction is usually ended by 
** either a COMMIT or a ROLLBACK operation. This routine may be called 
** after rollback of a hot-journal, or if an error occurs while opening
** the journal file or writing the very first journal-header of a
** database transaction.
** 
................................................................................
** database then the IO error code is returned to the user. If the 
** operation to finalize the journal file fails, then the code still
** tries to unlock the database file if not in exclusive mode. If the
** unlock operation fails as well, then the first error code related
** to the first error encountered (the journal finalization one) is
** returned.
*/
static int pager_end_transaction(Pager *pPager, int hasMaster){
  int rc = SQLITE_OK;      /* Error code from journal finalization operation */
  int rc2 = SQLITE_OK;     /* Error code from db file unlock operation */

  /* Do nothing if the pager does not have an open write transaction
  ** or at least a RESERVED lock. This function may be called when there
  ** is no write-transaction active but a RESERVED or greater lock is
  ** held under two circumstances:
................................................................................
  if( pagerUseWal(pPager) ){
    /* Drop the WAL write-lock, if any. Also, if the connection was in 
    ** locking_mode=exclusive mode but is no longer, drop the EXCLUSIVE 
    ** lock held on the database file.
    */
    rc2 = sqlite3WalEndWriteTransaction(pPager->pWal);
    assert( rc2==SQLITE_OK );









  }

  if( !pPager->exclusiveMode 
   && (!pagerUseWal(pPager) || sqlite3WalExclusiveMode(pPager->pWal, 0))
  ){
    rc2 = pagerUnlockDb(pPager, SHARED_LOCK);
    pPager->changeCountDone = 0;
  }
  pPager->eState = PAGER_READER;
................................................................................
    assert( assert_pager_state(pPager) );
    if( pPager->eState>=PAGER_WRITER_LOCKED ){
      sqlite3BeginBenignMalloc();
      sqlite3PagerRollback(pPager);
      sqlite3EndBenignMalloc();
    }else if( !pPager->exclusiveMode ){
      assert( pPager->eState==PAGER_READER );
      pager_end_transaction(pPager, 0);
    }
  }
  pager_unlock(pPager);
}

/*
** Parameter aData must point to a buffer of pPager->pageSize bytes
................................................................................
  }
  if( rc==SQLITE_OK
   && (pPager->eState>=PAGER_WRITER_DBMOD || pPager->eState==PAGER_OPEN)
  ){
    rc = sqlite3PagerSync(pPager);
  }
  if( rc==SQLITE_OK ){
    rc = pager_end_transaction(pPager, zMaster[0]!='\0');
    testcase( rc!=SQLITE_OK );
  }
  if( rc==SQLITE_OK && zMaster[0] && res ){
    /* If there was a master journal and this routine will return success,
    ** see if it is possible to delete the master journal.
    */
    rc = pager_delmaster(pPager, zMaster);
................................................................................
        }
      }
  #else
      rc = pager_incr_changecounter(pPager, 0);
  #endif
      if( rc!=SQLITE_OK ) goto commit_phase_one_exit;
  
      /* If this transaction has made the database smaller, then all pages
      ** being discarded by the truncation must be written to the journal
      ** file.
      **
      ** Before reading the pages with page numbers larger than the 
      ** current value of Pager.dbSize, set dbSize back to the value
      ** that it took at the start of the transaction. Otherwise, the
      ** calls to sqlite3PagerGet() return zeroed pages instead of 
      ** reading data from the database file.
      */
      if( pPager->dbSize<pPager->dbOrigSize 
       && pPager->journalMode!=PAGER_JOURNALMODE_OFF
      ){
        Pgno i;                                   /* Iterator variable */
        const Pgno iSkip = PAGER_MJ_PGNO(pPager); /* Pending lock page */
        const Pgno dbSize = pPager->dbSize;       /* Database image size */ 
        pPager->dbSize = pPager->dbOrigSize;
        for( i=dbSize+1; i<=pPager->dbOrigSize; i++ ){
          if( !sqlite3BitvecTest(pPager->pInJournal, i) && i!=iSkip ){
            PgHdr *pPage;             /* Page to journal */
            rc = sqlite3PagerGet(pPager, i, &pPage);
            if( rc!=SQLITE_OK ) goto commit_phase_one_exit;
            rc = sqlite3PagerWrite(pPage);
            sqlite3PagerUnref(pPage);
            if( rc!=SQLITE_OK ) goto commit_phase_one_exit;
          }
        }
        pPager->dbSize = dbSize;
      } 
  
      /* Write the master journal name into the journal file. If a master 
      ** journal file name has already been written to the journal file, 
      ** or if zMaster is NULL (no master journal), then this call is a no-op.
      */
      rc = writeMasterJournal(pPager, zMaster);
      if( rc!=SQLITE_OK ) goto commit_phase_one_exit;
  
................................................................................
  
      rc = pager_write_pagelist(pPager,sqlite3PcacheDirtyList(pPager->pPCache));
      if( rc!=SQLITE_OK ){
        assert( rc!=SQLITE_IOERR_BLOCKED );
        goto commit_phase_one_exit;
      }
      sqlite3PcacheCleanAll(pPager->pPCache);
  
      /* If the file on disk is not the same size as the database image,
      ** then use pager_truncate to grow or shrink the file here.
      */



      if( pPager->dbSize!=pPager->dbFileSize ){
        Pgno nNew = pPager->dbSize - (pPager->dbSize==PAGER_MJ_PGNO(pPager));
        assert( pPager->eState==PAGER_WRITER_DBMOD );
        rc = pager_truncate(pPager, nNew);
        if( rc!=SQLITE_OK ) goto commit_phase_one_exit;
      }
  
      /* Finally, sync the database file. */
................................................................................
  ){
    assert( pPager->journalOff==JOURNAL_HDR_SZ(pPager) || !pPager->journalOff );
    pPager->eState = PAGER_READER;
    return SQLITE_OK;
  }

  PAGERTRACE(("COMMIT %d\n", PAGERID(pPager)));
  rc = pager_end_transaction(pPager, pPager->setMaster);
  return pager_error(pPager, rc);
}

/*
** If a write transaction is open, then all changes made within the 
** transaction are reverted and the current write-transaction is closed.
** The pager falls back to PAGER_READER state if successful, or PAGER_ERROR
................................................................................
  assert( assert_pager_state(pPager) );
  if( pPager->eState==PAGER_ERROR ) return pPager->errCode;
  if( pPager->eState<=PAGER_READER ) return SQLITE_OK;

  if( pagerUseWal(pPager) ){
    int rc2;
    rc = sqlite3PagerSavepoint(pPager, SAVEPOINT_ROLLBACK, -1);
    rc2 = pager_end_transaction(pPager, pPager->setMaster);
    if( rc==SQLITE_OK ) rc = rc2;
  }else if( !isOpen(pPager->jfd) || pPager->eState==PAGER_WRITER_LOCKED ){
    int eState = pPager->eState;
    rc = pager_end_transaction(pPager, 0);
    if( !MEMDB && eState>PAGER_WRITER_LOCKED ){
      /* This can happen using journal_mode=off. Move the pager to the error 
      ** state to indicate that the contents of the cache may not be trusted.
      ** Any active readers will get SQLITE_ABORT.
      */
      pPager->errCode = SQLITE_ABORT;
      pPager->eState = PAGER_ERROR;
................................................................................
  sqlite3 *db;          /* Database connection currently using this Btree */
  BtCursor *pCursor;    /* A list of all open cursors */
  MemPage *pPage1;      /* First page of the database */
  u8 openFlags;         /* Flags to sqlite3BtreeOpen() */
#ifndef SQLITE_OMIT_AUTOVACUUM
  u8 autoVacuum;        /* True if auto-vacuum is enabled */
  u8 incrVacuum;        /* True if incr-vacuum is enabled */

#endif
  u8 inTransaction;     /* Transaction state */
  u8 max1bytePayload;   /* Maximum first byte of cell for a 1-byte payload */
  u16 btsFlags;         /* Boolean parameters.  See BTS_* macros below */
  u16 maxLocal;         /* Maximum local payload in non-LEAFDATA tables */
  u16 minLocal;         /* Minimum local payload in non-LEAFDATA tables */
  u16 maxLeaf;          /* Maximum local payload in a LEAFDATA table */
................................................................................
  /* If the btree is already in a write-transaction, or it
  ** is already in a read-transaction and a read-transaction
  ** is requested, this is a no-op.
  */
  if( p->inTrans==TRANS_WRITE || (p->inTrans==TRANS_READ && !wrflag) ){
    goto trans_begun;
  }


  /* Write transactions are not possible on a read-only database */
  if( (pBt->btsFlags & BTS_READ_ONLY)!=0 && wrflag ){
    rc = SQLITE_READONLY;
    goto trans_begun;
  }

................................................................................
    }
  }
  return rc;
}

/* Forward declaration required by incrVacuumStep(). */
static int allocateBtreePage(BtShared *, MemPage **, Pgno *, Pgno, u8);




/*
** Perform a single step of an incremental-vacuum. If successful,
** return SQLITE_OK. If there is no work to do (and therefore no
** point in calling this function again), return SQLITE_DONE.

**
** More specificly, this function attempts to re-organize the 
** database so that the last page of the file currently in use
** is no longer in use.
**



** If the nFin parameter is non-zero, this function assumes
** that the caller will keep calling incrVacuumStep() until
** it returns SQLITE_DONE or an error, and that nFin is the
** number of pages the database file will contain after this 
** process is complete.  If nFin is zero, it is assumed that
** incrVacuumStep() will be called a finite amount of times
** which may or may not empty the freelist.  A full autovacuum
** has nFin>0.  A "PRAGMA incremental_vacuum" has nFin==0.


*/
static int incrVacuumStep(BtShared *pBt, Pgno nFin, Pgno iLastPg){
  Pgno nFreeList;           /* Number of pages still on the free-list */
  int rc;

  assert( sqlite3_mutex_held(pBt->mutex) );
  assert( iLastPg>nFin );

  if( !PTRMAP_ISPAGE(pBt, iLastPg) && iLastPg!=PENDING_BYTE_PAGE(pBt) ){
................................................................................
      return rc;
    }
    if( eType==PTRMAP_ROOTPAGE ){
      return SQLITE_CORRUPT_BKPT;
    }

    if( eType==PTRMAP_FREEPAGE ){
      if( nFin==0 ){
        /* Remove the page from the files free-list. This is not required
        ** if nFin is non-zero. In that case, the free-list will be
        ** truncated to zero after this function returns, so it doesn't 
        ** matter if it still contains some garbage entries.
        */
        Pgno iFreePg;
        MemPage *pFreePg;
        rc = allocateBtreePage(pBt, &pFreePg, &iFreePg, iLastPg, 1);
        if( rc!=SQLITE_OK ){
          return rc;
        }
        assert( iFreePg==iLastPg );
        releasePage(pFreePg);
      }
    } else {
      Pgno iFreePg;             /* Index of free page to move pLastPg to */
      MemPage *pLastPg;



      rc = btreeGetPage(pBt, iLastPg, &pLastPg, 0);
      if( rc!=SQLITE_OK ){
        return rc;
      }

      /* If nFin is zero, this loop runs exactly once and page pLastPg
      ** is swapped with the first free page pulled off the free list.
      **
      ** On the other hand, if nFin is greater than zero, then keep
      ** looping until a free-page located within the first nFin pages
      ** of the file is found.
      */




      do {
        MemPage *pFreePg;
        rc = allocateBtreePage(pBt, &pFreePg, &iFreePg, 0, 0);
        if( rc!=SQLITE_OK ){
          releasePage(pLastPg);
          return rc;
        }
        releasePage(pFreePg);
      }while( nFin!=0 && iFreePg>nFin );
      assert( iFreePg<iLastPg );
      
      rc = sqlite3PagerWrite(pLastPg->pDbPage);
      if( rc==SQLITE_OK ){
        rc = relocatePage(pBt, pLastPg, eType, iPtrPage, iFreePg, nFin!=0);
      }
      releasePage(pLastPg);
      if( rc!=SQLITE_OK ){
        return rc;
      }
    }
  }

  if( nFin==0 ){

    iLastPg--;
    while( iLastPg==PENDING_BYTE_PAGE(pBt)||PTRMAP_ISPAGE(pBt, iLastPg) ){
      if( PTRMAP_ISPAGE(pBt, iLastPg) ){
        MemPage *pPg;
        rc = btreeGetPage(pBt, iLastPg, &pPg, 0);
        if( rc!=SQLITE_OK ){
          return rc;
        }
        rc = sqlite3PagerWrite(pPg->pDbPage);
        releasePage(pPg);
        if( rc!=SQLITE_OK ){
          return rc;
        }
      }
      iLastPg--;
    }
    sqlite3PagerTruncateImage(pBt->pPager, iLastPg);
    pBt->nPage = iLastPg;
  }
  return SQLITE_OK;
}
























/*
** A write-transaction must be opened before calling this function.
** It performs a single unit of work towards an incremental vacuum.
**
** If the incremental vacuum is finished after this function has run,
** SQLITE_DONE is returned. If it is not finished, but no error occurred,
................................................................................
  BtShared *pBt = p->pBt;

  sqlite3BtreeEnter(p);
  assert( pBt->inTransaction==TRANS_WRITE && p->inTrans==TRANS_WRITE );
  if( !pBt->autoVacuum ){
    rc = SQLITE_DONE;
  }else{







    invalidateAllOverflowCache(pBt);
    rc = incrVacuumStep(pBt, 0, btreePagecount(pBt));
    if( rc==SQLITE_OK ){
      rc = sqlite3PagerWrite(pBt->pPage1->pDbPage);
      put4byte(&pBt->pPage1->aData[28], pBt->nPage);



    }
  }
  sqlite3BtreeLeave(p);
  return rc;
}

/*
................................................................................

  assert( sqlite3_mutex_held(pBt->mutex) );
  invalidateAllOverflowCache(pBt);
  assert(pBt->autoVacuum);
  if( !pBt->incrVacuum ){
    Pgno nFin;         /* Number of pages in database after autovacuuming */
    Pgno nFree;        /* Number of pages on the freelist initially */
    Pgno nPtrmap;      /* Number of PtrMap pages to be freed */
    Pgno iFree;        /* The next page to be freed */
    int nEntry;        /* Number of entries on one ptrmap page */
    Pgno nOrig;        /* Database size before freeing */

    nOrig = btreePagecount(pBt);
    if( PTRMAP_ISPAGE(pBt, nOrig) || nOrig==PENDING_BYTE_PAGE(pBt) ){
      /* It is not possible to create a database for which the final page
      ** is either a pointer-map page or the pending-byte page. If one
      ** is encountered, this indicates corruption.
      */
      return SQLITE_CORRUPT_BKPT;
    }

    nFree = get4byte(&pBt->pPage1->aData[36]);
    nEntry = pBt->usableSize/5;
    nPtrmap = (nFree-nOrig+PTRMAP_PAGENO(pBt, nOrig)+nEntry)/nEntry;
    nFin = nOrig - nFree - nPtrmap;
    if( nOrig>PENDING_BYTE_PAGE(pBt) && nFin<PENDING_BYTE_PAGE(pBt) ){
      nFin--;
    }
    while( PTRMAP_ISPAGE(pBt, nFin) || nFin==PENDING_BYTE_PAGE(pBt) ){
      nFin--;
    }
    if( nFin>nOrig ) return SQLITE_CORRUPT_BKPT;

    for(iFree=nOrig; iFree>nFin && rc==SQLITE_OK; iFree--){
      rc = incrVacuumStep(pBt, nFin, iFree);
    }
    if( (rc==SQLITE_DONE || rc==SQLITE_OK) && nFree>0 ){
      rc = sqlite3PagerWrite(pBt->pPage1->pDbPage);
      put4byte(&pBt->pPage1->aData[32], 0);
      put4byte(&pBt->pPage1->aData[36], 0);
      put4byte(&pBt->pPage1->aData[28], nFin);
      sqlite3PagerTruncateImage(pBt->pPager, nFin);
      pBt->nPage = nFin;
    }
    if( rc!=SQLITE_OK ){
      sqlite3PagerRollback(pPager);
    }
  }

................................................................................
    if( pBt->autoVacuum ){
      rc = autoVacuumCommit(pBt);
      if( rc!=SQLITE_OK ){
        sqlite3BtreeLeave(p);
        return rc;
      }
    }



#endif
    rc = sqlite3PagerCommitPhaseOne(pBt->pPager, zMaster, 0);
    sqlite3BtreeLeave(p);
  }
  return rc;
}

................................................................................
** This function is called from both BtreeCommitPhaseTwo() and BtreeRollback()
** at the conclusion of a transaction.
*/
static void btreeEndTransaction(Btree *p){
  BtShared *pBt = p->pBt;
  assert( sqlite3BtreeHoldsMutex(p) );




  btreeClearHasContent(pBt);
  if( p->inTrans>TRANS_NONE && p->db->activeVdbeCnt>1 ){
    /* If there are other active statements that belong to this database
    ** handle, downgrade to a read-only transaction. The other statements
    ** may still be reading from the database.  */
    downgradeAllSharedCacheTableLocks(p);
    p->inTrans = TRANS_READ;
................................................................................
** is only used by auto-vacuum databases when allocating a new table.
*/
static int allocateBtreePage(
  BtShared *pBt, 
  MemPage **ppPage, 
  Pgno *pPgno, 
  Pgno nearby,
  u8 exact
){
  MemPage *pPage1;
  int rc;
  u32 n;     /* Number of pages on the freelist */
  u32 k;     /* Number of leaves on the trunk of the freelist */
  MemPage *pTrunk = 0;
  MemPage *pPrevTrunk = 0;
................................................................................
    u8 searchList = 0; /* If the free-list must be searched for 'nearby' */
    
    /* If the 'exact' parameter was true and a query of the pointer-map
    ** shows that the page 'nearby' is somewhere on the free-list, then
    ** the entire-list will be searched for that page.
    */
#ifndef SQLITE_OMIT_AUTOVACUUM

    if( exact && nearby<=mxPage ){
      u8 eType;
      assert( nearby>0 );
      assert( pBt->autoVacuum );
      rc = ptrmapGet(pBt, nearby, &eType, 0);
      if( rc ) return rc;
      if( eType==PTRMAP_FREEPAGE ){
        searchList = 1;
      }
      *pPgno = nearby;



    }
#endif

    /* Decrement the free-list count by 1. Set iTrunk to the index of the
    ** first free-list trunk page. iPrevTrunk is initially 1.
    */
    rc = sqlite3PagerWrite(pPage1->pDbPage);
................................................................................
        pTrunk = 0;
        TRACE(("ALLOCATE: %d trunk - %d free pages left\n", *pPgno, n-1));
      }else if( k>(u32)(pBt->usableSize/4 - 2) ){
        /* Value of k is out of range.  Database corruption */
        rc = SQLITE_CORRUPT_BKPT;
        goto end_allocate_page;
#ifndef SQLITE_OMIT_AUTOVACUUM
      }else if( searchList && nearby==iTrunk ){


        /* The list is being searched and this trunk page is the page
        ** to allocate, regardless of whether it has leaves.
        */
        assert( *pPgno==iTrunk );
        *ppPage = pTrunk;
        searchList = 0;
        rc = sqlite3PagerWrite(pTrunk->pDbPage);
        if( rc ){
          goto end_allocate_page;
        }
        if( k==0 ){
................................................................................
      }else if( k>0 ){
        /* Extract a leaf from the trunk */
        u32 closest;
        Pgno iPage;
        unsigned char *aData = pTrunk->aData;
        if( nearby>0 ){
          u32 i;
          int dist;




          closest = 0;





          dist = sqlite3AbsInt32(get4byte(&aData[8]) - nearby);
          for(i=1; i<k; i++){
            int d2 = sqlite3AbsInt32(get4byte(&aData[8+i*4]) - nearby);
            if( d2<dist ){
              closest = i;
              dist = d2;

            }
          }
        }else{
          closest = 0;
        }

        iPage = get4byte(&aData[8+closest*4]);
        testcase( iPage==mxPage );
        if( iPage>mxPage ){
          rc = SQLITE_CORRUPT_BKPT;
          goto end_allocate_page;
        }
        testcase( iPage==mxPage );
        if( !searchList || iPage==nearby ){


          int noContent;
          *pPgno = iPage;
          TRACE(("ALLOCATE: %d was leaf %d of %d on trunk %d"
                 ": %d more free pages\n",
                 *pPgno, closest+1, k, pTrunk->pgno, n-1));
          rc = sqlite3PagerWrite(pTrunk->pDbPage);
          if( rc ) goto end_allocate_page;
................................................................................
          searchList = 0;
        }
      }
      releasePage(pPrevTrunk);
      pPrevTrunk = 0;
    }while( searchList );
  }else{
    /* There are no pages on the freelist, so create a new page at the
    ** end of the file */


















    rc = sqlite3PagerWrite(pBt->pPage1->pDbPage);
    if( rc ) return rc;
    pBt->nPage++;
    if( pBt->nPage==PENDING_BYTE_PAGE(pBt) ) pBt->nPage++;

#ifndef SQLITE_OMIT_AUTOVACUUM
    if( pBt->autoVacuum && PTRMAP_ISPAGE(pBt, pBt->nPage) ){
................................................................................
      /* If *pPgno refers to a pointer-map page, allocate two new pages
      ** at the end of the file instead of one. The first allocated page
      ** becomes a new pointer-map page, the second is used by the caller.
      */
      MemPage *pPg = 0;
      TRACE(("ALLOCATE: %d from end of file (pointer-map page)\n", pBt->nPage));
      assert( pBt->nPage!=PENDING_BYTE_PAGE(pBt) );
      rc = btreeGetPage(pBt, pBt->nPage, &pPg, 1);
      if( rc==SQLITE_OK ){
        rc = sqlite3PagerWrite(pPg->pDbPage);
        releasePage(pPg);
      }
      if( rc ) return rc;
      pBt->nPage++;
      if( pBt->nPage==PENDING_BYTE_PAGE(pBt) ){ pBt->nPage++; }
    }
#endif
    put4byte(28 + (u8*)pBt->pPage1->aData, pBt->nPage);
    *pPgno = pBt->nPage;

    assert( *pPgno!=PENDING_BYTE_PAGE(pBt) );
    rc = btreeGetPage(pBt, *pPgno, ppPage, 1);
    if( rc ) return rc;
    rc = sqlite3PagerWrite((*ppPage)->pDbPage);
    if( rc!=SQLITE_OK ){
      releasePage(*ppPage);
    }
    TRACE(("ALLOCATE: %d from end of file\n", *pPgno));
  }
................................................................................
    }
    assert( pgnoRoot>=3 );

    /* Allocate a page. The page that currently resides at pgnoRoot will
    ** be moved to the allocated page (unless the allocated page happens
    ** to reside at pgnoRoot).
    */
    rc = allocateBtreePage(pBt, &pPageMove, &pgnoMove, pgnoRoot, 1);
    if( rc!=SQLITE_OK ){
      return rc;
    }

    if( pgnoMove!=pgnoRoot ){
      /* pgnoRoot is the page that will be used for the root-page of
      ** the new table (assuming an error did not occur). But we were
................................................................................
          if( nDestTruncate==(int)PENDING_BYTE_PAGE(p->pDest->pBt) ){
            nDestTruncate--;
          }
        }else{
          nDestTruncate = nSrcPage * (pgszSrc/pgszDest);
        }
        assert( nDestTruncate>0 );
        sqlite3PagerTruncateImage(pDestPager, nDestTruncate);

        if( pgszSrc<pgszDest ){
          /* If the source page-size is smaller than the destination page-size,
          ** two extra things may need to happen:
          **
          **   * The destination may need to be truncated, and
          **
          **   * Data stored on the pages immediately following the 
          **     pending-byte page in the source database may need to be
          **     copied into the destination database.
          */
          const i64 iSize = (i64)pgszSrc * (i64)nSrcPage;
          sqlite3_file * const pFile = sqlite3PagerFile(pDestPager);


          i64 iOff;
          i64 iEnd;

          assert( pFile );
          assert( nDestTruncate==0 
              || (i64)nDestTruncate*(i64)pgszDest >= iSize || (
                nDestTruncate==(int)(PENDING_BYTE_PAGE(p->pDest->pBt)-1)
             && iSize>=PENDING_BYTE && iSize<=PENDING_BYTE+pgszDest
          ));

          /* This call ensures that all data required to recreate the original
          ** database has been stored in the journal for pDestPager and the
          ** journal synced to disk. So at this point we may safely modify
          ** the database file in any way, knowing that if a power failure
          ** occurs, the original database will be reconstructed from the 
          ** journal file.  */












          rc = sqlite3PagerCommitPhaseOne(pDestPager, 0, 1);


          /* Write the extra pages and truncate the database file as required */
          iEnd = MIN(PENDING_BYTE + pgszDest, iSize);
          for(
            iOff=PENDING_BYTE+pgszSrc; 
            rc==SQLITE_OK && iOff<iEnd; 
            iOff+=pgszSrc
................................................................................
          }

          /* Sync the database file to disk. */
          if( rc==SQLITE_OK ){
            rc = sqlite3PagerSync(pDestPager);
          }
        }else{

          rc = sqlite3PagerCommitPhaseOne(pDestPager, 0, 0);
        }
    
        /* Finish committing the transaction to the destination database. */
        if( SQLITE_OK==rc
         && SQLITE_OK==(rc = sqlite3BtreeCommitPhaseTwo(p->pDest, 0))
        ){
................................................................................
  const char *zSpan,
  const char *zCol,
  const char *zTab,
  const char *zDb
){
  int n;
  for(n=0; ALWAYS(zSpan[n]) && zSpan[n]!='.'; n++){}
  if( zDb && sqlite3StrNICmp(zSpan, zDb, n)!=0 ){
    return 0;
  }
  zSpan += n+1;
  for(n=0; ALWAYS(zSpan[n]) && zSpan[n]!='.'; n++){}
  if( zTab && sqlite3StrNICmp(zSpan, zTab, n)!=0 ){
    return 0;
  }
  zSpan += n+1;
  if( zCol && sqlite3StrICmp(zSpan, zCol)!=0 ){
    return 0;
  }
  return 1;
................................................................................
    default: {
      assert( sqlite3_value_type(argv[0])==SQLITE_NULL );
      sqlite3_result_text(context, "NULL", 4, SQLITE_STATIC);
      break;
    }
  }
}



















































/*
** The hex() function.  Interpret the argument as a blob.  Return
** a hexadecimal rendering as text.
*/
static void hexFunc(
  sqlite3_context *context,
................................................................................
    FUNCTION(max,                0, 1, 1, 0                ),
    AGGREGATE(max,               1, 1, 1, minmaxStep,      minMaxFinalize ),
    FUNCTION2(typeof,            1, 0, 0, typeofFunc,  SQLITE_FUNC_TYPEOF),
    FUNCTION2(length,            1, 0, 0, lengthFunc,  SQLITE_FUNC_LENGTH),
    FUNCTION(instr,              2, 0, 0, instrFunc        ),
    FUNCTION(substr,             2, 0, 0, substrFunc       ),
    FUNCTION(substr,             3, 0, 0, substrFunc       ),


    FUNCTION(abs,                1, 0, 0, absFunc          ),
#ifndef SQLITE_OMIT_FLOATING_POINT
    FUNCTION(round,              1, 0, 0, roundFunc        ),
    FUNCTION(round,              2, 0, 0, roundFunc        ),
#endif
    FUNCTION(upper,              1, 0, 0, upperFunc        ),
    FUNCTION(lower,              1, 0, 0, lowerFunc        ),
................................................................................
  ** operator is UNION, EXCEPT, or INTERSECT (but not UNION ALL).
  */
  if( op==TK_ALL ){
    regPrev = 0;
  }else{
    int nExpr = p->pEList->nExpr;
    assert( nOrderBy>=nExpr || db->mallocFailed );
    regPrev = sqlite3GetTempRange(pParse, nExpr+1);

    sqlite3VdbeAddOp2(v, OP_Integer, 0, regPrev);
    pKeyDup = sqlite3DbMallocZero(db,
                  sizeof(*pKeyDup) + nExpr*(sizeof(CollSeq*)+1) );
    if( pKeyDup ){
      pKeyDup->aSortOrder = (u8*)&pKeyDup->aColl[nExpr];
      pKeyDup->nField = (u16)nExpr;
      pKeyDup->enc = ENC(db);
................................................................................
  sqlite3VdbeResolveLabel(v, labelCmpr);
  sqlite3VdbeAddOp4(v, OP_Permutation, 0, 0, 0, (char*)aPermute, P4_INTARRAY);
  sqlite3VdbeAddOp4(v, OP_Compare, destA.iSdst, destB.iSdst, nOrderBy,
                         (char*)pKeyMerge, P4_KEYINFO_HANDOFF);
  sqlite3VdbeChangeP5(v, OPFLAG_PERMUTE);
  sqlite3VdbeAddOp3(v, OP_Jump, addrAltB, addrAeqB, addrAgtB);

  /* Release temporary registers
  */
  if( regPrev ){
    sqlite3ReleaseTempRange(pParse, regPrev, nOrderBy+1);
  }

  /* Jump to the this point in order to terminate the query.
  */
  sqlite3VdbeResolveLabel(v, labelEnd);

  /* Set the number of output columns
  */
  if( pDest->eDest==SRT_Output ){
................................................................................
      sqlite3BtreeRollback(p, tripCode);
      db->aDb[i].inTrans = 0;
    }
  }
  sqlite3VtabRollback(db);
  sqlite3EndBenignMalloc();

  if( db->flags&SQLITE_InternChanges ){
    sqlite3ExpirePreparedStatements(db);
    sqlite3ResetAllSchemasOfConnection(db);
  }

  /* Any deferred constraint violations have now been resolved. */
  db->nDeferredCons = 0;

** Later we learn that _XOPEN_SOURCE is poorly or incorrectly
** implemented on some systems.  So we avoid defining it at all
** if it is already defined or if it is unneeded because we are
** not doing a threadsafe build.  Ticket #2681.
**
** See also ticket #2741.
*/
#if !defined(_XOPEN_SOURCE) && !defined(__DARWIN__) \
 && !defined(__APPLE__) && SQLITE_THREADSAFE
#  define _XOPEN_SOURCE 500  /* Needed to enable pthread recursive mutexes */
#endif

/*
** The TCL headers are only needed when compiling the TCL bindings.
*/
#if defined(SQLITE_TCL) || defined(TCLSH)
................................................................................
**
** See also: [sqlite3_libversion()],
** [sqlite3_libversion_number()], [sqlite3_sourceid()],
** [sqlite_version()] and [sqlite_source_id()].
*/
#define SQLITE_VERSION        "3.7.16"
#define SQLITE_VERSION_NUMBER 3007016
#define SQLITE_SOURCE_ID      "2013-02-26 18:54:18 cd8067238439638bcfd3966d55d2a3990f36d702"

/*
** 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
................................................................................
SQLITE_PRIVATE Select *sqlite3SelectNew(Parse*,ExprList*,SrcList*,Expr*,ExprList*,
                         Expr*,ExprList*,u16,Expr*,Expr*);
SQLITE_PRIVATE void sqlite3SelectDelete(sqlite3*, Select*);
SQLITE_PRIVATE Table *sqlite3SrcListLookup(Parse*, SrcList*);
SQLITE_PRIVATE int sqlite3IsReadOnly(Parse*, Table*, int);
SQLITE_PRIVATE void sqlite3OpenTable(Parse*, int iCur, int iDb, Table*, int);
#if defined(SQLITE_ENABLE_UPDATE_DELETE_LIMIT) && !defined(SQLITE_OMIT_SUBQUERY)
SQLITE_PRIVATE Expr *sqlite3LimitWhere(Parse*,SrcList*,Expr*,ExprList*,Expr*,Expr*,char*);
#endif
SQLITE_PRIVATE void sqlite3DeleteFrom(Parse*, SrcList*, Expr*);
SQLITE_PRIVATE void sqlite3Update(Parse*, SrcList*, ExprList*, Expr*, int);
SQLITE_PRIVATE WhereInfo *sqlite3WhereBegin(Parse*,SrcList*,Expr*,ExprList*,ExprList*,u16,int);
SQLITE_PRIVATE void sqlite3WhereEnd(WhereInfo*);
SQLITE_PRIVATE int sqlite3ExprCodeGetColumn(Parse*, Table*, int, int, int, u8);
SQLITE_PRIVATE void sqlite3ExprCodeGetColumnOfTable(Vdbe*, Table*, int, int, int);
................................................................................
**     x = sqlite3GetVarint32( A, &B );
**     x = sqlite3PutVarint32( A, B );
**
**     x = getVarint32( A, B );
**     x = putVarint32( A, B );
**
*/
#define getVarint32(A,B)  \
  (u8)((*(A)<(u8)0x80)?((B)=(u32)*(A)),1:sqlite3GetVarint32((A),(u32 *)&(B)))
#define putVarint32(A,B)  \
  (u8)(((u32)(B)<(u32)0x80)?(*(A)=(unsigned char)(B)),1:\
  sqlite3PutVarint32((A),(B)))
#define getVarint    sqlite3GetVarint
#define putVarint    sqlite3PutVarint


SQLITE_PRIVATE const char *sqlite3IndexAffinityStr(Vdbe *, Index *);
SQLITE_PRIVATE void sqlite3TableAffinityStr(Vdbe *, Table *);
SQLITE_PRIVATE char sqlite3CompareAffinity(Expr *pExpr, char aff2);
................................................................................
  if( rc2==SQLITE_FULL || rc2==SQLITE_IOERR ){
    pPager->errCode = rc;
    pPager->eState = PAGER_ERROR;
  }
  return rc;
}

static int pager_truncate(Pager *pPager, Pgno nPage);

/*
** This routine ends a transaction. A transaction is usually ended by 
** either a COMMIT or a ROLLBACK operation. This routine may be called 
** after rollback of a hot-journal, or if an error occurs while opening
** the journal file or writing the very first journal-header of a
** database transaction.
** 
................................................................................
** database then the IO error code is returned to the user. If the 
** operation to finalize the journal file fails, then the code still
** tries to unlock the database file if not in exclusive mode. If the
** unlock operation fails as well, then the first error code related
** to the first error encountered (the journal finalization one) is
** returned.
*/
static int pager_end_transaction(Pager *pPager, int hasMaster, int bCommit){
  int rc = SQLITE_OK;      /* Error code from journal finalization operation */
  int rc2 = SQLITE_OK;     /* Error code from db file unlock operation */

  /* Do nothing if the pager does not have an open write transaction
  ** or at least a RESERVED lock. This function may be called when there
  ** is no write-transaction active but a RESERVED or greater lock is
  ** held under two circumstances:
................................................................................
  if( pagerUseWal(pPager) ){
    /* Drop the WAL write-lock, if any. Also, if the connection was in 
    ** locking_mode=exclusive mode but is no longer, drop the EXCLUSIVE 
    ** lock held on the database file.
    */
    rc2 = sqlite3WalEndWriteTransaction(pPager->pWal);
    assert( rc2==SQLITE_OK );
  }else if( rc==SQLITE_OK && bCommit && pPager->dbFileSize>pPager->dbSize ){
    /* This branch is taken when committing a transaction in rollback-journal
    ** mode if the database file on disk is larger than the database image.
    ** At this point the journal has been finalized and the transaction 
    ** successfully committed, but the EXCLUSIVE lock is still held on the
    ** file. So it is safe to truncate the database file to its minimum
    ** required size.  */
    assert( pPager->eLock==EXCLUSIVE_LOCK );
    rc = pager_truncate(pPager, pPager->dbSize);
  }

  if( !pPager->exclusiveMode 
   && (!pagerUseWal(pPager) || sqlite3WalExclusiveMode(pPager->pWal, 0))
  ){
    rc2 = pagerUnlockDb(pPager, SHARED_LOCK);
    pPager->changeCountDone = 0;
  }
  pPager->eState = PAGER_READER;
................................................................................
    assert( assert_pager_state(pPager) );
    if( pPager->eState>=PAGER_WRITER_LOCKED ){
      sqlite3BeginBenignMalloc();
      sqlite3PagerRollback(pPager);
      sqlite3EndBenignMalloc();
    }else if( !pPager->exclusiveMode ){
      assert( pPager->eState==PAGER_READER );
      pager_end_transaction(pPager, 0, 0);
    }
  }
  pager_unlock(pPager);
}

/*
** Parameter aData must point to a buffer of pPager->pageSize bytes
................................................................................
  }
  if( rc==SQLITE_OK
   && (pPager->eState>=PAGER_WRITER_DBMOD || pPager->eState==PAGER_OPEN)
  ){
    rc = sqlite3PagerSync(pPager);
  }
  if( rc==SQLITE_OK ){
    rc = pager_end_transaction(pPager, zMaster[0]!='\0', 0);
    testcase( rc!=SQLITE_OK );
  }
  if( rc==SQLITE_OK && zMaster[0] && res ){
    /* If there was a master journal and this routine will return success,
    ** see if it is possible to delete the master journal.
    */
    rc = pager_delmaster(pPager, zMaster);
................................................................................
        }
      }
  #else
      rc = pager_incr_changecounter(pPager, 0);
  #endif
      if( rc!=SQLITE_OK ) goto commit_phase_one_exit;
  






























      /* Write the master journal name into the journal file. If a master 
      ** journal file name has already been written to the journal file, 
      ** or if zMaster is NULL (no master journal), then this call is a no-op.
      */
      rc = writeMasterJournal(pPager, zMaster);
      if( rc!=SQLITE_OK ) goto commit_phase_one_exit;
  
................................................................................
  
      rc = pager_write_pagelist(pPager,sqlite3PcacheDirtyList(pPager->pPCache));
      if( rc!=SQLITE_OK ){
        assert( rc!=SQLITE_IOERR_BLOCKED );
        goto commit_phase_one_exit;
      }
      sqlite3PcacheCleanAll(pPager->pPCache);

      /* If the file on disk is smaller than the database image, use 
      ** pager_truncate to grow the file here. This can happen if the database
      ** image was extended as part of the current transaction and then the
      ** last page in the db image moved to the free-list. In this case the
      ** last page is never written out to disk, leaving the database file
      ** undersized. Fix this now if it is the case.  */
      if( pPager->dbSize>pPager->dbFileSize ){
        Pgno nNew = pPager->dbSize - (pPager->dbSize==PAGER_MJ_PGNO(pPager));
        assert( pPager->eState==PAGER_WRITER_DBMOD );
        rc = pager_truncate(pPager, nNew);
        if( rc!=SQLITE_OK ) goto commit_phase_one_exit;
      }
  
      /* Finally, sync the database file. */
................................................................................
  ){
    assert( pPager->journalOff==JOURNAL_HDR_SZ(pPager) || !pPager->journalOff );
    pPager->eState = PAGER_READER;
    return SQLITE_OK;
  }

  PAGERTRACE(("COMMIT %d\n", PAGERID(pPager)));
  rc = pager_end_transaction(pPager, pPager->setMaster, 1);
  return pager_error(pPager, rc);
}

/*
** If a write transaction is open, then all changes made within the 
** transaction are reverted and the current write-transaction is closed.
** The pager falls back to PAGER_READER state if successful, or PAGER_ERROR
................................................................................
  assert( assert_pager_state(pPager) );
  if( pPager->eState==PAGER_ERROR ) return pPager->errCode;
  if( pPager->eState<=PAGER_READER ) return SQLITE_OK;

  if( pagerUseWal(pPager) ){
    int rc2;
    rc = sqlite3PagerSavepoint(pPager, SAVEPOINT_ROLLBACK, -1);
    rc2 = pager_end_transaction(pPager, pPager->setMaster, 0);
    if( rc==SQLITE_OK ) rc = rc2;
  }else if( !isOpen(pPager->jfd) || pPager->eState==PAGER_WRITER_LOCKED ){
    int eState = pPager->eState;
    rc = pager_end_transaction(pPager, 0, 0);
    if( !MEMDB && eState>PAGER_WRITER_LOCKED ){
      /* This can happen using journal_mode=off. Move the pager to the error 
      ** state to indicate that the contents of the cache may not be trusted.
      ** Any active readers will get SQLITE_ABORT.
      */
      pPager->errCode = SQLITE_ABORT;
      pPager->eState = PAGER_ERROR;
................................................................................
  sqlite3 *db;          /* Database connection currently using this Btree */
  BtCursor *pCursor;    /* A list of all open cursors */
  MemPage *pPage1;      /* First page of the database */
  u8 openFlags;         /* Flags to sqlite3BtreeOpen() */
#ifndef SQLITE_OMIT_AUTOVACUUM
  u8 autoVacuum;        /* True if auto-vacuum is enabled */
  u8 incrVacuum;        /* True if incr-vacuum is enabled */
  u8 bDoTruncate;       /* True to truncate db on commit */
#endif
  u8 inTransaction;     /* Transaction state */
  u8 max1bytePayload;   /* Maximum first byte of cell for a 1-byte payload */
  u16 btsFlags;         /* Boolean parameters.  See BTS_* macros below */
  u16 maxLocal;         /* Maximum local payload in non-LEAFDATA tables */
  u16 minLocal;         /* Minimum local payload in non-LEAFDATA tables */
  u16 maxLeaf;          /* Maximum local payload in a LEAFDATA table */
................................................................................
  /* If the btree is already in a write-transaction, or it
  ** is already in a read-transaction and a read-transaction
  ** is requested, this is a no-op.
  */
  if( p->inTrans==TRANS_WRITE || (p->inTrans==TRANS_READ && !wrflag) ){
    goto trans_begun;
  }
  assert( pBt->bDoTruncate==0 );

  /* Write transactions are not possible on a read-only database */
  if( (pBt->btsFlags & BTS_READ_ONLY)!=0 && wrflag ){
    rc = SQLITE_READONLY;
    goto trans_begun;
  }

................................................................................
    }
  }
  return rc;
}

/* Forward declaration required by incrVacuumStep(). */
static int allocateBtreePage(BtShared *, MemPage **, Pgno *, Pgno, u8);
#define BTALLOC_ANY   0           /* Allocate any page */
#define BTALLOC_EXACT 1           /* Allocate exact page if possible */
#define BTALLOC_LE    2           /* Allocate any page <= the parameter */

/*
** Perform a single step of an incremental-vacuum. If successful, return
** SQLITE_OK. If there is no work to do (and therefore no point in 
** calling this function again), return SQLITE_DONE. Or, if an error 
** occurs, return some other error code.
**
** More specificly, this function attempts to re-organize the database so 
** that the last page of the file currently in use is no longer in use.

**
** Parameter nFin is the number of pages that this database would contain
** were this function called until it returns SQLITE_DONE.
**
** If the bCommit parameter is non-zero, this function assumes that the 
** caller will keep calling incrVacuumStep() until it returns SQLITE_DONE 






** or an error. bCommit is passed true for an auto-vacuum-on-commmit 
** operation, or false for an incremental vacuum.
*/
static int incrVacuumStep(BtShared *pBt, Pgno nFin, Pgno iLastPg, int bCommit){
  Pgno nFreeList;           /* Number of pages still on the free-list */
  int rc;

  assert( sqlite3_mutex_held(pBt->mutex) );
  assert( iLastPg>nFin );

  if( !PTRMAP_ISPAGE(pBt, iLastPg) && iLastPg!=PENDING_BYTE_PAGE(pBt) ){
................................................................................
      return rc;
    }
    if( eType==PTRMAP_ROOTPAGE ){
      return SQLITE_CORRUPT_BKPT;
    }

    if( eType==PTRMAP_FREEPAGE ){
      if( bCommit==0 ){
        /* Remove the page from the files free-list. This is not required
        ** if bCommit is non-zero. In that case, the free-list will be
        ** truncated to zero after this function returns, so it doesn't 
        ** matter if it still contains some garbage entries.
        */
        Pgno iFreePg;
        MemPage *pFreePg;
        rc = allocateBtreePage(pBt, &pFreePg, &iFreePg, iLastPg, BTALLOC_EXACT);
        if( rc!=SQLITE_OK ){
          return rc;
        }
        assert( iFreePg==iLastPg );
        releasePage(pFreePg);
      }
    } else {
      Pgno iFreePg;             /* Index of free page to move pLastPg to */
      MemPage *pLastPg;
      u8 eMode = BTALLOC_ANY;   /* Mode parameter for allocateBtreePage() */
      Pgno iNear = 0;           /* nearby parameter for allocateBtreePage() */

      rc = btreeGetPage(pBt, iLastPg, &pLastPg, 0);
      if( rc!=SQLITE_OK ){
        return rc;
      }

      /* If bCommit is zero, this loop runs exactly once and page pLastPg
      ** is swapped with the first free page pulled off the free list.
      **
      ** On the other hand, if bCommit is greater than zero, then keep
      ** looping until a free-page located within the first nFin pages
      ** of the file is found.
      */
      if( bCommit==0 ){
        eMode = BTALLOC_LE;
        iNear = nFin;
      }
      do {
        MemPage *pFreePg;
        rc = allocateBtreePage(pBt, &pFreePg, &iFreePg, iNear, eMode);
        if( rc!=SQLITE_OK ){
          releasePage(pLastPg);
          return rc;
        }
        releasePage(pFreePg);
      }while( bCommit && iFreePg>nFin );
      assert( iFreePg<iLastPg );
      


      rc = relocatePage(pBt, pLastPg, eType, iPtrPage, iFreePg, nFin!=0);

      releasePage(pLastPg);
      if( rc!=SQLITE_OK ){
        return rc;
      }
    }
  }

  if( bCommit==0 ){
    do {
      iLastPg--;
    }while( iLastPg==PENDING_BYTE_PAGE(pBt) || PTRMAP_ISPAGE(pBt, iLastPg) );

    pBt->bDoTruncate = 1;













    pBt->nPage = iLastPg;
  }
  return SQLITE_OK;
}

/*
** The database opened by the first argument is an auto-vacuum database
** nOrig pages in size containing nFree free pages. Return the expected 
** size of the database in pages following an auto-vacuum operation.
*/
static Pgno finalDbSize(BtShared *pBt, Pgno nOrig, Pgno nFree){
  int nEntry;                     /* Number of entries on one ptrmap page */
  Pgno nPtrmap;                   /* Number of PtrMap pages to be freed */
  Pgno nFin;                      /* Return value */

  nEntry = pBt->usableSize/5;
  nPtrmap = (nFree-nOrig+PTRMAP_PAGENO(pBt, nOrig)+nEntry)/nEntry;
  nFin = nOrig - nFree - nPtrmap;
  if( nOrig>PENDING_BYTE_PAGE(pBt) && nFin<PENDING_BYTE_PAGE(pBt) ){
    nFin--;
  }
  while( PTRMAP_ISPAGE(pBt, nFin) || nFin==PENDING_BYTE_PAGE(pBt) ){
    nFin--;
  }

  return nFin;
}

/*
** A write-transaction must be opened before calling this function.
** It performs a single unit of work towards an incremental vacuum.
**
** If the incremental vacuum is finished after this function has run,
** SQLITE_DONE is returned. If it is not finished, but no error occurred,
................................................................................
  BtShared *pBt = p->pBt;

  sqlite3BtreeEnter(p);
  assert( pBt->inTransaction==TRANS_WRITE && p->inTrans==TRANS_WRITE );
  if( !pBt->autoVacuum ){
    rc = SQLITE_DONE;
  }else{
    Pgno nOrig = btreePagecount(pBt);
    Pgno nFree = get4byte(&pBt->pPage1->aData[36]);
    Pgno nFin = finalDbSize(pBt, nOrig, nFree);

    if( nOrig<nFin ){
      rc = SQLITE_CORRUPT_BKPT;
    }else if( nFree>0 ){
      invalidateAllOverflowCache(pBt);
      rc = incrVacuumStep(pBt, nFin, nOrig, 0);
      if( rc==SQLITE_OK ){
        rc = sqlite3PagerWrite(pBt->pPage1->pDbPage);
        put4byte(&pBt->pPage1->aData[28], pBt->nPage);
      }
    }else{
      rc = SQLITE_DONE;
    }
  }
  sqlite3BtreeLeave(p);
  return rc;
}

/*
................................................................................

  assert( sqlite3_mutex_held(pBt->mutex) );
  invalidateAllOverflowCache(pBt);
  assert(pBt->autoVacuum);
  if( !pBt->incrVacuum ){
    Pgno nFin;         /* Number of pages in database after autovacuuming */
    Pgno nFree;        /* Number of pages on the freelist initially */

    Pgno iFree;        /* The next page to be freed */

    Pgno nOrig;        /* Database size before freeing */

    nOrig = btreePagecount(pBt);
    if( PTRMAP_ISPAGE(pBt, nOrig) || nOrig==PENDING_BYTE_PAGE(pBt) ){
      /* It is not possible to create a database for which the final page
      ** is either a pointer-map page or the pending-byte page. If one
      ** is encountered, this indicates corruption.
      */
      return SQLITE_CORRUPT_BKPT;
    }

    nFree = get4byte(&pBt->pPage1->aData[36]);
    nFin = finalDbSize(pBt, nOrig, nFree);








    if( nFin>nOrig ) return SQLITE_CORRUPT_BKPT;

    for(iFree=nOrig; iFree>nFin && rc==SQLITE_OK; iFree--){
      rc = incrVacuumStep(pBt, nFin, iFree, 1);
    }
    if( (rc==SQLITE_DONE || rc==SQLITE_OK) && nFree>0 ){
      rc = sqlite3PagerWrite(pBt->pPage1->pDbPage);
      put4byte(&pBt->pPage1->aData[32], 0);
      put4byte(&pBt->pPage1->aData[36], 0);
      put4byte(&pBt->pPage1->aData[28], nFin);
      pBt->bDoTruncate = 1;
      pBt->nPage = nFin;
    }
    if( rc!=SQLITE_OK ){
      sqlite3PagerRollback(pPager);
    }
  }

................................................................................
    if( pBt->autoVacuum ){
      rc = autoVacuumCommit(pBt);
      if( rc!=SQLITE_OK ){
        sqlite3BtreeLeave(p);
        return rc;
      }
    }
    if( pBt->bDoTruncate ){
      sqlite3PagerTruncateImage(pBt->pPager, pBt->nPage);
    }
#endif
    rc = sqlite3PagerCommitPhaseOne(pBt->pPager, zMaster, 0);
    sqlite3BtreeLeave(p);
  }
  return rc;
}

................................................................................
** This function is called from both BtreeCommitPhaseTwo() and BtreeRollback()
** at the conclusion of a transaction.
*/
static void btreeEndTransaction(Btree *p){
  BtShared *pBt = p->pBt;
  assert( sqlite3BtreeHoldsMutex(p) );

#ifndef SQLITE_OMIT_AUTOVACUUM
  pBt->bDoTruncate = 0;
#endif
  btreeClearHasContent(pBt);
  if( p->inTrans>TRANS_NONE && p->db->activeVdbeCnt>1 ){
    /* If there are other active statements that belong to this database
    ** handle, downgrade to a read-only transaction. The other statements
    ** may still be reading from the database.  */
    downgradeAllSharedCacheTableLocks(p);
    p->inTrans = TRANS_READ;
................................................................................
** is only used by auto-vacuum databases when allocating a new table.
*/
static int allocateBtreePage(
  BtShared *pBt, 
  MemPage **ppPage, 
  Pgno *pPgno, 
  Pgno nearby,
  u8 eMode
){
  MemPage *pPage1;
  int rc;
  u32 n;     /* Number of pages on the freelist */
  u32 k;     /* Number of leaves on the trunk of the freelist */
  MemPage *pTrunk = 0;
  MemPage *pPrevTrunk = 0;
................................................................................
    u8 searchList = 0; /* If the free-list must be searched for 'nearby' */
    
    /* If the 'exact' parameter was true and a query of the pointer-map
    ** shows that the page 'nearby' is somewhere on the free-list, then
    ** the entire-list will be searched for that page.
    */
#ifndef SQLITE_OMIT_AUTOVACUUM
    if( eMode==BTALLOC_EXACT ){
      if( nearby<=mxPage ){
        u8 eType;
        assert( nearby>0 );
        assert( pBt->autoVacuum );
        rc = ptrmapGet(pBt, nearby, &eType, 0);
        if( rc ) return rc;
        if( eType==PTRMAP_FREEPAGE ){
          searchList = 1;
        }

      }
    }else if( eMode==BTALLOC_LE ){
      searchList = 1;
    }
#endif

    /* Decrement the free-list count by 1. Set iTrunk to the index of the
    ** first free-list trunk page. iPrevTrunk is initially 1.
    */
    rc = sqlite3PagerWrite(pPage1->pDbPage);
................................................................................
        pTrunk = 0;
        TRACE(("ALLOCATE: %d trunk - %d free pages left\n", *pPgno, n-1));
      }else if( k>(u32)(pBt->usableSize/4 - 2) ){
        /* Value of k is out of range.  Database corruption */
        rc = SQLITE_CORRUPT_BKPT;
        goto end_allocate_page;
#ifndef SQLITE_OMIT_AUTOVACUUM
      }else if( searchList 
            && (nearby==iTrunk || (iTrunk<nearby && eMode==BTALLOC_LE)) 
      ){
        /* The list is being searched and this trunk page is the page
        ** to allocate, regardless of whether it has leaves.
        */
        *pPgno = iTrunk;
        *ppPage = pTrunk;
        searchList = 0;
        rc = sqlite3PagerWrite(pTrunk->pDbPage);
        if( rc ){
          goto end_allocate_page;
        }
        if( k==0 ){
................................................................................
      }else if( k>0 ){
        /* Extract a leaf from the trunk */
        u32 closest;
        Pgno iPage;
        unsigned char *aData = pTrunk->aData;
        if( nearby>0 ){
          u32 i;
          closest = 0;
          if( eMode==BTALLOC_LE ){
            for(i=0; i<k; i++){
              iPage = get4byte(&aData[8+i*4]);
              if( iPage<=nearby ){
                closest = i;
                break;
              }
            }
          }else{
            int dist;
            dist = sqlite3AbsInt32(get4byte(&aData[8]) - nearby);
            for(i=1; i<k; i++){
              int d2 = sqlite3AbsInt32(get4byte(&aData[8+i*4]) - nearby);
              if( d2<dist ){
                closest = i;
                dist = d2;
              }
            }
          }
        }else{
          closest = 0;
        }

        iPage = get4byte(&aData[8+closest*4]);
        testcase( iPage==mxPage );
        if( iPage>mxPage ){
          rc = SQLITE_CORRUPT_BKPT;
          goto end_allocate_page;
        }
        testcase( iPage==mxPage );
        if( !searchList 
         || (iPage==nearby || (iPage<nearby && eMode==BTALLOC_LE)) 
        ){
          int noContent;
          *pPgno = iPage;
          TRACE(("ALLOCATE: %d was leaf %d of %d on trunk %d"
                 ": %d more free pages\n",
                 *pPgno, closest+1, k, pTrunk->pgno, n-1));
          rc = sqlite3PagerWrite(pTrunk->pDbPage);
          if( rc ) goto end_allocate_page;
................................................................................
          searchList = 0;
        }
      }
      releasePage(pPrevTrunk);
      pPrevTrunk = 0;
    }while( searchList );
  }else{
    /* There are no pages on the freelist, so append a new page to the
    ** database image.
    **
    ** Normally, new pages allocated by this block can be requested from the
    ** pager layer with the 'no-content' flag set. This prevents the pager
    ** from trying to read the pages content from disk. However, if the
    ** current transaction has already run one or more incremental-vacuum
    ** steps, then the page we are about to allocate may contain content
    ** that is required in the event of a rollback. In this case, do
    ** not set the no-content flag. This causes the pager to load and journal
    ** the current page content before overwriting it.
    **
    ** Note that the pager will not actually attempt to load or journal 
    ** content for any page that really does lie past the end of the database
    ** file on disk. So the effects of disabling the no-content optimization
    ** here are confined to those pages that lie between the end of the
    ** database image and the end of the database file.
    */
    int bNoContent = (0==pBt->bDoTruncate);

    rc = sqlite3PagerWrite(pBt->pPage1->pDbPage);
    if( rc ) return rc;
    pBt->nPage++;
    if( pBt->nPage==PENDING_BYTE_PAGE(pBt) ) pBt->nPage++;

#ifndef SQLITE_OMIT_AUTOVACUUM
    if( pBt->autoVacuum && PTRMAP_ISPAGE(pBt, pBt->nPage) ){
................................................................................
      /* If *pPgno refers to a pointer-map page, allocate two new pages
      ** at the end of the file instead of one. The first allocated page
      ** becomes a new pointer-map page, the second is used by the caller.
      */
      MemPage *pPg = 0;
      TRACE(("ALLOCATE: %d from end of file (pointer-map page)\n", pBt->nPage));
      assert( pBt->nPage!=PENDING_BYTE_PAGE(pBt) );
      rc = btreeGetPage(pBt, pBt->nPage, &pPg, bNoContent);
      if( rc==SQLITE_OK ){
        rc = sqlite3PagerWrite(pPg->pDbPage);
        releasePage(pPg);
      }
      if( rc ) return rc;
      pBt->nPage++;
      if( pBt->nPage==PENDING_BYTE_PAGE(pBt) ){ pBt->nPage++; }
    }
#endif
    put4byte(28 + (u8*)pBt->pPage1->aData, pBt->nPage);
    *pPgno = pBt->nPage;

    assert( *pPgno!=PENDING_BYTE_PAGE(pBt) );
    rc = btreeGetPage(pBt, *pPgno, ppPage, bNoContent);
    if( rc ) return rc;
    rc = sqlite3PagerWrite((*ppPage)->pDbPage);
    if( rc!=SQLITE_OK ){
      releasePage(*ppPage);
    }
    TRACE(("ALLOCATE: %d from end of file\n", *pPgno));
  }
................................................................................
    }
    assert( pgnoRoot>=3 );

    /* Allocate a page. The page that currently resides at pgnoRoot will
    ** be moved to the allocated page (unless the allocated page happens
    ** to reside at pgnoRoot).
    */
    rc = allocateBtreePage(pBt, &pPageMove, &pgnoMove, pgnoRoot, BTALLOC_EXACT);
    if( rc!=SQLITE_OK ){
      return rc;
    }

    if( pgnoMove!=pgnoRoot ){
      /* pgnoRoot is the page that will be used for the root-page of
      ** the new table (assuming an error did not occur). But we were
................................................................................
          if( nDestTruncate==(int)PENDING_BYTE_PAGE(p->pDest->pBt) ){
            nDestTruncate--;
          }
        }else{
          nDestTruncate = nSrcPage * (pgszSrc/pgszDest);
        }
        assert( nDestTruncate>0 );


        if( pgszSrc<pgszDest ){
          /* If the source page-size is smaller than the destination page-size,
          ** two extra things may need to happen:
          **
          **   * The destination may need to be truncated, and
          **
          **   * Data stored on the pages immediately following the 
          **     pending-byte page in the source database may need to be
          **     copied into the destination database.
          */
          const i64 iSize = (i64)pgszSrc * (i64)nSrcPage;
          sqlite3_file * const pFile = sqlite3PagerFile(pDestPager);
          Pgno iPg;
          int nDstPage;
          i64 iOff;
          i64 iEnd;

          assert( pFile );
          assert( nDestTruncate==0 
              || (i64)nDestTruncate*(i64)pgszDest >= iSize || (
                nDestTruncate==(int)(PENDING_BYTE_PAGE(p->pDest->pBt)-1)
             && iSize>=PENDING_BYTE && iSize<=PENDING_BYTE+pgszDest
          ));

          /* This block ensures that all data required to recreate the original
          ** database has been stored in the journal for pDestPager and the
          ** journal synced to disk. So at this point we may safely modify
          ** the database file in any way, knowing that if a power failure
          ** occurs, the original database will be reconstructed from the 
          ** journal file.  */
          sqlite3PagerPagecount(pDestPager, &nDstPage);
          for(iPg=nDestTruncate; rc==SQLITE_OK && iPg<=(Pgno)nDstPage; iPg++){
            if( iPg!=PENDING_BYTE_PAGE(p->pDest->pBt) ){
              DbPage *pPg;
              rc = sqlite3PagerGet(pDestPager, iPg, &pPg);
              if( rc==SQLITE_OK ){
                rc = sqlite3PagerWrite(pPg);
                sqlite3PagerUnref(pPg);
              }
            }
          }
          if( rc==SQLITE_OK ){
            rc = sqlite3PagerCommitPhaseOne(pDestPager, 0, 1);
          }

          /* Write the extra pages and truncate the database file as required */
          iEnd = MIN(PENDING_BYTE + pgszDest, iSize);
          for(
            iOff=PENDING_BYTE+pgszSrc; 
            rc==SQLITE_OK && iOff<iEnd; 
            iOff+=pgszSrc
................................................................................
          }

          /* Sync the database file to disk. */
          if( rc==SQLITE_OK ){
            rc = sqlite3PagerSync(pDestPager);
          }
        }else{
          sqlite3PagerTruncateImage(pDestPager, nDestTruncate);
          rc = sqlite3PagerCommitPhaseOne(pDestPager, 0, 0);
        }
    
        /* Finish committing the transaction to the destination database. */
        if( SQLITE_OK==rc
         && SQLITE_OK==(rc = sqlite3BtreeCommitPhaseTwo(p->pDest, 0))
        ){
................................................................................
  const char *zSpan,
  const char *zCol,
  const char *zTab,
  const char *zDb
){
  int n;
  for(n=0; ALWAYS(zSpan[n]) && zSpan[n]!='.'; n++){}
  if( zDb && (sqlite3StrNICmp(zSpan, zDb, n)!=0 || zDb[n]!=0) ){
    return 0;
  }
  zSpan += n+1;
  for(n=0; ALWAYS(zSpan[n]) && zSpan[n]!='.'; n++){}
  if( zTab && (sqlite3StrNICmp(zSpan, zTab, n)!=0 || zTab[n]!=0) ){
    return 0;
  }
  zSpan += n+1;
  if( zCol && sqlite3StrICmp(zSpan, zCol)!=0 ){
    return 0;
  }
  return 1;
................................................................................
    default: {
      assert( sqlite3_value_type(argv[0])==SQLITE_NULL );
      sqlite3_result_text(context, "NULL", 4, SQLITE_STATIC);
      break;
    }
  }
}

/*
** The unicode() function.  Return the integer unicode code-point value
** for the first character of the input string. 
*/
static void unicodeFunc(
  sqlite3_context *context,
  int argc,
  sqlite3_value **argv
){
  const unsigned char *z = sqlite3_value_text(argv[0]);
  if( z && z[0] ) sqlite3_result_int(context, sqlite3Utf8Read(&z));
}

/*
** The char() function takes zero or more arguments, each of which is
** an integer.  It constructs a string where each character of the string
** is the unicode character for the corresponding integer argument.
*/
static void charFunc(
  sqlite3_context *context,
  int argc,
  sqlite3_value **argv
){
  unsigned char *z, *zOut;
  int i;
  zOut = z = sqlite3_malloc( argc*4 );
  if( z==0 ){
    sqlite3_result_error_nomem(context);
    return;
  }
  for(i=0; i<argc; i++){
    sqlite3_int64 x;
    unsigned c;
    x = sqlite3_value_int64(argv[i]);
    if( x<0 || x>0x10ffff ) x = 0xfffd;
    c = (unsigned)(x & 0x1fffff);
    if( c<=0xFFFF ){
      *zOut++ = (u8)(c&0x00FF);
      *zOut++ = (u8)((c>>8)&0x00FF);
    }else{
      if( c>=0xd800 && c<=0xdbff ) c = 0xfffd;
      *zOut++ = (u8)(((c>>10)&0x003F) + (((c-0x10000)>>10)&0x00C0));
      *zOut++ = (u8)(0x00D8 + (((c-0x10000)>>18)&0x03));
      *zOut++ = (u8)(c&0x00FF);
      *zOut++ = (u8)(0x00DC + ((c>>8)&0x03));
    }
  }
  sqlite3_result_text16le(context, (char*)z, (int)(zOut-z), sqlite3_free);
}

/*
** The hex() function.  Interpret the argument as a blob.  Return
** a hexadecimal rendering as text.
*/
static void hexFunc(
  sqlite3_context *context,
................................................................................
    FUNCTION(max,                0, 1, 1, 0                ),
    AGGREGATE(max,               1, 1, 1, minmaxStep,      minMaxFinalize ),
    FUNCTION2(typeof,            1, 0, 0, typeofFunc,  SQLITE_FUNC_TYPEOF),
    FUNCTION2(length,            1, 0, 0, lengthFunc,  SQLITE_FUNC_LENGTH),
    FUNCTION(instr,              2, 0, 0, instrFunc        ),
    FUNCTION(substr,             2, 0, 0, substrFunc       ),
    FUNCTION(substr,             3, 0, 0, substrFunc       ),
    FUNCTION(unicode,            1, 0, 0, unicodeFunc      ),
    FUNCTION(char,              -1, 0, 0, charFunc         ),
    FUNCTION(abs,                1, 0, 0, absFunc          ),
#ifndef SQLITE_OMIT_FLOATING_POINT
    FUNCTION(round,              1, 0, 0, roundFunc        ),
    FUNCTION(round,              2, 0, 0, roundFunc        ),
#endif
    FUNCTION(upper,              1, 0, 0, upperFunc        ),
    FUNCTION(lower,              1, 0, 0, lowerFunc        ),
................................................................................
  ** operator is UNION, EXCEPT, or INTERSECT (but not UNION ALL).
  */
  if( op==TK_ALL ){
    regPrev = 0;
  }else{
    int nExpr = p->pEList->nExpr;
    assert( nOrderBy>=nExpr || db->mallocFailed );
    regPrev = pParse->nMem+1;
    pParse->nMem += nExpr+1;
    sqlite3VdbeAddOp2(v, OP_Integer, 0, regPrev);
    pKeyDup = sqlite3DbMallocZero(db,
                  sizeof(*pKeyDup) + nExpr*(sizeof(CollSeq*)+1) );
    if( pKeyDup ){
      pKeyDup->aSortOrder = (u8*)&pKeyDup->aColl[nExpr];
      pKeyDup->nField = (u16)nExpr;
      pKeyDup->enc = ENC(db);
................................................................................
  sqlite3VdbeResolveLabel(v, labelCmpr);
  sqlite3VdbeAddOp4(v, OP_Permutation, 0, 0, 0, (char*)aPermute, P4_INTARRAY);
  sqlite3VdbeAddOp4(v, OP_Compare, destA.iSdst, destB.iSdst, nOrderBy,
                         (char*)pKeyMerge, P4_KEYINFO_HANDOFF);
  sqlite3VdbeChangeP5(v, OPFLAG_PERMUTE);
  sqlite3VdbeAddOp3(v, OP_Jump, addrAltB, addrAeqB, addrAgtB);







  /* Jump to the this point in order to terminate the query.
  */
  sqlite3VdbeResolveLabel(v, labelEnd);

  /* Set the number of output columns
  */
  if( pDest->eDest==SRT_Output ){
................................................................................
      sqlite3BtreeRollback(p, tripCode);
      db->aDb[i].inTrans = 0;
    }
  }
  sqlite3VtabRollback(db);
  sqlite3EndBenignMalloc();

  if( (db->flags&SQLITE_InternChanges)!=0 && db->init.busy==0 ){
    sqlite3ExpirePreparedStatements(db);
    sqlite3ResetAllSchemasOfConnection(db);
  }

  /* Any deferred constraint violations have now been resolved. */
  db->nDeferredCons = 0;

**
** See also: [sqlite3_libversion()],
** [sqlite3_libversion_number()], [sqlite3_sourceid()],
** [sqlite_version()] and [sqlite_source_id()].
*/
#define SQLITE_VERSION        "3.7.16"
#define SQLITE_VERSION_NUMBER 3007016
#define SQLITE_SOURCE_ID      "2013-02-15 04:21:01 843e1c543aabab8cd62f28742d5818887d36bcb7"

/*
** 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

**
** See also: [sqlite3_libversion()],
** [sqlite3_libversion_number()], [sqlite3_sourceid()],
** [sqlite_version()] and [sqlite_source_id()].
*/
#define SQLITE_VERSION        "3.7.16"
#define SQLITE_VERSION_NUMBER 3007016
#define SQLITE_SOURCE_ID      "2013-02-26 18:54:18 cd8067238439638bcfd3966d55d2a3990f36d702"

/*
** 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

    <li>Support automatic value conversions for columns with a declared type of BIGUINT, INTEGER8, INTEGER16, INTEGER32, INTEGER64, SMALLUINT, TINYSINT, UNSIGNEDINTEGER, UNSIGNEDINTEGER8, UNSIGNEDINTEGER16, UNSIGNEDINTEGER32, UNSIGNEDINTEGER64, INT8, INT16, INT32, INT64, UINT, UINT8, UINT16, UINT32, UINT64, or ULONG.</li>
    <li>Add BindUInt32AsInt64 connection flag to force binding of UInt32 values as Int64 instead. Pursuant to [c010fa6584].</li>
    <li>Add BindAllAsText connection flag to force binding of all values as text.</li>
    <li>Remove AUTOINCREMENT from the column type name map.&nbsp;<b>** Potentially Incompatible Change **</b></li>
    <li>Avoid throwing overflow exceptions from the SQLite3.GetValue method for integral column types. Partial fix for [c010fa6584].&nbsp;<b>** Potentially Incompatible Change **</b></li>
    <li>Use the legacy connection closing algorithm when built with the INTEROP_LEGACY_CLOSE compile-time option.</li>
    <li>Support using the directory containing the primary managed-only assembly as the basis for native library pre-loading.</li>

</ul>
<p>
    <b>1.0.84.0 - January 9, 2013</b>
</p>
<ul>
    <li>Updated to <a href="http://www.sqlite.org/releaselog/3_7_15_2.html">SQLite 3.7.15.2</a>.</li>
    <li>Explicitly dispose of all SQLiteCommand objects managed by the DbDataAdapter class. Fix for [6434e23a0f].</li>

    <li>Support automatic value conversions for columns with a declared type of BIGUINT, INTEGER8, INTEGER16, INTEGER32, INTEGER64, SMALLUINT, TINYSINT, UNSIGNEDINTEGER, UNSIGNEDINTEGER8, UNSIGNEDINTEGER16, UNSIGNEDINTEGER32, UNSIGNEDINTEGER64, INT8, INT16, INT32, INT64, UINT, UINT8, UINT16, UINT32, UINT64, or ULONG.</li>
    <li>Add BindUInt32AsInt64 connection flag to force binding of UInt32 values as Int64 instead. Pursuant to [c010fa6584].</li>
    <li>Add BindAllAsText connection flag to force binding of all values as text.</li>
    <li>Remove AUTOINCREMENT from the column type name map.&nbsp;<b>** Potentially Incompatible Change **</b></li>
    <li>Avoid throwing overflow exceptions from the SQLite3.GetValue method for integral column types. Partial fix for [c010fa6584].&nbsp;<b>** Potentially Incompatible Change **</b></li>
    <li>Use the legacy connection closing algorithm when built with the INTEROP_LEGACY_CLOSE compile-time option.</li>
    <li>Support using the directory containing the primary managed-only assembly as the basis for native library pre-loading.</li>
    <li>Still further enhancements to the build and test automation.</li>
</ul>
<p>
    <b>1.0.84.0 - January 9, 2013</b>
</p>
<ul>
    <li>Updated to <a href="http://www.sqlite.org/releaselog/3_7_15_2.html">SQLite 3.7.15.2</a>.</li>
    <li>Explicitly dispose of all SQLiteCommand objects managed by the DbDataAdapter class. Fix for [6434e23a0f].</li>

    <li>Support automatic value conversions for columns with a declared type of BIGUINT, INTEGER8, INTEGER16, INTEGER32, INTEGER64, SMALLUINT, TINYSINT, UNSIGNEDINTEGER, UNSIGNEDINTEGER8, UNSIGNEDINTEGER16, UNSIGNEDINTEGER32, UNSIGNEDINTEGER64, INT8, INT16, INT32, INT64, UINT, UINT8, UINT16, UINT32, UINT64, or ULONG.</li>
    <li>Add BindUInt32AsInt64 connection flag to force binding of UInt32 values as Int64 instead. Pursuant to [c010fa6584].</li>
    <li>Add BindAllAsText connection flag to force binding of all values as text.</li>
    <li>Remove AUTOINCREMENT from the column type name map.&nbsp;<b>** Potentially Incompatible Change **</b></li>
    <li>Avoid throwing overflow exceptions from the SQLite3.GetValue method for integral column types. Partial fix for [c010fa6584].&nbsp;<b>** Potentially Incompatible Change **</b></li>
    <li>Use the legacy connection closing algorithm when built with the INTEROP_LEGACY_CLOSE compile-time option.</li>
    <li>Support using the directory containing the primary managed-only assembly as the basis for native library pre-loading.</li>

</ul>
<p>
    <b>1.0.84.0 - January 9, 2013</b>
</p>
<ul>
    <li>Updated to [http://www.sqlite.org/releaselog/3_7_15_2.html|SQLite 3.7.15.2].</li>
    <li>Explicitly dispose of all SQLiteCommand objects managed by the DbDataAdapter class. Fix for [6434e23a0f].</li>

    <li>Support automatic value conversions for columns with a declared type of BIGUINT, INTEGER8, INTEGER16, INTEGER32, INTEGER64, SMALLUINT, TINYSINT, UNSIGNEDINTEGER, UNSIGNEDINTEGER8, UNSIGNEDINTEGER16, UNSIGNEDINTEGER32, UNSIGNEDINTEGER64, INT8, INT16, INT32, INT64, UINT, UINT8, UINT16, UINT32, UINT64, or ULONG.</li>
    <li>Add BindUInt32AsInt64 connection flag to force binding of UInt32 values as Int64 instead. Pursuant to [c010fa6584].</li>
    <li>Add BindAllAsText connection flag to force binding of all values as text.</li>
    <li>Remove AUTOINCREMENT from the column type name map.&nbsp;<b>** Potentially Incompatible Change **</b></li>
    <li>Avoid throwing overflow exceptions from the SQLite3.GetValue method for integral column types. Partial fix for [c010fa6584].&nbsp;<b>** Potentially Incompatible Change **</b></li>
    <li>Use the legacy connection closing algorithm when built with the INTEROP_LEGACY_CLOSE compile-time option.</li>
    <li>Support using the directory containing the primary managed-only assembly as the basis for native library pre-loading.</li>
    <li>Still further enhancements to the build and test automation.</li>
</ul>
<p>
    <b>1.0.84.0 - January 9, 2013</b>
</p>
<ul>
    <li>Updated to [http://www.sqlite.org/releaselog/3_7_15_2.html|SQLite 3.7.15.2].</li>
    <li>Explicitly dispose of all SQLiteCommand objects managed by the DbDataAdapter class. Fix for [6434e23a0f].</li>