SQLite

Check-in [ca9d86baf7]
Login

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

Overview
Comment:Optimization: Convert an ORDER BY clause into a no-op if the query also contains a GROUP BY clause that will force the same output order.
Downloads: Tarball | ZIP archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: ca9d86baf70f210d331ce93102177c8005c494cb
User & Date: drh 2010-04-26 19:17:27.000
Context
2010-04-30
05:57
Zero the "immediate FK constraint counter" associated with a statement object when sqlite3_reset() is called. Fix for [c39ff61c43]. (check-in: f660be615a user: dan tags: trunk)
2010-04-27
01:56
Merge in recent changes from the trunk (check-in: 7a0ac682c3 user: drh tags: wal)
2010-04-26
19:17
Optimization: Convert an ORDER BY clause into a no-op if the query also contains a GROUP BY clause that will force the same output order. (check-in: ca9d86baf7 user: drh tags: trunk)
17:36
Change the default_cache_size pragma to always store a positive value. (check-in: 36fb2cae75 user: drh tags: trunk)
Changes
Unified Diff Ignore Whitespace Patch
Changes to src/expr.c.
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483

























3484
3485
3486
3487
3488
3489
3490
** expressions are the same.  But if you get a 0 or 1 return, then you
** can be sure the expressions are the same.  In the places where
** this routine is used, it does not hurt to get an extra 2 - that
** just might result in some slightly slower code.  But returning
** an incorrect 0 or 1 could lead to a malfunction.
*/
int sqlite3ExprCompare(Expr *pA, Expr *pB){
  int i;
  if( pA==0||pB==0 ){
    return pB==pA ? 0 : 2;
  }
  assert( !ExprHasAnyProperty(pA, EP_TokenOnly|EP_Reduced) );
  assert( !ExprHasAnyProperty(pB, EP_TokenOnly|EP_Reduced) );
  if( ExprHasProperty(pA, EP_xIsSelect) || ExprHasProperty(pB, EP_xIsSelect) ){
    return 2;
  }
  if( (pA->flags & EP_Distinct)!=(pB->flags & EP_Distinct) ) return 2;
  if( pA->op!=pB->op ) return 2;
  if( sqlite3ExprCompare(pA->pLeft, pB->pLeft) ) return 2;
  if( sqlite3ExprCompare(pA->pRight, pB->pRight) ) return 2;

  if( pA->x.pList && pB->x.pList ){
    if( pA->x.pList->nExpr!=pB->x.pList->nExpr ) return 2;
    for(i=0; i<pA->x.pList->nExpr; i++){
      Expr *pExprA = pA->x.pList->a[i].pExpr;
      Expr *pExprB = pB->x.pList->a[i].pExpr;
      if( sqlite3ExprCompare(pExprA, pExprB) ) return 2;
    }
  }else if( pA->x.pList || pB->x.pList ){
    return 2;
  }

  if( pA->iTable!=pB->iTable || pA->iColumn!=pB->iColumn ) return 2;
  if( ExprHasProperty(pA, EP_IntValue) ){
    if( !ExprHasProperty(pB, EP_IntValue) || pA->u.iValue!=pB->u.iValue ){
      return 2;
    }
  }else if( pA->op!=TK_COLUMN && pA->u.zToken ){
    if( ExprHasProperty(pB, EP_IntValue) || NEVER(pB->u.zToken==0) ) return 2;
    if( sqlite3StrICmp(pA->u.zToken,pB->u.zToken)!=0 ){
      return 2;
    }
  }
  if( (pA->flags & EP_ExpCollate)!=(pB->flags & EP_ExpCollate) ) return 1;
  if( (pA->flags & EP_ExpCollate)!=0 && pA->pColl!=pB->pColl ) return 2;
  return 0;
}



























/*
** Add a new element to the pAggInfo->aCol[] array.  Return the index of
** the new element.  Return a negative number if malloc fails.
*/
static int addAggInfoColumn(sqlite3 *db, AggInfo *pInfo){
  int i;







<












|
<
<
<
<
<
<
<
<
<
<
<
















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







3436
3437
3438
3439
3440
3441
3442

3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455











3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
** expressions are the same.  But if you get a 0 or 1 return, then you
** can be sure the expressions are the same.  In the places where
** this routine is used, it does not hurt to get an extra 2 - that
** just might result in some slightly slower code.  But returning
** an incorrect 0 or 1 could lead to a malfunction.
*/
int sqlite3ExprCompare(Expr *pA, Expr *pB){

  if( pA==0||pB==0 ){
    return pB==pA ? 0 : 2;
  }
  assert( !ExprHasAnyProperty(pA, EP_TokenOnly|EP_Reduced) );
  assert( !ExprHasAnyProperty(pB, EP_TokenOnly|EP_Reduced) );
  if( ExprHasProperty(pA, EP_xIsSelect) || ExprHasProperty(pB, EP_xIsSelect) ){
    return 2;
  }
  if( (pA->flags & EP_Distinct)!=(pB->flags & EP_Distinct) ) return 2;
  if( pA->op!=pB->op ) return 2;
  if( sqlite3ExprCompare(pA->pLeft, pB->pLeft) ) return 2;
  if( sqlite3ExprCompare(pA->pRight, pB->pRight) ) return 2;
  if( sqlite3ExprListCompare(pA->x.pList, pB->x.pList) ) return 2;











  if( pA->iTable!=pB->iTable || pA->iColumn!=pB->iColumn ) return 2;
  if( ExprHasProperty(pA, EP_IntValue) ){
    if( !ExprHasProperty(pB, EP_IntValue) || pA->u.iValue!=pB->u.iValue ){
      return 2;
    }
  }else if( pA->op!=TK_COLUMN && pA->u.zToken ){
    if( ExprHasProperty(pB, EP_IntValue) || NEVER(pB->u.zToken==0) ) return 2;
    if( sqlite3StrICmp(pA->u.zToken,pB->u.zToken)!=0 ){
      return 2;
    }
  }
  if( (pA->flags & EP_ExpCollate)!=(pB->flags & EP_ExpCollate) ) return 1;
  if( (pA->flags & EP_ExpCollate)!=0 && pA->pColl!=pB->pColl ) return 2;
  return 0;
}

/*
** Compare two ExprList objects.  Return 0 if they are identical and 
** non-zero if they differ in any way.
**
** This routine might return non-zero for equivalent ExprLists.  The
** only consequence will be disabled optimizations.  But this routine
** must never return 0 if the two ExprList objects are different, or
** a malfunction will result.
**
** Two NULL pointers are considered to be the same.  But a NULL pointer
** always differs from a non-NULL pointer.
*/
int sqlite3ExprListCompare(ExprList *pA, ExprList *pB){
  int i;
  if( pA==0 && pB==0 ) return 0;
  if( pA==0 || pB==0 ) return 1;
  if( pA->nExpr!=pB->nExpr ) return 1;
  for(i=0; i<pA->nExpr; i++){
    Expr *pExprA = pA->a[i].pExpr;
    Expr *pExprB = pB->a[i].pExpr;
    if( pA->a[i].sortOrder!=pB->a[i].sortOrder ) return 1;
    if( sqlite3ExprCompare(pExprA, pExprB) ) return 1;
  }
  return 0;
}

/*
** Add a new element to the pAggInfo->aCol[] array.  Return the index of
** the new element.  Return a negative number if malloc fails.
*/
static int addAggInfoColumn(sqlite3 *db, AggInfo *pInfo){
  int i;
Changes to src/select.c.
3713
3714
3715
3716
3717
3718
3719












3720
3721
3722
3723
3724
3725
3726
  assert( p->pGroupBy==0 || (p->selFlags & SF_Aggregate)!=0 );
  if( (p->selFlags & (SF_Distinct|SF_Aggregate))==SF_Distinct ){
    p->pGroupBy = sqlite3ExprListDup(db, p->pEList, 0);
    pGroupBy = p->pGroupBy;
    p->selFlags &= ~SF_Distinct;
    isDistinct = 0;
  }













  /* If there is an ORDER BY clause, then this sorting
  ** index might end up being unused if the data can be 
  ** extracted in pre-sorted order.  If that is the case, then the
  ** OP_OpenEphemeral instruction will be changed to an OP_Noop once
  ** we figure out that the sorting index is not needed.  The addrSortIndex
  ** variable is used to facilitate that change.







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







3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
  assert( p->pGroupBy==0 || (p->selFlags & SF_Aggregate)!=0 );
  if( (p->selFlags & (SF_Distinct|SF_Aggregate))==SF_Distinct ){
    p->pGroupBy = sqlite3ExprListDup(db, p->pEList, 0);
    pGroupBy = p->pGroupBy;
    p->selFlags &= ~SF_Distinct;
    isDistinct = 0;
  }

  /* If there is both a GROUP BY and an ORDER BY clause and they are
  ** identical, then disable the ORDER BY clause since the GROUP BY
  ** will cause elements to come out in the correct order.  This is
  ** an optimization - the correct answer should result regardless.
  ** Use the SQLITE_GroupByOrder flag with SQLITE_TESTCTRL_OPTIMIZER
  ** to disable this optimization for testing purposes.
  */
  if( sqlite3ExprListCompare(p->pGroupBy, pOrderBy)==0
         && (db->flags & SQLITE_GroupByOrder)==0 ){
    pOrderBy = 0;
  }

  /* If there is an ORDER BY clause, then this sorting
  ** index might end up being unused if the data can be 
  ** extracted in pre-sorted order.  If that is the case, then the
  ** OP_OpenEphemeral instruction will be changed to an OP_Noop once
  ** we figure out that the sorting index is not needed.  The addrSortIndex
  ** variable is used to facilitate that change.
Changes to src/sqliteInt.h.
921
922
923
924
925
926
927

928
929
930
931
932
933
934
935
** These must be the low-order bits of the flags field.
*/
#define SQLITE_QueryFlattener 0x01        /* Disable query flattening */
#define SQLITE_ColumnCache    0x02        /* Disable the column cache */
#define SQLITE_IndexSort      0x04        /* Disable indexes for sorting */
#define SQLITE_IndexSearch    0x08        /* Disable indexes for searching */
#define SQLITE_IndexCover     0x10        /* Disable index covering table */

#define SQLITE_OptMask        0x1f        /* Mask of all disablable opts */

/*
** Possible values for the sqlite.magic field.
** The numbers are obtained at random and have no special meaning, other
** than being distinct from one another.
*/
#define SQLITE_MAGIC_OPEN     0xa029a697  /* Database is open */







>
|







921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
** These must be the low-order bits of the flags field.
*/
#define SQLITE_QueryFlattener 0x01        /* Disable query flattening */
#define SQLITE_ColumnCache    0x02        /* Disable the column cache */
#define SQLITE_IndexSort      0x04        /* Disable indexes for sorting */
#define SQLITE_IndexSearch    0x08        /* Disable indexes for searching */
#define SQLITE_IndexCover     0x10        /* Disable index covering table */
#define SQLITE_GroupByOrder   0x20        /* Disable GROUPBY cover of ORDERBY */
#define SQLITE_OptMask        0xff        /* Mask of all disablable opts */

/*
** Possible values for the sqlite.magic field.
** The numbers are obtained at random and have no special meaning, other
** than being distinct from one another.
*/
#define SQLITE_MAGIC_OPEN     0xa029a697  /* Database is open */
2685
2686
2687
2688
2689
2690
2691

2692
2693
2694
2695
2696
2697
2698
Index *sqlite3FindIndex(sqlite3*,const char*, const char*);
void sqlite3UnlinkAndDeleteTable(sqlite3*,int,const char*);
void sqlite3UnlinkAndDeleteIndex(sqlite3*,int,const char*);
void sqlite3Vacuum(Parse*);
int sqlite3RunVacuum(char**, sqlite3*);
char *sqlite3NameFromToken(sqlite3*, Token*);
int sqlite3ExprCompare(Expr*, Expr*);

void sqlite3ExprAnalyzeAggregates(NameContext*, Expr*);
void sqlite3ExprAnalyzeAggList(NameContext*,ExprList*);
Vdbe *sqlite3GetVdbe(Parse*);
void sqlite3PrngSaveState(void);
void sqlite3PrngRestoreState(void);
void sqlite3PrngResetState(void);
void sqlite3RollbackAll(sqlite3*);







>







2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
Index *sqlite3FindIndex(sqlite3*,const char*, const char*);
void sqlite3UnlinkAndDeleteTable(sqlite3*,int,const char*);
void sqlite3UnlinkAndDeleteIndex(sqlite3*,int,const char*);
void sqlite3Vacuum(Parse*);
int sqlite3RunVacuum(char**, sqlite3*);
char *sqlite3NameFromToken(sqlite3*, Token*);
int sqlite3ExprCompare(Expr*, Expr*);
int sqlite3ExprListCompare(ExprList*, ExprList*);
void sqlite3ExprAnalyzeAggregates(NameContext*, Expr*);
void sqlite3ExprAnalyzeAggList(NameContext*,ExprList*);
Vdbe *sqlite3GetVdbe(Parse*);
void sqlite3PrngSaveState(void);
void sqlite3PrngRestoreState(void);
void sqlite3PrngResetState(void);
void sqlite3RollbackAll(sqlite3*);