selfuncs.c

来自「PostgreSQL7.4.6 for Linux」· C语言 代码 · 共 2,384 行 · 第 1/5 页

		 */
		Oid			eqopr = get_opclass_member(opclass, BTEqualStrategyNumber);
		List	   *eqargs;

		if (eqopr == InvalidOid)
			elog(ERROR, "no = operator for opclass %u", opclass);
		eqargs = makeList2(var, prefix);
		result = DatumGetFloat8(DirectFunctionCall4(eqsel,
													PointerGetDatum(root),
												 ObjectIdGetDatum(eqopr),
												 PointerGetDatum(eqargs),
											   Int32GetDatum(varRelid)));
	}
	else
	{
		/*
		 * Not exact-match pattern.  We estimate selectivity of the fixed
		 * prefix and remainder of pattern separately, then combine the
		 * two.
		 */
		Selectivity prefixsel;
		Selectivity restsel;
		Selectivity selec;

		if (pstatus == Pattern_Prefix_Partial)
			prefixsel = prefix_selectivity(root, var, opclass, prefix);
		else
			prefixsel = 1.0;
		restsel = pattern_selectivity(rest, ptype);
		selec = prefixsel * restsel;
		/* result should be in range, but make sure... */
		CLAMP_PROBABILITY(selec);
		result = selec;
	}

	if (prefix)
	{
		pfree(DatumGetPointer(prefix->constvalue));
		pfree(prefix);
	}

	return result;
}

/*
 *		regexeqsel		- Selectivity of regular-expression pattern match.
 */
Datum
regexeqsel(PG_FUNCTION_ARGS)
{
	PG_RETURN_FLOAT8(patternsel(fcinfo, Pattern_Type_Regex));
}

/*
 *		icregexeqsel	- Selectivity of case-insensitive regex match.
 */
Datum
icregexeqsel(PG_FUNCTION_ARGS)
{
	PG_RETURN_FLOAT8(patternsel(fcinfo, Pattern_Type_Regex_IC));
}

/*
 *		likesel			- Selectivity of LIKE pattern match.
 */
Datum
likesel(PG_FUNCTION_ARGS)
{
	PG_RETURN_FLOAT8(patternsel(fcinfo, Pattern_Type_Like));
}

/*
 *		iclikesel			- Selectivity of ILIKE pattern match.
 */
Datum
iclikesel(PG_FUNCTION_ARGS)
{
	PG_RETURN_FLOAT8(patternsel(fcinfo, Pattern_Type_Like_IC));
}

/*
 *		regexnesel		- Selectivity of regular-expression pattern non-match.
 */
Datum
regexnesel(PG_FUNCTION_ARGS)
{
	double		result;

	result = patternsel(fcinfo, Pattern_Type_Regex);
	result = 1.0 - result;
	PG_RETURN_FLOAT8(result);
}

/*
 *		icregexnesel	- Selectivity of case-insensitive regex non-match.
 */
Datum
icregexnesel(PG_FUNCTION_ARGS)
{
	double		result;

	result = patternsel(fcinfo, Pattern_Type_Regex_IC);
	result = 1.0 - result;
	PG_RETURN_FLOAT8(result);
}

/*
 *		nlikesel		- Selectivity of LIKE pattern non-match.
 */
Datum
nlikesel(PG_FUNCTION_ARGS)
{
	double		result;

	result = patternsel(fcinfo, Pattern_Type_Like);
	result = 1.0 - result;
	PG_RETURN_FLOAT8(result);
}

/*
 *		icnlikesel		- Selectivity of ILIKE pattern non-match.
 */
Datum
icnlikesel(PG_FUNCTION_ARGS)
{
	double		result;

	result = patternsel(fcinfo, Pattern_Type_Like_IC);
	result = 1.0 - result;
	PG_RETURN_FLOAT8(result);
}

/*
 *		booltestsel		- Selectivity of BooleanTest Node.
 */
Selectivity
booltestsel(Query *root, BoolTestType booltesttype, Node *arg,
			int varRelid, JoinType jointype)
{
	Var		   *var;
	Oid			relid;
	HeapTuple	statsTuple;
	Datum	   *values;
	int			nvalues;
	float4	   *numbers;
	int			nnumbers;
	double		selec;

	/*
	 * Ignore any binary-compatible relabeling (probably unnecessary, but
	 * can't hurt)
	 */
	if (IsA(arg, RelabelType))
		arg = (Node *) ((RelabelType *) arg)->arg;

	if (IsA(arg, Var) &&
		(varRelid == 0 || varRelid == ((Var *) arg)->varno))
		var = (Var *) arg;
	else
	{
		/*
		 * If argument is not a Var, we can't get statistics for it, but
		 * perhaps clause_selectivity can do something with it.  We ignore
		 * the possibility of a NULL value when using clause_selectivity,
		 * and just assume the value is either TRUE or FALSE.
		 */
		switch (booltesttype)
		{
			case IS_UNKNOWN:
				selec = DEFAULT_UNK_SEL;
				break;
			case IS_NOT_UNKNOWN:
				selec = DEFAULT_NOT_UNK_SEL;
				break;
			case IS_TRUE:
			case IS_NOT_FALSE:
				selec = (double) clause_selectivity(root, arg,
													varRelid, jointype);
				break;
			case IS_FALSE:
			case IS_NOT_TRUE:
				selec = 1.0 - (double) clause_selectivity(root, arg,
													 varRelid, jointype);
				break;
			default:
				elog(ERROR, "unrecognized booltesttype: %d",
					 (int) booltesttype);
				selec = 0.0;	/* Keep compiler quiet */
				break;
		}
		return (Selectivity) selec;
	}

	/* get stats for the attribute, if available */
	relid = getrelid(var->varno, root->rtable);
	if (relid == InvalidOid)
		statsTuple = NULL;
	else
		statsTuple = SearchSysCache(STATRELATT,
									ObjectIdGetDatum(relid),
									Int16GetDatum(var->varattno),
									0, 0);

	if (HeapTupleIsValid(statsTuple))
	{
		Form_pg_statistic stats;
		double		freq_null;

		stats = (Form_pg_statistic) GETSTRUCT(statsTuple);

		freq_null = stats->stanullfrac;

		if (get_attstatsslot(statsTuple, var->vartype, var->vartypmod,
							 STATISTIC_KIND_MCV, InvalidOid,
							 &values, &nvalues,
							 &numbers, &nnumbers)
			&& nnumbers > 0)
		{
			double		freq_true;
			double		freq_false;

			/*
			 * Get first MCV frequency and derive frequency for true.
			 */
			if (DatumGetBool(values[0]))
				freq_true = numbers[0];
			else
				freq_true = 1.0 - numbers[0] - freq_null;

			/*
			 * Next derive freqency for false. Then use these as
			 * appropriate to derive frequency for each case.
			 */
			freq_false = 1.0 - freq_true - freq_null;

			switch (booltesttype)
			{
				case IS_UNKNOWN:
					/* select only NULL values */
					selec = freq_null;
					break;
				case IS_NOT_UNKNOWN:
					/* select non-NULL values */
					selec = 1.0 - freq_null;
					break;
				case IS_TRUE:
					/* select only TRUE values */
					selec = freq_true;
					break;
				case IS_NOT_TRUE:
					/* select non-TRUE values */
					selec = 1.0 - freq_true;
					break;
				case IS_FALSE:
					/* select only FALSE values */
					selec = freq_false;
					break;
				case IS_NOT_FALSE:
					/* select non-FALSE values */
					selec = 1.0 - freq_false;
					break;
				default:
					elog(ERROR, "unrecognized booltesttype: %d",
						 (int) booltesttype);
					selec = 0.0;	/* Keep compiler quiet */
					break;
			}

			free_attstatsslot(var->vartype, values, nvalues,
							  numbers, nnumbers);
		}
		else
		{
			/*
			 * No most-common-value info available. Still have null
			 * fraction information, so use it for IS [NOT] UNKNOWN.
			 * Otherwise adjust for null fraction and assume an even split
			 * for boolean tests.
			 */
			switch (booltesttype)
			{
				case IS_UNKNOWN:

					/*
					 * Use freq_null directly.
					 */
					selec = freq_null;
					break;
				case IS_NOT_UNKNOWN:

					/*
					 * Select not unknown (not null) values. Calculate
					 * from freq_null.
					 */
					selec = 1.0 - freq_null;
					break;
				case IS_TRUE:
				case IS_NOT_TRUE:
				case IS_FALSE:
				case IS_NOT_FALSE:
					selec = (1.0 - freq_null) / 2.0;
					break;
				default:
					elog(ERROR, "unrecognized booltesttype: %d",
						 (int) booltesttype);
					selec = 0.0;	/* Keep compiler quiet */
					break;
			}
		}

		ReleaseSysCache(statsTuple);
	}
	else
	{
		/*
		 * No VACUUM ANALYZE stats available, so use a default value.
		 * (Note: not much point in recursing to clause_selectivity here.)
		 */
		switch (booltesttype)
		{
			case IS_UNKNOWN:
				selec = DEFAULT_UNK_SEL;
				break;
			case IS_NOT_UNKNOWN:
				selec = DEFAULT_NOT_UNK_SEL;
				break;
			case IS_TRUE:
			case IS_NOT_TRUE:
			case IS_FALSE:
			case IS_NOT_FALSE:
				selec = DEFAULT_BOOL_SEL;
				break;
			default:
				elog(ERROR, "unrecognized booltesttype: %d",
					 (int) booltesttype);
				selec = 0.0;	/* Keep compiler quiet */
				break;
		}
	}

	/* result should be in range, but make sure... */
	CLAMP_PROBABILITY(selec);

	return (Selectivity) selec;
}

/*
 *		nulltestsel		- Selectivity of NullTest Node.
 */
Selectivity
nulltestsel(Query *root, NullTestType nulltesttype, Node *arg, int varRelid)
{
	Var		   *var;
	Oid			relid;
	HeapTuple	statsTuple;
	double		selec;
	double		defselec;
	double		freq_null;

	switch (nulltesttype)
	{
		case IS_NULL:
			defselec = DEFAULT_UNK_SEL;
			break;
		case IS_NOT_NULL:
			defselec = DEFAULT_NOT_UNK_SEL;
			break;
		default:
			elog(ERROR, "unrecognized nulltesttype: %d",
				 (int) nulltesttype);
			return (Selectivity) 0;		/* keep compiler quiet */
	}

	/*
	 * Ignore any binary-compatible relabeling
	 */
	if (IsA(arg, RelabelType))
		arg = (Node *) ((RelabelType *) arg)->arg;

	if (IsA(arg, Var) &&
		(varRelid == 0 || varRelid == ((Var *) arg)->varno))
		var = (Var *) arg;
	else
	{
		/* punt if non-Var argument */
		return (Selectivity) defselec;
	}

	relid = getrelid(var->varno, root->rtable);
	if (relid == InvalidOid)
		return (Selectivity) defselec;

	/* get stats for the attribute, if available */
	statsTuple = SearchSysCache(STATRELATT,
								ObjectIdGetDatum(relid),
								Int16GetDatum(var->varattno),
								0, 0);
	if (HeapTupleIsValid(statsTuple))
	{
		Form_pg_statistic stats;

		stats = (Form_pg_statistic) GETSTRUCT(statsTuple);
		freq_null = stats->stanullfrac;

		switch (nulltesttype)
		{
			case IS_NULL:

				/*
				 * Use freq_null directly.
				 */
				selec = freq_null;
				break;
			case IS_NOT_NULL:

				/*
				 * Select not unknown (not null) values. Calculate from
				 * freq_null.
				 */
				selec = 1.0 - freq_null;
				break;
			default:
				elog(ERROR, "unrecognized nulltesttype: %d",
					 (int) nulltesttype);
				return (Selectivity) 0; /* keep compiler quiet */
		}

		ReleaseSysCache(statsTuple);
	}
	else
	{
		/*
		 * No VACUUM ANALYZE stats available, so make a guess
		 */
		selec = defselec;
	}

	/* result should be in range, but make sure... */
	CLAMP_PROBABILITY(selec);

	return (Selectivity) selec;
}

/*
 *		eqjoinsel		- Join selectivity of "="
 */
Datum
eqjoinsel(PG_FUNCTION_ARGS)
{
	Query	   *root = (Query *) PG_GETARG_POINTER(0);
	Oid			operator = PG_GETARG_OID(1);
	List	   *args = (List *) PG_GETARG_POINTER(2);
	JoinType	jointype = (JoinType) PG_GETARG_INT16(3);
	Var		   *var1;
	Var		   *var2;
	double		selec;

	get_join_vars(args, &var1, &var2);

	if (var1 == NULL && var2 == NULL)
		selec = DEFAULT_EQ_SEL;
	else
	{
		HeapTuple	statsTuple1 = NULL;
		HeapTuple	statsTuple2 = NULL;
		Form_pg_statistic stats1 = NULL;
		Form_pg_statistic stats2 = NULL;
		double		nd1 = DEFAULT_NUM_DISTINCT;
		double		nd2 = DEFAULT_NUM_DISTINCT;
		bool		have_mcvs1 = false;
		Datum	   *values1 = NULL;
		int			nvalues1 = 0;
		float4	   *numbers1 = NULL;
		int			nnumbers1 = 0;
		bool		have_mcvs2 = false;
		Datum	   *values2 = NULL;
		int			nvalues2 = 0;