selfuncs.c

PostgreSQL 8.1.4的源码 适用于Linux下的开源数据库系统

										   constval)))
				mcv_selec += numbers[i];
			sumcommon += numbers[i];
		}
		free_attstatsslot(vardata->atttype, values, nvalues,
						  numbers, nnumbers);
	}

	/*
	 * If there is a histogram, determine which bin the constant falls in, and
	 * compute the resulting contribution to selectivity.
	 *
	 * Someday, VACUUM might store more than one histogram per rel/att,
	 * corresponding to more than one possible sort ordering defined for the
	 * column type.  However, to make that work we will need to figure out
	 * which staop to search for --- it's not necessarily the one we have at
	 * hand!  (For example, we might have a '<=' operator rather than the '<'
	 * operator that will appear in staop.)  For now, assume that whatever
	 * appears in pg_statistic is sorted the same way our operator sorts, or
	 * the reverse way if isgt is TRUE.
	 */
	hist_selec = 0.0;

	if (get_attstatsslot(vardata->statsTuple,
						 vardata->atttype, vardata->atttypmod,
						 STATISTIC_KIND_HISTOGRAM, InvalidOid,
						 &values, &nvalues,
						 NULL, NULL))
	{
		if (nvalues > 1)
		{
			double		histfrac;
			bool		ltcmp;

			ltcmp = DatumGetBool(FunctionCall2(&opproc,
											   values[0],
											   constval));
			if (isgt)
				ltcmp = !ltcmp;
			if (!ltcmp)
			{
				/* Constant is below lower histogram boundary. */
				histfrac = 0.0;
			}
			else
			{
				/*
				 * Scan to find proper location.  This could be made faster by
				 * using a binary-search method, but it's probably not worth
				 * the trouble for typical histogram sizes.
				 */
				for (i = 1; i < nvalues; i++)
				{
					ltcmp = DatumGetBool(FunctionCall2(&opproc,
													   values[i],
													   constval));
					if (isgt)
						ltcmp = !ltcmp;
					if (!ltcmp)
						break;
				}
				if (i >= nvalues)
				{
					/* Constant is above upper histogram boundary. */
					histfrac = 1.0;
				}
				else
				{
					double		val,
								high,
								low;
					double		binfrac;

					/*
					 * We have values[i-1] < constant < values[i].
					 *
					 * Convert the constant and the two nearest bin boundary
					 * values to a uniform comparison scale, and do a linear
					 * interpolation within this bin.
					 */
					if (convert_to_scalar(constval, consttype, &val,
										  values[i - 1], values[i],
										  vardata->vartype,
										  &low, &high))
					{
						if (high <= low)
						{
							/* cope if bin boundaries appear identical */
							binfrac = 0.5;
						}
						else if (val <= low)
							binfrac = 0.0;
						else if (val >= high)
							binfrac = 1.0;
						else
						{
							binfrac = (val - low) / (high - low);

							/*
							 * Watch out for the possibility that we got a NaN
							 * or Infinity from the division.  This can happen
							 * despite the previous checks, if for example
							 * "low" is -Infinity.
							 */
							if (isnan(binfrac) ||
								binfrac < 0.0 || binfrac > 1.0)
								binfrac = 0.5;
						}
					}
					else
					{
						/*
						 * Ideally we'd produce an error here, on the grounds
						 * that the given operator shouldn't have scalarXXsel
						 * registered as its selectivity func unless we can
						 * deal with its operand types.  But currently, all
						 * manner of stuff is invoking scalarXXsel, so give a
						 * default estimate until that can be fixed.
						 */
						binfrac = 0.5;
					}

					/*
					 * Now, compute the overall selectivity across the values
					 * represented by the histogram.  We have i-1 full bins
					 * and binfrac partial bin below the constant.
					 */
					histfrac = (double) (i - 1) + binfrac;
					histfrac /= (double) (nvalues - 1);
				}
			}

			/*
			 * Now histfrac = fraction of histogram entries below the
			 * constant.
			 *
			 * Account for "<" vs ">"
			 */
			hist_selec = isgt ? (1.0 - histfrac) : histfrac;

			/*
			 * The histogram boundaries are only approximate to begin with,
			 * and may well be out of date anyway.	Therefore, don't believe
			 * extremely small or large selectivity estimates.
			 */
			if (hist_selec < 0.0001)
				hist_selec = 0.0001;
			else if (hist_selec > 0.9999)
				hist_selec = 0.9999;
		}

		free_attstatsslot(vardata->atttype, values, nvalues, NULL, 0);
	}

	/*
	 * Now merge the results from the MCV and histogram calculations,
	 * realizing that the histogram covers only the non-null values that are
	 * not listed in MCV.
	 */
	selec = 1.0 - stats->stanullfrac - sumcommon;

	if (hist_selec > 0.0)
		selec *= hist_selec;
	else
	{
		/*
		 * If no histogram but there are values not accounted for by MCV,
		 * arbitrarily assume half of them will match.
		 */
		selec *= 0.5;
	}

	selec += mcv_selec;

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

	return selec;
}

/*
 *		scalarltsel		- Selectivity of "<" (also "<=") for scalars.
 */
Datum
scalarltsel(PG_FUNCTION_ARGS)
{
	PlannerInfo *root = (PlannerInfo *) PG_GETARG_POINTER(0);
	Oid			operator = PG_GETARG_OID(1);
	List	   *args = (List *) PG_GETARG_POINTER(2);
	int			varRelid = PG_GETARG_INT32(3);
	VariableStatData vardata;
	Node	   *other;
	bool		varonleft;
	Datum		constval;
	Oid			consttype;
	bool		isgt;
	double		selec;

	/*
	 * If expression is not variable op something or something op variable,
	 * then punt and return a default estimate.
	 */
	if (!get_restriction_variable(root, args, varRelid,
								  &vardata, &other, &varonleft))
		PG_RETURN_FLOAT8(DEFAULT_INEQ_SEL);

	/*
	 * Can't do anything useful if the something is not a constant, either.
	 */
	if (!IsA(other, Const))
	{
		ReleaseVariableStats(vardata);
		PG_RETURN_FLOAT8(DEFAULT_INEQ_SEL);
	}

	/*
	 * If the constant is NULL, assume operator is strict and return zero, ie,
	 * operator will never return TRUE.
	 */
	if (((Const *) other)->constisnull)
	{
		ReleaseVariableStats(vardata);
		PG_RETURN_FLOAT8(0.0);
	}
	constval = ((Const *) other)->constvalue;
	consttype = ((Const *) other)->consttype;

	/*
	 * Force the var to be on the left to simplify logic in scalarineqsel.
	 */
	if (varonleft)
	{
		/* we have var < other */
		isgt = false;
	}
	else
	{
		/* we have other < var, commute to make var > other */
		operator = get_commutator(operator);
		if (!operator)
		{
			/* Use default selectivity (should we raise an error instead?) */
			ReleaseVariableStats(vardata);
			PG_RETURN_FLOAT8(DEFAULT_INEQ_SEL);
		}
		isgt = true;
	}

	selec = scalarineqsel(root, operator, isgt, &vardata, constval, consttype);

	ReleaseVariableStats(vardata);

	PG_RETURN_FLOAT8((float8) selec);
}

/*
 *		scalargtsel		- Selectivity of ">" (also ">=") for integers.
 */
Datum
scalargtsel(PG_FUNCTION_ARGS)
{
	PlannerInfo *root = (PlannerInfo *) PG_GETARG_POINTER(0);
	Oid			operator = PG_GETARG_OID(1);
	List	   *args = (List *) PG_GETARG_POINTER(2);
	int			varRelid = PG_GETARG_INT32(3);
	VariableStatData vardata;
	Node	   *other;
	bool		varonleft;
	Datum		constval;
	Oid			consttype;
	bool		isgt;
	double		selec;

	/*
	 * If expression is not variable op something or something op variable,
	 * then punt and return a default estimate.
	 */
	if (!get_restriction_variable(root, args, varRelid,
								  &vardata, &other, &varonleft))
		PG_RETURN_FLOAT8(DEFAULT_INEQ_SEL);

	/*
	 * Can't do anything useful if the something is not a constant, either.
	 */
	if (!IsA(other, Const))
	{
		ReleaseVariableStats(vardata);
		PG_RETURN_FLOAT8(DEFAULT_INEQ_SEL);
	}

	/*
	 * If the constant is NULL, assume operator is strict and return zero, ie,
	 * operator will never return TRUE.
	 */
	if (((Const *) other)->constisnull)
	{
		ReleaseVariableStats(vardata);
		PG_RETURN_FLOAT8(0.0);
	}
	constval = ((Const *) other)->constvalue;
	consttype = ((Const *) other)->consttype;

	/*
	 * Force the var to be on the left to simplify logic in scalarineqsel.
	 */
	if (varonleft)
	{
		/* we have var > other */
		isgt = true;
	}
	else
	{
		/* we have other > var, commute to make var < other */
		operator = get_commutator(operator);
		if (!operator)
		{
			/* Use default selectivity (should we raise an error instead?) */
			ReleaseVariableStats(vardata);
			PG_RETURN_FLOAT8(DEFAULT_INEQ_SEL);
		}
		isgt = false;
	}

	selec = scalarineqsel(root, operator, isgt, &vardata, constval, consttype);

	ReleaseVariableStats(vardata);

	PG_RETURN_FLOAT8((float8) selec);
}

/*
 * patternsel			- Generic code for pattern-match selectivity.
 */
static double
patternsel(PG_FUNCTION_ARGS, Pattern_Type ptype)
{
	PlannerInfo *root = (PlannerInfo *) PG_GETARG_POINTER(0);

#ifdef NOT_USED
	Oid			operator = PG_GETARG_OID(1);
#endif
	List	   *args = (List *) PG_GETARG_POINTER(2);
	int			varRelid = PG_GETARG_INT32(3);
	VariableStatData vardata;
	Node	   *variable;
	Node	   *other;
	bool		varonleft;
	Datum		constval;
	Oid			consttype;
	Oid			vartype;
	Oid			opclass;
	Pattern_Prefix_Status pstatus;
	Const	   *patt = NULL;
	Const	   *prefix = NULL;
	Const	   *rest = NULL;
	double		result;

	/*
	 * If expression is not variable op constant, then punt and return a
	 * default estimate.
	 */
	if (!get_restriction_variable(root, args, varRelid,
								  &vardata, &other, &varonleft))
		return DEFAULT_MATCH_SEL;
	if (!varonleft || !IsA(other, Const))
	{
		ReleaseVariableStats(vardata);
		return DEFAULT_MATCH_SEL;
	}
	variable = (Node *) linitial(args);

	/*
	 * If the constant is NULL, assume operator is strict and return zero, ie,
	 * operator will never return TRUE.
	 */
	if (((Const *) other)->constisnull)
	{
		ReleaseVariableStats(vardata);
		return 0.0;
	}
	constval = ((Const *) other)->constvalue;
	consttype = ((Const *) other)->consttype;

	/*
	 * The right-hand const is type text or bytea for all supported operators.
	 * We do not expect to see binary-compatible types here, since
	 * const-folding should have relabeled the const to exactly match the
	 * operator's declared type.
	 */
	if (consttype != TEXTOID && consttype != BYTEAOID)
	{
		ReleaseVariableStats(vardata);
		return DEFAULT_MATCH_SEL;
	}

	/*
	 * Similarly, the exposed type of the left-hand side should be one of
	 * those we know.  (Do not look at vardata.atttype, which might be
	 * something binary-compatible but different.)	We can use it to choose
	 * the index opclass from which we must draw the comparison operators.
	 *
	 * NOTE: It would be more correct to use the PATTERN opclasses than the
	 * simple ones, but at the moment ANALYZE will not generate statistics for
	 * the PATTERN operators.  But our results are so approximate anyway that
	 * it probably hardly matters.
	 */
	vartype = vardata.vartype;

	switch (vartype)
	{
		case TEXTOID:
			opclass = TEXT_BTREE_OPS_OID;
			break;
		case VARCHAROID:
			opclass = VARCHAR_BTREE_OPS_OID;
			break;
		case BPCHAROID:
			opclass = BPCHAR_BTREE_OPS_OID;
			break;
		case NAMEOID:
			opclass = NAME_BTREE_OPS_OID;
			break;
		case BYTEAOID:
			opclass = BYTEA_BTREE_OPS_OID;
			break;
		default:
			ReleaseVariableStats(vardata);
			return DEFAULT_MATCH_SEL;
	}

	/* divide pattern into fixed prefix and remainder */
	patt = (Const *) other;
	pstatus = pattern_fixed_prefix(patt, ptype, &prefix, &rest);

	/*
	 * If necessary, coerce the prefix constant to the right type. (The "rest"
	 * constant need not be changed.)
	 */
	if (prefix && prefix->consttype != vartype)
	{
		char	   *prefixstr;

		switch (prefix->consttype)
		{
			case TEXTOID:
				prefixstr = DatumGetCString(DirectFunctionCall1(textout,
														prefix->constvalue));
				break;
			case BYTEAOID:
				prefixstr = DatumGetCString(DirectFunctionCall1(byteaout,
														prefix->constvalue));
				break;
			default:
				elog(ERROR, "unrecognized consttype: %u",
					 prefix->consttype);
				ReleaseVariableStats(vardata);
				return DEFAULT_MATCH_SEL;
		}
		prefix = string_to_const(prefixstr, vartype);
		pfree(prefixstr);
	}

	if (pstatus == Pattern_Prefix_Exact)
	{
		/*
		 * Pattern specifies an exact match, so pretend operator is '='