selfuncs.c File Reference

#include "postgres.h"
#include <ctype.h>
#include <math.h>
#include "access/brin.h"
#include "access/brin_page.h"
#include "access/gin.h"
#include "access/table.h"
#include "access/tableam.h"
#include "access/visibilitymap.h"
#include "catalog/pg_am.h"
#include "catalog/pg_collation.h"
#include "catalog/pg_operator.h"
#include "catalog/pg_statistic.h"
#include "catalog/pg_statistic_ext.h"
#include "executor/nodeAgg.h"
#include "miscadmin.h"
#include "nodes/makefuncs.h"
#include "nodes/nodeFuncs.h"
#include "optimizer/clauses.h"
#include "optimizer/cost.h"
#include "optimizer/optimizer.h"
#include "optimizer/pathnode.h"
#include "optimizer/paths.h"
#include "optimizer/plancat.h"
#include "parser/parse_clause.h"
#include "parser/parse_relation.h"
#include "parser/parsetree.h"
#include "statistics/statistics.h"
#include "storage/bufmgr.h"
#include "utils/acl.h"
#include "utils/array.h"
#include "utils/builtins.h"
#include "utils/date.h"
#include "utils/datum.h"
#include "utils/fmgroids.h"
#include "utils/index_selfuncs.h"
#include "utils/lsyscache.h"
#include "utils/memutils.h"
#include "utils/pg_locale.h"
#include "utils/rel.h"
#include "utils/selfuncs.h"
#include "utils/snapmgr.h"
#include "utils/spccache.h"
#include "utils/syscache.h"
#include "utils/timestamp.h"
#include "utils/typcache.h"

Include dependency graph for selfuncs.c:

Go to the source code of this file.

Data Structures
struct	GroupVarInfo
struct	GinQualCounts

Macros
#define	DEFAULT_PAGE_CPU_MULTIPLIER   50.0
#define	VISITED_PAGES_LIMIT   100

Functions
static double	eqsel_internal (PG_FUNCTION_ARGS, bool negate)
static double	eqjoinsel_inner (Oid opfuncoid, Oid collation, VariableStatData *vardata1, VariableStatData *vardata2, double nd1, double nd2, bool isdefault1, bool isdefault2, AttStatsSlot *sslot1, AttStatsSlot *sslot2, Form_pg_statistic stats1, Form_pg_statistic stats2, bool have_mcvs1, bool have_mcvs2)
static double	eqjoinsel_semi (Oid opfuncoid, Oid collation, VariableStatData *vardata1, VariableStatData *vardata2, double nd1, double nd2, bool isdefault1, bool isdefault2, AttStatsSlot *sslot1, AttStatsSlot *sslot2, Form_pg_statistic stats1, Form_pg_statistic stats2, bool have_mcvs1, bool have_mcvs2, RelOptInfo *inner_rel)
static bool	estimate_multivariate_ndistinct (PlannerInfo *root, RelOptInfo *rel, List **varinfos, double *ndistinct)
static bool	convert_to_scalar (Datum value, Oid valuetypid, Oid collid, double *scaledvalue, Datum lobound, Datum hibound, Oid boundstypid, double *scaledlobound, double *scaledhibound)
static double	convert_numeric_to_scalar (Datum value, Oid typid, bool *failure)
static void	convert_string_to_scalar (char *value, double *scaledvalue, char *lobound, double *scaledlobound, char *hibound, double *scaledhibound)
static void	convert_bytea_to_scalar (Datum value, double *scaledvalue, Datum lobound, double *scaledlobound, Datum hibound, double *scaledhibound)
static double	convert_one_string_to_scalar (char *value, int rangelo, int rangehi)
static double	convert_one_bytea_to_scalar (unsigned char *value, int valuelen, int rangelo, int rangehi)
static char *	convert_string_datum (Datum value, Oid typid, Oid collid, bool *failure)
static double	convert_timevalue_to_scalar (Datum value, Oid typid, bool *failure)
static void	examine_simple_variable (PlannerInfo *root, Var *var, VariableStatData *vardata)
static bool	get_variable_range (PlannerInfo *root, VariableStatData *vardata, Oid sortop, Oid collation, Datum *min, Datum *max)
static void	get_stats_slot_range (AttStatsSlot *sslot, Oid opfuncoid, FmgrInfo *opproc, Oid collation, int16 typLen, bool typByVal, Datum *min, Datum *max, bool *p_have_data)
static bool	get_actual_variable_range (PlannerInfo *root, VariableStatData *vardata, Oid sortop, Oid collation, Datum *min, Datum *max)
static bool	get_actual_variable_endpoint (Relation heapRel, Relation indexRel, ScanDirection indexscandir, ScanKey scankeys, int16 typLen, bool typByVal, TupleTableSlot *tableslot, MemoryContext outercontext, Datum *endpointDatum)
static RelOptInfo *	find_join_input_rel (PlannerInfo *root, Relids relids)
Datum	eqsel (PG_FUNCTION_ARGS)
double	var_eq_const (VariableStatData *vardata, Oid oproid, Oid collation, Datum constval, bool constisnull, bool varonleft, bool negate)
double	var_eq_non_const (VariableStatData *vardata, Oid oproid, Oid collation, Node *other, bool varonleft, bool negate)
Datum	neqsel (PG_FUNCTION_ARGS)
static double	scalarineqsel (PlannerInfo *root, Oid operator, bool isgt, bool iseq, Oid collation, VariableStatData *vardata, Datum constval, Oid consttype)
double	mcv_selectivity (VariableStatData *vardata, FmgrInfo *opproc, Oid collation, Datum constval, bool varonleft, double *sumcommonp)
double	histogram_selectivity (VariableStatData *vardata, FmgrInfo *opproc, Oid collation, Datum constval, bool varonleft, int min_hist_size, int n_skip, int *hist_size)
double	generic_restriction_selectivity (PlannerInfo *root, Oid oproid, Oid collation, List *args, int varRelid, double default_selectivity)
double	ineq_histogram_selectivity (PlannerInfo *root, VariableStatData *vardata, Oid opoid, FmgrInfo *opproc, bool isgt, bool iseq, Oid collation, Datum constval, Oid consttype)
static Datum	scalarineqsel_wrapper (PG_FUNCTION_ARGS, bool isgt, bool iseq)
Datum	scalarltsel (PG_FUNCTION_ARGS)
Datum	scalarlesel (PG_FUNCTION_ARGS)
Datum	scalargtsel (PG_FUNCTION_ARGS)
Datum	scalargesel (PG_FUNCTION_ARGS)
Selectivity	boolvarsel (PlannerInfo *root, Node *arg, int varRelid)
Selectivity	booltestsel (PlannerInfo *root, BoolTestType booltesttype, Node *arg, int varRelid, JoinType jointype, SpecialJoinInfo *sjinfo)
Selectivity	nulltestsel (PlannerInfo *root, NullTestType nulltesttype, Node *arg, int varRelid, JoinType jointype, SpecialJoinInfo *sjinfo)
static Node *	strip_array_coercion (Node *node)
Selectivity	scalararraysel (PlannerInfo *root, ScalarArrayOpExpr *clause, bool is_join_clause, int varRelid, JoinType jointype, SpecialJoinInfo *sjinfo)
double	estimate_array_length (PlannerInfo *root, Node *arrayexpr)
Selectivity	rowcomparesel (PlannerInfo *root, RowCompareExpr *clause, int varRelid, JoinType jointype, SpecialJoinInfo *sjinfo)
Datum	eqjoinsel (PG_FUNCTION_ARGS)
Datum	neqjoinsel (PG_FUNCTION_ARGS)
Datum	scalarltjoinsel (PG_FUNCTION_ARGS)
Datum	scalarlejoinsel (PG_FUNCTION_ARGS)
Datum	scalargtjoinsel (PG_FUNCTION_ARGS)
Datum	scalargejoinsel (PG_FUNCTION_ARGS)
void	mergejoinscansel (PlannerInfo *root, Node *clause, Oid opfamily, int strategy, bool nulls_first, Selectivity *leftstart, Selectivity *leftend, Selectivity *rightstart, Selectivity *rightend)
Datum	matchingsel (PG_FUNCTION_ARGS)
Datum	matchingjoinsel (PG_FUNCTION_ARGS)
static List *	add_unique_group_var (PlannerInfo *root, List *varinfos, Node *var, VariableStatData *vardata)
double	estimate_num_groups (PlannerInfo *root, List *groupExprs, double input_rows, List **pgset, EstimationInfo *estinfo)
void	estimate_hash_bucket_stats (PlannerInfo *root, Node *hashkey, double nbuckets, Selectivity *mcv_freq, Selectivity *bucketsize_frac)
double	estimate_hashagg_tablesize (PlannerInfo *root, Path *path, const AggClauseCosts *agg_costs, double dNumGroups)
bool	get_restriction_variable (PlannerInfo *root, List *args, int varRelid, VariableStatData *vardata, Node **other, bool *varonleft)
void	get_join_variables (PlannerInfo *root, List *args, SpecialJoinInfo *sjinfo, VariableStatData *vardata1, VariableStatData *vardata2, bool *join_is_reversed)
static void	ReleaseDummy (HeapTuple tuple)
void	examine_variable (PlannerInfo *root, Node *node, int varRelid, VariableStatData *vardata)
bool	statistic_proc_security_check (VariableStatData *vardata, Oid func_oid)
double	get_variable_numdistinct (VariableStatData *vardata, bool *isdefault)
List *	get_quals_from_indexclauses (List *indexclauses)
Cost	index_other_operands_eval_cost (PlannerInfo *root, List *indexquals)
void	genericcostestimate (PlannerInfo *root, IndexPath *path, double loop_count, GenericCosts *costs)
List *	add_predicate_to_index_quals (IndexOptInfo *index, List *indexQuals)
void	btcostestimate (PlannerInfo *root, IndexPath *path, double loop_count, Cost *indexStartupCost, Cost *indexTotalCost, Selectivity *indexSelectivity, double *indexCorrelation, double *indexPages)
void	hashcostestimate (PlannerInfo *root, IndexPath *path, double loop_count, Cost *indexStartupCost, Cost *indexTotalCost, Selectivity *indexSelectivity, double *indexCorrelation, double *indexPages)
void	gistcostestimate (PlannerInfo *root, IndexPath *path, double loop_count, Cost *indexStartupCost, Cost *indexTotalCost, Selectivity *indexSelectivity, double *indexCorrelation, double *indexPages)
void	spgcostestimate (PlannerInfo *root, IndexPath *path, double loop_count, Cost *indexStartupCost, Cost *indexTotalCost, Selectivity *indexSelectivity, double *indexCorrelation, double *indexPages)
static bool	gincost_pattern (IndexOptInfo *index, int indexcol, Oid clause_op, Datum query, GinQualCounts *counts)
static bool	gincost_opexpr (PlannerInfo *root, IndexOptInfo *index, int indexcol, OpExpr *clause, GinQualCounts *counts)
static bool	gincost_scalararrayopexpr (PlannerInfo *root, IndexOptInfo *index, int indexcol, ScalarArrayOpExpr *clause, double numIndexEntries, GinQualCounts *counts)
void	gincostestimate (PlannerInfo *root, IndexPath *path, double loop_count, Cost *indexStartupCost, Cost *indexTotalCost, Selectivity *indexSelectivity, double *indexCorrelation, double *indexPages)
void	brincostestimate (PlannerInfo *root, IndexPath *path, double loop_count, Cost *indexStartupCost, Cost *indexTotalCost, Selectivity *indexSelectivity, double *indexCorrelation, double *indexPages)

Variables
get_relation_stats_hook_type	get_relation_stats_hook = NULL
get_index_stats_hook_type	get_index_stats_hook = NULL

Macro Definition Documentation

DEFAULT_PAGE_CPU_MULTIPLIER

#define DEFAULT_PAGE_CPU_MULTIPLIER   50.0

Definition at line 144 of file selfuncs.c.

VISITED_PAGES_LIMIT

#define VISITED_PAGES_LIMIT   100

Function Documentation

add_predicate_to_index_quals()

List* add_predicate_to_index_quals	(	IndexOptInfo *	index,
		List *	indexQuals
	)

Definition at line 6775 of file selfuncs.c.

{
    List       *predExtraQuals = NIL;
    ListCell   *lc;
 
    if (index->indpred == NIL)
        return indexQuals;
 
    foreach(lc, index->indpred)
    {
        Node       *predQual = (Node *) lfirst(lc);
        List       *oneQual = list_make1(predQual);
 
        if (!predicate_implied_by(oneQual, indexQuals, false))
            predExtraQuals = list_concat(predExtraQuals, oneQual);
    }
    return list_concat(predExtraQuals, indexQuals);
}

References lfirst, list_concat(), list_make1, NIL, and predicate_implied_by().

Referenced by btcostestimate(), genericcostestimate(), and gincostestimate().

add_unique_group_var()

static List* add_unique_group_var ( PlannerInfo *  root, List *  varinfos, Node *  var, VariableStatData *  vardata  )

static

Definition at line 3296 of file selfuncs.c.

{
    GroupVarInfo *varinfo;
    double      ndistinct;
    bool        isdefault;
    ListCell   *lc;
 
    ndistinct = get_variable_numdistinct(vardata, &isdefault);
 
    foreach(lc, varinfos)
    {
        varinfo = (GroupVarInfo *) lfirst(lc);
 
        /* Drop exact duplicates */
        if (equal(var, varinfo->var))
            return varinfos;
 
        /*
         * Drop known-equal vars, but only if they belong to different
         * relations (see comments for estimate_num_groups).  We aren't too
         * fussy about the semantics of "equal" here.
         */
        if (vardata->rel != varinfo->rel &&
            exprs_known_equal(root, var, varinfo->var, InvalidOid))
        {
            if (varinfo->ndistinct <= ndistinct)
            {
                /* Keep older item, forget new one */
                return varinfos;
            }
            else
            {
                /* Delete the older item */
                varinfos = foreach_delete_current(varinfos, lc);
            }
        }
    }
 
    varinfo = (GroupVarInfo *) palloc(sizeof(GroupVarInfo));
 
    varinfo->var = var;
    varinfo->rel = vardata->rel;
    varinfo->ndistinct = ndistinct;
    varinfo->isdefault = isdefault;
    varinfos = lappend(varinfos, varinfo);
    return varinfos;
}

References equal(), exprs_known_equal(), foreach_delete_current, get_variable_numdistinct(), InvalidOid, GroupVarInfo::isdefault, lappend(), lfirst, GroupVarInfo::ndistinct, palloc(), GroupVarInfo::rel, VariableStatData::rel, root, and GroupVarInfo::var.

Referenced by estimate_num_groups().

booltestsel()

Selectivity booltestsel	(	PlannerInfo *	root,
		BoolTestType	booltesttype,
		Node *	arg,
		int	varRelid,
		JoinType	jointype,
		SpecialJoinInfo *	sjinfo
	)

Definition at line 1540 of file selfuncs.c.

{
    VariableStatData vardata;
    double      selec;
 
    examine_variable(root, arg, varRelid, &vardata);
 
    if (HeapTupleIsValid(vardata.statsTuple))
    {
        Form_pg_statistic stats;
        double      freq_null;
        AttStatsSlot sslot;
 
        stats = (Form_pg_statistic) GETSTRUCT(vardata.statsTuple);
        freq_null = stats->stanullfrac;
 
        if (get_attstatsslot(&sslot, vardata.statsTuple,
                             STATISTIC_KIND_MCV, InvalidOid,
                             ATTSTATSSLOT_VALUES | ATTSTATSSLOT_NUMBERS)
            && sslot.nnumbers > 0)
        {
            double      freq_true;
            double      freq_false;
 
            /*
             * Get first MCV frequency and derive frequency for true.
             */
            if (DatumGetBool(sslot.values[0]))
                freq_true = sslot.numbers[0];
            else
                freq_true = 1.0 - sslot.numbers[0] - freq_null;
 
            /*
             * Next derive frequency 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(&sslot);
        }
        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 a 50-50 split of TRUE and FALSE.
             */
            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:
                case IS_FALSE:
                    /* Assume we select half of the non-NULL values */
                    selec = (1.0 - freq_null) / 2.0;
                    break;
                case IS_NOT_TRUE:
                case IS_NOT_FALSE:
                    /* Assume we select NULLs plus half of the non-NULLs */
                    /* equiv. to freq_null + (1.0 - freq_null) / 2.0 */
                    selec = (freq_null + 1.0) / 2.0;
                    break;
                default:
                    elog(ERROR, "unrecognized booltesttype: %d",
                         (int) booltesttype);
                    selec = 0.0;    /* Keep compiler quiet */
                    break;
            }
        }
    }
    else
    {
        /*
         * If we can't get variable statistics for the argument, 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, sjinfo);
                break;
            case IS_FALSE:
            case IS_NOT_TRUE:
                selec = 1.0 - (double) clause_selectivity(root, arg,
                                                          varRelid,
                                                          jointype, sjinfo);
                break;
            default:
                elog(ERROR, "unrecognized booltesttype: %d",
                     (int) booltesttype);
                selec = 0.0;    /* Keep compiler quiet */
                break;
        }
    }
 
    ReleaseVariableStats(vardata);
 
    /* result should be in range, but make sure... */
    CLAMP_PROBABILITY(selec);
 
    return (Selectivity) selec;
}

References arg, ATTSTATSSLOT_NUMBERS, ATTSTATSSLOT_VALUES, CLAMP_PROBABILITY, clause_selectivity(), DatumGetBool(), DEFAULT_NOT_UNK_SEL, DEFAULT_UNK_SEL, elog, ERROR, examine_variable(), free_attstatsslot(), get_attstatsslot(), GETSTRUCT, HeapTupleIsValid, InvalidOid, IS_FALSE, IS_NOT_FALSE, IS_NOT_TRUE, IS_NOT_UNKNOWN, IS_TRUE, IS_UNKNOWN, AttStatsSlot::nnumbers, AttStatsSlot::numbers, ReleaseVariableStats, root, VariableStatData::statsTuple, and AttStatsSlot::values.

Referenced by clause_selectivity_ext().

boolvarsel()

Selectivity boolvarsel	(	PlannerInfo *	root,
		Node *	arg,
		int	varRelid
	)

Definition at line 1512 of file selfuncs.c.

{
    VariableStatData vardata;
    double      selec;
 
    examine_variable(root, arg, varRelid, &vardata);
    if (HeapTupleIsValid(vardata.statsTuple))
    {
        /*
         * A boolean variable V is equivalent to the clause V = 't', so we
         * compute the selectivity as if that is what we have.
         */
        selec = var_eq_const(&vardata, BooleanEqualOperator, InvalidOid,
                             BoolGetDatum(true), false, true, false);
    }
    else
    {
        /* Otherwise, the default estimate is 0.5 */
        selec = 0.5;
    }
    ReleaseVariableStats(vardata);
    return selec;
}

References arg, BoolGetDatum(), examine_variable(), HeapTupleIsValid, InvalidOid, ReleaseVariableStats, root, VariableStatData::statsTuple, and var_eq_const().

Referenced by clause_selectivity_ext().

brincostestimate()

void brincostestimate	(	PlannerInfo *	root,
		IndexPath *	path,
		double	loop_count,
		Cost *	indexStartupCost,
		Cost *	indexTotalCost,
		Selectivity *	indexSelectivity,
		double *	indexCorrelation,
		double *	indexPages
	)

Definition at line 7981 of file selfuncs.c.

{
    IndexOptInfo *index = path->indexinfo;
    List       *indexQuals = get_quals_from_indexclauses(path->indexclauses);
    double      numPages = index->pages;
    RelOptInfo *baserel = index->rel;
    RangeTblEntry *rte = planner_rt_fetch(baserel->relid, root);
    Cost        spc_seq_page_cost;
    Cost        spc_random_page_cost;
    double      qual_arg_cost;
    double      qualSelectivity;
    BrinStatsData statsData;
    double      indexRanges;
    double      minimalRanges;
    double      estimatedRanges;
    double      selec;
    Relation    indexRel;
    ListCell   *l;
    VariableStatData vardata;
 
    Assert(rte->rtekind == RTE_RELATION);
 
    /* fetch estimated page cost for the tablespace containing the index */
    get_tablespace_page_costs(index->reltablespace,
                              &spc_random_page_cost,
                              &spc_seq_page_cost);
 
    /*
     * Obtain some data from the index itself, if possible.  Otherwise invent
     * some plausible internal statistics based on the relation page count.
     */
    if (!index->hypothetical)
    {
        /*
         * A lock should have already been obtained on the index in plancat.c.
         */
        indexRel = index_open(index->indexoid, NoLock);
        brinGetStats(indexRel, &statsData);
        index_close(indexRel, NoLock);
 
        /* work out the actual number of ranges in the index */
        indexRanges = Max(ceil((double) baserel->pages /
                               statsData.pagesPerRange), 1.0);
    }
    else
    {
        /*
         * Assume default number of pages per range, and estimate the number
         * of ranges based on that.
         */
        indexRanges = Max(ceil((double) baserel->pages /
                               BRIN_DEFAULT_PAGES_PER_RANGE), 1.0);
 
        statsData.pagesPerRange = BRIN_DEFAULT_PAGES_PER_RANGE;
        statsData.revmapNumPages = (indexRanges / REVMAP_PAGE_MAXITEMS) + 1;
    }
 
    /*
     * Compute index correlation
     *
     * Because we can use all index quals equally when scanning, we can use
     * the largest correlation (in absolute value) among columns used by the
     * query.  Start at zero, the worst possible case.  If we cannot find any
     * correlation statistics, we will keep it as 0.
     */
    *indexCorrelation = 0;
 
    foreach(l, path->indexclauses)
    {
        IndexClause *iclause = lfirst_node(IndexClause, l);
        AttrNumber  attnum = index->indexkeys[iclause->indexcol];
 
        /* attempt to lookup stats in relation for this index column */
        if (attnum != 0)
        {
            /* Simple variable -- look to stats for the underlying table */
            if (get_relation_stats_hook &&
                (*get_relation_stats_hook) (root, rte, attnum, &vardata))
            {
                /*
                 * The hook took control of acquiring a stats tuple.  If it
                 * did supply a tuple, it'd better have supplied a freefunc.
                 */
                if (HeapTupleIsValid(vardata.statsTuple) && !vardata.freefunc)
                    elog(ERROR,
                         "no function provided to release variable stats with");
            }
            else
            {
                vardata.statsTuple =
                    SearchSysCache3(STATRELATTINH,
                                    ObjectIdGetDatum(rte->relid),
                                    Int16GetDatum(attnum),
                                    BoolGetDatum(false));
                vardata.freefunc = ReleaseSysCache;
            }
        }
        else
        {
            /*
             * Looks like we've found an expression column in the index. Let's
             * see if there's any stats for it.
             */
 
            /* get the attnum from the 0-based index. */
            attnum = iclause->indexcol + 1;
 
            if (get_index_stats_hook &&
                (*get_index_stats_hook) (root, index->indexoid, attnum, &vardata))
            {
                /*
                 * The hook took control of acquiring a stats tuple.  If it
                 * did supply a tuple, it'd better have supplied a freefunc.
                 */
                if (HeapTupleIsValid(vardata.statsTuple) &&
                    !vardata.freefunc)
                    elog(ERROR, "no function provided to release variable stats with");
            }
            else
            {
                vardata.statsTuple = SearchSysCache3(STATRELATTINH,
                                                     ObjectIdGetDatum(index->indexoid),
                                                     Int16GetDatum(attnum),
                                                     BoolGetDatum(false));
                vardata.freefunc = ReleaseSysCache;
            }
        }
 
        if (HeapTupleIsValid(vardata.statsTuple))
        {
            AttStatsSlot sslot;
 
            if (get_attstatsslot(&sslot, vardata.statsTuple,
                                 STATISTIC_KIND_CORRELATION, InvalidOid,
                                 ATTSTATSSLOT_NUMBERS))
            {
                double      varCorrelation = 0.0;
 
                if (sslot.nnumbers > 0)
                    varCorrelation = fabs(sslot.numbers[0]);
 
                if (varCorrelation > *indexCorrelation)
                    *indexCorrelation = varCorrelation;
 
                free_attstatsslot(&sslot);
            }
        }
 
        ReleaseVariableStats(vardata);
    }
 
    qualSelectivity = clauselist_selectivity(root, indexQuals,
                                             baserel->relid,
                                             JOIN_INNER, NULL);
 
    /*
     * Now calculate the minimum possible ranges we could match with if all of
     * the rows were in the perfect order in the table's heap.
     */
    minimalRanges = ceil(indexRanges * qualSelectivity);
 
    /*
     * Now estimate the number of ranges that we'll touch by using the
     * indexCorrelation from the stats. Careful not to divide by zero (note
     * we're using the absolute value of the correlation).
     */
    if (*indexCorrelation < 1.0e-10)
        estimatedRanges = indexRanges;
    else
        estimatedRanges = Min(minimalRanges / *indexCorrelation, indexRanges);
 
    /* we expect to visit this portion of the table */
    selec = estimatedRanges / indexRanges;
 
    CLAMP_PROBABILITY(selec);
 
    *indexSelectivity = selec;
 
    /*
     * Compute the index qual costs, much as in genericcostestimate, to add to
     * the index costs.  We can disregard indexorderbys, since BRIN doesn't
     * support those.
     */
    qual_arg_cost = index_other_operands_eval_cost(root, indexQuals);
 
    /*
     * Compute the startup cost as the cost to read the whole revmap
     * sequentially, including the cost to execute the index quals.
     */
    *indexStartupCost =
        spc_seq_page_cost * statsData.revmapNumPages * loop_count;
    *indexStartupCost += qual_arg_cost;
 
    /*
     * To read a BRIN index there might be a bit of back and forth over
     * regular pages, as revmap might point to them out of sequential order;
     * calculate the total cost as reading the whole index in random order.
     */
    *indexTotalCost = *indexStartupCost +
        spc_random_page_cost * (numPages - statsData.revmapNumPages) * loop_count;
 
    /*
     * Charge a small amount per range tuple which we expect to match to. This
     * is meant to reflect the costs of manipulating the bitmap. The BRIN scan
     * will set a bit for each page in the range when we find a matching
     * range, so we must multiply the charge by the number of pages in the
     * range.
     */
    *indexTotalCost += 0.1 * cpu_operator_cost * estimatedRanges *
        statsData.pagesPerRange;
 
    *indexPages = index->pages;
}

References Assert, attnum, ATTSTATSSLOT_NUMBERS, BoolGetDatum(), BRIN_DEFAULT_PAGES_PER_RANGE, brinGetStats(), CLAMP_PROBABILITY, clauselist_selectivity(), cpu_operator_cost, elog, ERROR, free_attstatsslot(), VariableStatData::freefunc, get_attstatsslot(), get_index_stats_hook, get_quals_from_indexclauses(), get_relation_stats_hook, get_tablespace_page_costs(), HeapTupleIsValid, index_close(), index_open(), index_other_operands_eval_cost(), IndexPath::indexclauses, IndexClause::indexcol, IndexPath::indexinfo, Int16GetDatum(), InvalidOid, JOIN_INNER, lfirst_node, Max, Min, AttStatsSlot::nnumbers, NoLock, AttStatsSlot::numbers, ObjectIdGetDatum(), RelOptInfo::pages, BrinStatsData::pagesPerRange, planner_rt_fetch, ReleaseSysCache(), ReleaseVariableStats, RangeTblEntry::relid, RelOptInfo::relid, REVMAP_PAGE_MAXITEMS, BrinStatsData::revmapNumPages, root, RTE_RELATION, RangeTblEntry::rtekind, SearchSysCache3(), and VariableStatData::statsTuple.

Referenced by brinhandler().

btcostestimate()

void btcostestimate	(	PlannerInfo *	root,
		IndexPath *	path,
		double	loop_count,
		Cost *	indexStartupCost,
		Cost *	indexTotalCost,
		Selectivity *	indexSelectivity,
		double *	indexCorrelation,
		double *	indexPages
	)

Definition at line 6796 of file selfuncs.c.

{
    IndexOptInfo *index = path->indexinfo;
    GenericCosts costs = {0};
    Oid         relid;
    AttrNumber  colnum;
    VariableStatData vardata = {0};
    double      numIndexTuples;
    Cost        descentCost;
    List       *indexBoundQuals;
    int         indexcol;
    bool        eqQualHere;
    bool        found_saop;
    bool        found_is_null_op;
    double      num_sa_scans;
    ListCell   *lc;
 
    /*
     * For a btree scan, only leading '=' quals plus inequality quals for the
     * immediately next attribute contribute to index selectivity (these are
     * the "boundary quals" that determine the starting and stopping points of
     * the index scan).  Additional quals can suppress visits to the heap, so
     * it's OK to count them in indexSelectivity, but they should not count
     * for estimating numIndexTuples.  So we must examine the given indexquals
     * to find out which ones count as boundary quals.  We rely on the
     * knowledge that they are given in index column order.
     *
     * For a RowCompareExpr, we consider only the first column, just as
     * rowcomparesel() does.
     *
     * If there's a ScalarArrayOpExpr in the quals, we'll actually perform up
     * to N index descents (not just one), but the ScalarArrayOpExpr's
     * operator can be considered to act the same as it normally does.
     */
    indexBoundQuals = NIL;
    indexcol = 0;
    eqQualHere = false;
    found_saop = false;
    found_is_null_op = false;
    num_sa_scans = 1;
    foreach(lc, path->indexclauses)
    {
        IndexClause *iclause = lfirst_node(IndexClause, lc);
        ListCell   *lc2;
 
        if (indexcol != iclause->indexcol)
        {
            /* Beginning of a new column's quals */
            if (!eqQualHere)
                break;          /* done if no '=' qual for indexcol */
            eqQualHere = false;
            indexcol++;
            if (indexcol != iclause->indexcol)
                break;          /* no quals at all for indexcol */
        }
 
        /* Examine each indexqual associated with this index clause */
        foreach(lc2, iclause->indexquals)
        {
            RestrictInfo *rinfo = lfirst_node(RestrictInfo, lc2);
            Expr       *clause = rinfo->clause;
            Oid         clause_op = InvalidOid;
            int         op_strategy;
 
            if (IsA(clause, OpExpr))
            {
                OpExpr     *op = (OpExpr *) clause;
 
                clause_op = op->opno;
            }
            else if (IsA(clause, RowCompareExpr))
            {
                RowCompareExpr *rc = (RowCompareExpr *) clause;
 
                clause_op = linitial_oid(rc->opnos);
            }
            else if (IsA(clause, ScalarArrayOpExpr))
            {
                ScalarArrayOpExpr *saop = (ScalarArrayOpExpr *) clause;
                Node       *other_operand = (Node *) lsecond(saop->args);
                double      alength = estimate_array_length(root, other_operand);
 
                clause_op = saop->opno;
                found_saop = true;
                /* estimate SA descents by indexBoundQuals only */
                if (alength > 1)
                    num_sa_scans *= alength;
            }
            else if (IsA(clause, NullTest))
            {
                NullTest   *nt = (NullTest *) clause;
 
                if (nt->nulltesttype == IS_NULL)
                {
                    found_is_null_op = true;
                    /* IS NULL is like = for selectivity purposes */
                    eqQualHere = true;
                }
            }
            else
                elog(ERROR, "unsupported indexqual type: %d",
                     (int) nodeTag(clause));
 
            /* check for equality operator */
            if (OidIsValid(clause_op))
            {
                op_strategy = get_op_opfamily_strategy(clause_op,
                                                       index->opfamily[indexcol]);
                Assert(op_strategy != 0);   /* not a member of opfamily?? */
                if (op_strategy == BTEqualStrategyNumber)
                    eqQualHere = true;
            }
 
            indexBoundQuals = lappend(indexBoundQuals, rinfo);
        }
    }
 
    /*
     * If index is unique and we found an '=' clause for each column, we can
     * just assume numIndexTuples = 1 and skip the expensive
     * clauselist_selectivity calculations.  However, a ScalarArrayOp or
     * NullTest invalidates that theory, even though it sets eqQualHere.
     */
    if (index->unique &&
        indexcol == index->nkeycolumns - 1 &&
        eqQualHere &&
        !found_saop &&
        !found_is_null_op)
        numIndexTuples = 1.0;
    else
    {
        List       *selectivityQuals;
        Selectivity btreeSelectivity;
 
        /*
         * If the index is partial, AND the index predicate with the
         * index-bound quals to produce a more accurate idea of the number of
         * rows covered by the bound conditions.
         */
        selectivityQuals = add_predicate_to_index_quals(index, indexBoundQuals);
 
        btreeSelectivity = clauselist_selectivity(root, selectivityQuals,
                                                  index->rel->relid,
                                                  JOIN_INNER,
                                                  NULL);
        numIndexTuples = btreeSelectivity * index->rel->tuples;
 
        /*
         * btree automatically combines individual ScalarArrayOpExpr primitive
         * index scans whenever the tuples covered by the next set of array
         * keys are close to tuples covered by the current set.  That puts a
         * natural ceiling on the worst case number of descents -- there
         * cannot possibly be more than one descent per leaf page scanned.
         *
         * Clamp the number of descents to at most 1/3 the number of index
         * pages.  This avoids implausibly high estimates with low selectivity
         * paths, where scans usually require only one or two descents.  This
         * is most likely to help when there are several SAOP clauses, where
         * naively accepting the total number of distinct combinations of
         * array elements as the number of descents would frequently lead to
         * wild overestimates.
         *
         * We somewhat arbitrarily don't just make the cutoff the total number
         * of leaf pages (we make it 1/3 the total number of pages instead) to
         * give the btree code credit for its ability to continue on the leaf
         * level with low selectivity scans.
         */
        num_sa_scans = Min(num_sa_scans, ceil(index->pages * 0.3333333));
        num_sa_scans = Max(num_sa_scans, 1);
 
        /*
         * As in genericcostestimate(), we have to adjust for any
         * ScalarArrayOpExpr quals included in indexBoundQuals, and then round
         * to integer.
         *
         * It is tempting to make genericcostestimate behave as if SAOP
         * clauses work in almost the same way as scalar operators during
         * btree scans, making the top-level scan look like a continuous scan
         * (as opposed to num_sa_scans-many primitive index scans).  After
         * all, btree scans mostly work like that at runtime.  However, such a
         * scheme would badly bias genericcostestimate's simplistic approach
         * to calculating numIndexPages through prorating.
         *
         * Stick with the approach taken by non-native SAOP scans for now.
         * genericcostestimate will use the Mackert-Lohman formula to
         * compensate for repeat page fetches, even though that definitely
         * won't happen during btree scans (not for leaf pages, at least).
         * We're usually very pessimistic about the number of primitive index
         * scans that will be required, but it's not clear how to do better.
         */
        numIndexTuples = rint(numIndexTuples / num_sa_scans);
    }
 
    /*
     * Now do generic index cost estimation.
     */
    costs.numIndexTuples = numIndexTuples;
    costs.num_sa_scans = num_sa_scans;
 
    genericcostestimate(root, path, loop_count, &costs);
 
    /*
     * Add a CPU-cost component to represent the costs of initial btree
     * descent.  We don't charge any I/O cost for touching upper btree levels,
     * since they tend to stay in cache, but we still have to do about log2(N)
     * comparisons to descend a btree of N leaf tuples.  We charge one
     * cpu_operator_cost per comparison.
     *
     * If there are ScalarArrayOpExprs, charge this once per estimated SA
     * index descent.  The ones after the first one are not startup cost so
     * far as the overall plan goes, so just add them to "total" cost.
     */
    if (index->tuples > 1)      /* avoid computing log(0) */
    {
        descentCost = ceil(log(index->tuples) / log(2.0)) * cpu_operator_cost;
        costs.indexStartupCost += descentCost;
        costs.indexTotalCost += costs.num_sa_scans * descentCost;
    }
 
    /*
     * Even though we're not charging I/O cost for touching upper btree pages,
     * it's still reasonable to charge some CPU cost per page descended
     * through.  Moreover, if we had no such charge at all, bloated indexes
     * would appear to have the same search cost as unbloated ones, at least
     * in cases where only a single leaf page is expected to be visited.  This
     * cost is somewhat arbitrarily set at 50x cpu_operator_cost per page
     * touched.  The number of such pages is btree tree height plus one (ie,
     * we charge for the leaf page too).  As above, charge once per estimated
     * SA index descent.
     */
    descentCost = (index->tree_height + 1) * DEFAULT_PAGE_CPU_MULTIPLIER * cpu_operator_cost;
    costs.indexStartupCost += descentCost;
    costs.indexTotalCost += costs.num_sa_scans * descentCost;
 
    /*
     * If we can get an estimate of the first column's ordering correlation C
     * from pg_statistic, estimate the index correlation as C for a
     * single-column index, or C * 0.75 for multiple columns. (The idea here
     * is that multiple columns dilute the importance of the first column's
     * ordering, but don't negate it entirely.  Before 8.0 we divided the
     * correlation by the number of columns, but that seems too strong.)
     */
    if (index->indexkeys[0] != 0)
    {
        /* Simple variable --- look to stats for the underlying table */
        RangeTblEntry *rte = planner_rt_fetch(index->rel->relid, root);
 
        Assert(rte->rtekind == RTE_RELATION);
        relid = rte->relid;
        Assert(relid != InvalidOid);
        colnum = index->indexkeys[0];
 
        if (get_relation_stats_hook &&
            (*get_relation_stats_hook) (root, rte, colnum, &vardata))
        {
            /*
             * The hook took control of acquiring a stats tuple.  If it did
             * supply a tuple, it'd better have supplied a freefunc.
             */
            if (HeapTupleIsValid(vardata.statsTuple) &&
                !vardata.freefunc)
                elog(ERROR, "no function provided to release variable stats with");
        }
        else
        {
            vardata.statsTuple = SearchSysCache3(STATRELATTINH,
                                                 ObjectIdGetDatum(relid),
                                                 Int16GetDatum(colnum),
                                                 BoolGetDatum(rte->inh));
            vardata.freefunc = ReleaseSysCache;
        }
    }
    else
    {
        /* Expression --- maybe there are stats for the index itself */
        relid = index->indexoid;
        colnum = 1;
 
        if (get_index_stats_hook &&
            (*get_index_stats_hook) (root, relid, colnum, &vardata))
        {
            /*
             * The hook took control of acquiring a stats tuple.  If it did
             * supply a tuple, it'd better have supplied a freefunc.
             */
            if (HeapTupleIsValid(vardata.statsTuple) &&
                !vardata.freefunc)
                elog(ERROR, "no function provided to release variable stats with");
        }
        else
        {
            vardata.statsTuple = SearchSysCache3(STATRELATTINH,
                                                 ObjectIdGetDatum(relid),
                                                 Int16GetDatum(colnum),
                                                 BoolGetDatum(false));
            vardata.freefunc = ReleaseSysCache;
        }
    }
 
    if (HeapTupleIsValid(vardata.statsTuple))
    {
        Oid         sortop;
        AttStatsSlot sslot;
 
        sortop = get_opfamily_member(index->opfamily[0],
                                     index->opcintype[0],
                                     index->opcintype[0],
                                     BTLessStrategyNumber);
        if (OidIsValid(sortop) &&
            get_attstatsslot(&sslot, vardata.statsTuple,
                             STATISTIC_KIND_CORRELATION, sortop,
                             ATTSTATSSLOT_NUMBERS))
        {
            double      varCorrelation;
 
            Assert(sslot.nnumbers == 1);
            varCorrelation = sslot.numbers[0];
 
            if (index->reverse_sort[0])
                varCorrelation = -varCorrelation;
 
            if (index->nkeycolumns > 1)
                costs.indexCorrelation = varCorrelation * 0.75;
            else
                costs.indexCorrelation = varCorrelation;
 
            free_attstatsslot(&sslot);
        }
    }
 
    ReleaseVariableStats(vardata);
 
    *indexStartupCost = costs.indexStartupCost;
    *indexTotalCost = costs.indexTotalCost;
    *indexSelectivity = costs.indexSelectivity;
    *indexCorrelation = costs.indexCorrelation;
    *indexPages = costs.numIndexPages;
}

References add_predicate_to_index_quals(), ScalarArrayOpExpr::args, Assert, ATTSTATSSLOT_NUMBERS, BoolGetDatum(), BTEqualStrategyNumber, BTLessStrategyNumber, RestrictInfo::clause, clauselist_selectivity(), cpu_operator_cost, DEFAULT_PAGE_CPU_MULTIPLIER, elog, ERROR, estimate_array_length(), free_attstatsslot(), VariableStatData::freefunc, genericcostestimate(), get_attstatsslot(), get_index_stats_hook, get_op_opfamily_strategy(), get_opfamily_member(), get_relation_stats_hook, HeapTupleIsValid, IndexPath::indexclauses, IndexClause::indexcol, GenericCosts::indexCorrelation, IndexPath::indexinfo, IndexClause::indexquals, GenericCosts::indexSelectivity, GenericCosts::indexStartupCost, GenericCosts::indexTotalCost, RangeTblEntry::inh, Int16GetDatum(), InvalidOid, IS_NULL, IsA, JOIN_INNER, lappend(), lfirst_node, linitial_oid, lsecond, Max, Min, NIL, AttStatsSlot::nnumbers, nodeTag, NullTest::nulltesttype, GenericCosts::num_sa_scans, AttStatsSlot::numbers, GenericCosts::numIndexPages, GenericCosts::numIndexTuples, ObjectIdGetDatum(), OidIsValid, OpExpr::opno, ScalarArrayOpExpr::opno, planner_rt_fetch, ReleaseSysCache(), ReleaseVariableStats, RangeTblEntry::relid, root, RTE_RELATION, RangeTblEntry::rtekind, SearchSysCache3(), and VariableStatData::statsTuple.

Referenced by bthandler().

convert_bytea_to_scalar()

static void convert_bytea_to_scalar ( Datum  value, double *  scaledvalue, Datum  lobound, double *  scaledlobound, Datum  hibound, double *  scaledhibound  )

static

Definition at line 4734 of file selfuncs.c.

{
    bytea      *valuep = DatumGetByteaPP(value);
    bytea      *loboundp = DatumGetByteaPP(lobound);
    bytea      *hiboundp = DatumGetByteaPP(hibound);
    int         rangelo,
                rangehi,
                valuelen = VARSIZE_ANY_EXHDR(valuep),
                loboundlen = VARSIZE_ANY_EXHDR(loboundp),
                hiboundlen = VARSIZE_ANY_EXHDR(hiboundp),
                i,
                minlen;
    unsigned char *valstr = (unsigned char *) VARDATA_ANY(valuep);
    unsigned char *lostr = (unsigned char *) VARDATA_ANY(loboundp);
    unsigned char *histr = (unsigned char *) VARDATA_ANY(hiboundp);
 
    /*
     * Assume bytea data is uniformly distributed across all byte values.
     */
    rangelo = 0;
    rangehi = 255;
 
    /*
     * Now strip any common prefix of the three strings.
     */
    minlen = Min(Min(valuelen, loboundlen), hiboundlen);
    for (i = 0; i < minlen; i++)
    {
        if (*lostr != *histr || *lostr != *valstr)
            break;
        lostr++, histr++, valstr++;
        loboundlen--, hiboundlen--, valuelen--;
    }
 
    /*
     * Now we can do the conversions.
     */
    *scaledvalue = convert_one_bytea_to_scalar(valstr, valuelen, rangelo, rangehi);
    *scaledlobound = convert_one_bytea_to_scalar(lostr, loboundlen, rangelo, rangehi);
    *scaledhibound = convert_one_bytea_to_scalar(histr, hiboundlen, rangelo, rangehi);
}

References convert_one_bytea_to_scalar(), DatumGetByteaPP, i, Min, value, VARDATA_ANY, and VARSIZE_ANY_EXHDR.

Referenced by convert_to_scalar().

convert_numeric_to_scalar()

static double convert_numeric_to_scalar ( Datum  value, Oid  typid, bool *  failure  )

static

Definition at line 4453 of file selfuncs.c.

{
    switch (typid)
    {
        case BOOLOID:
            return (double) DatumGetBool(value);
        case INT2OID:
            return (double) DatumGetInt16(value);
        case INT4OID:
            return (double) DatumGetInt32(value);
        case INT8OID:
            return (double) DatumGetInt64(value);
        case FLOAT4OID:
            return (double) DatumGetFloat4(value);
        case FLOAT8OID:
            return (double) DatumGetFloat8(value);
        case NUMERICOID:
            /* Note: out-of-range values will be clamped to +-HUGE_VAL */
            return (double)
                DatumGetFloat8(DirectFunctionCall1(numeric_float8_no_overflow,
                                                   value));
        case OIDOID:
        case REGPROCOID:
        case REGPROCEDUREOID:
        case REGOPEROID:
        case REGOPERATOROID:
        case REGCLASSOID:
        case REGTYPEOID:
        case REGCOLLATIONOID:
        case REGCONFIGOID:
        case REGDICTIONARYOID:
        case REGROLEOID:
        case REGNAMESPACEOID:
            /* we can treat OIDs as integers... */
            return (double) DatumGetObjectId(value);
    }
 
    *failure = true;
    return 0;
}

References DatumGetBool(), DatumGetFloat4(), DatumGetFloat8(), DatumGetInt16(), DatumGetInt32(), DatumGetInt64(), DatumGetObjectId(), DirectFunctionCall1, numeric_float8_no_overflow(), and value.

Referenced by convert_to_scalar().

convert_one_bytea_to_scalar()

static double convert_one_bytea_to_scalar ( unsigned char *  value, int  valuelen, int  rangelo, int  rangehi  )

static

Definition at line 4782 of file selfuncs.c.

{
    double      num,
                denom,
                base;
 
    if (valuelen <= 0)
        return 0.0;             /* empty string has scalar value 0 */
 
    /*
     * Since base is 256, need not consider more than about 10 chars (even
     * this many seems like overkill)
     */
    if (valuelen > 10)
        valuelen = 10;
 
    /* Convert initial characters to fraction */
    base = rangehi - rangelo + 1;
    num = 0.0;
    denom = base;
    while (valuelen-- > 0)
    {
        int         ch = *value++;
 
        if (ch < rangelo)
            ch = rangelo - 1;
        else if (ch > rangehi)
            ch = rangehi + 1;
        num += ((double) (ch - rangelo)) / denom;
        denom *= base;
    }
 
    return num;
}

References value.

Referenced by convert_bytea_to_scalar().

convert_one_string_to_scalar()

static double convert_one_string_to_scalar ( char *  value, int  rangelo, int  rangehi  )

static

Definition at line 4595 of file selfuncs.c.

{
    int         slen = strlen(value);
    double      num,
                denom,
                base;
 
    if (slen <= 0)
        return 0.0;             /* empty string has scalar value 0 */
 
    /*
     * There seems little point in considering more than a dozen bytes from
     * the string.  Since base is at least 10, that will give us nominal
     * resolution of at least 12 decimal digits, which is surely far more
     * precision than this estimation technique has got anyway (especially in
     * non-C locales).  Also, even with the maximum possible base of 256, this
     * ensures denom cannot grow larger than 256^13 = 2.03e31, which will not
     * overflow on any known machine.
     */
    if (slen > 12)
        slen = 12;
 
    /* Convert initial characters to fraction */
    base = rangehi - rangelo + 1;
    num = 0.0;
    denom = base;
    while (slen-- > 0)
    {
        int         ch = (unsigned char) *value++;
 
        if (ch < rangelo)
            ch = rangelo - 1;
        else if (ch > rangehi)
            ch = rangehi + 1;
        num += ((double) (ch - rangelo)) / denom;
        denom *= base;
    }
 
    return num;
}

References value.

Referenced by convert_string_to_scalar().

convert_string_datum()

static char * convert_string_datum ( Datum  value, Oid  typid, Oid  collid, bool *  failure  )

static

Definition at line 4646 of file selfuncs.c.

{
    char       *val;
    pg_locale_t mylocale;
 
    switch (typid)
    {
        case CHAROID:
            val = (char *) palloc(2);
            val[0] = DatumGetChar(value);
            val[1] = '\0';
            break;
        case BPCHAROID:
        case VARCHAROID:
        case TEXTOID:
            val = TextDatumGetCString(value);
            break;
        case NAMEOID:
            {
                NameData   *nm = (NameData *) DatumGetPointer(value);
 
                val = pstrdup(NameStr(*nm));
                break;
            }
        default:
            *failure = true;
            return NULL;
    }
 
    mylocale = pg_newlocale_from_collation(collid);
 
    if (!mylocale->collate_is_c)
    {
        char       *xfrmstr;
        size_t      xfrmlen;
        size_t      xfrmlen2 PG_USED_FOR_ASSERTS_ONLY;
 
        /*
         * XXX: We could guess at a suitable output buffer size and only call
         * pg_strxfrm() twice if our guess is too small.
         *
         * XXX: strxfrm doesn't support UTF-8 encoding on Win32, it can return
         * bogus data or set an error. This is not really a problem unless it
         * crashes since it will only give an estimation error and nothing
         * fatal.
         *
         * XXX: we do not check pg_strxfrm_enabled(). On some platforms and in
         * some cases, libc strxfrm() may return the wrong results, but that
         * will only lead to an estimation error.
         */
        xfrmlen = pg_strxfrm(NULL, val, 0, mylocale);
#ifdef WIN32
 
        /*
         * On Windows, strxfrm returns INT_MAX when an error occurs. Instead
         * of trying to allocate this much memory (and fail), just return the
         * original string unmodified as if we were in the C locale.
         */
        if (xfrmlen == INT_MAX)
            return val;
#endif
        xfrmstr = (char *) palloc(xfrmlen + 1);
        xfrmlen2 = pg_strxfrm(xfrmstr, val, xfrmlen + 1, mylocale);
 
        /*
         * Some systems (e.g., glibc) can return a smaller value from the
         * second call than the first; thus the Assert must be <= not ==.
         */
        Assert(xfrmlen2 <= xfrmlen);
        pfree(val);
        val = xfrmstr;
    }
 
    return val;
}

References Assert, pg_locale_struct::collate_is_c, collid, DatumGetChar(), DatumGetPointer(), NameStr, palloc(), pfree(), pg_newlocale_from_collation(), pg_strxfrm(), PG_USED_FOR_ASSERTS_ONLY, pstrdup(), TextDatumGetCString, val, and value.

Referenced by convert_to_scalar().

convert_string_to_scalar()

static void convert_string_to_scalar ( char *  value, double *  scaledvalue, char *  lobound, double *  scaledlobound, char *  hibound, double *  scaledhibound  )

static

Definition at line 4515 of file selfuncs.c.

{
    int         rangelo,
                rangehi;
    char       *sptr;
 
    rangelo = rangehi = (unsigned char) hibound[0];
    for (sptr = lobound; *sptr; sptr++)
    {
        if (rangelo > (unsigned char) *sptr)
            rangelo = (unsigned char) *sptr;
        if (rangehi < (unsigned char) *sptr)
            rangehi = (unsigned char) *sptr;
    }
    for (sptr = hibound; *sptr; sptr++)
    {
        if (rangelo > (unsigned char) *sptr)
            rangelo = (unsigned char) *sptr;
        if (rangehi < (unsigned char) *sptr)
            rangehi = (unsigned char) *sptr;
    }
    /* If range includes any upper-case ASCII chars, make it include all */
    if (rangelo <= 'Z' && rangehi >= 'A')
    {
        if (rangelo > 'A')
            rangelo = 'A';
        if (rangehi < 'Z')
            rangehi = 'Z';
    }
    /* Ditto lower-case */
    if (rangelo <= 'z' && rangehi >= 'a')
    {
        if (rangelo > 'a')
            rangelo = 'a';
        if (rangehi < 'z')
            rangehi = 'z';
    }
    /* Ditto digits */
    if (rangelo <= '9' && rangehi >= '0')
    {
        if (rangelo > '0')
            rangelo = '0';
        if (rangehi < '9')
            rangehi = '9';
    }
 
    /*
     * If range includes less than 10 chars, assume we have not got enough
     * data, and make it include regular ASCII set.
     */
    if (rangehi - rangelo < 9)
    {
        rangelo = ' ';
        rangehi = 127;
    }
 
    /*
     * Now strip any common prefix of the three strings.
     */
    while (*lobound)
    {
        if (*lobound != *hibound || *lobound != *value)
            break;
        lobound++, hibound++, value++;
    }
 
    /*
     * Now we can do the conversions.
     */
    *scaledvalue = convert_one_string_to_scalar(value, rangelo, rangehi);
    *scaledlobound = convert_one_string_to_scalar(lobound, rangelo, rangehi);
    *scaledhibound = convert_one_string_to_scalar(hibound, rangelo, rangehi);
}

References convert_one_string_to_scalar(), and value.

Referenced by convert_to_scalar().

convert_timevalue_to_scalar()

static double convert_timevalue_to_scalar ( Datum  value, Oid  typid, bool *  failure  )

static

Definition at line 4825 of file selfuncs.c.

{
    switch (typid)
    {
        case TIMESTAMPOID:
            return DatumGetTimestamp(value);
        case TIMESTAMPTZOID:
            return DatumGetTimestampTz(value);
        case DATEOID:
            return date2timestamp_no_overflow(DatumGetDateADT(value));
        case INTERVALOID:
            {
                Interval   *interval = DatumGetIntervalP(value);
 
                /*
                 * Convert the month part of Interval to days using assumed
                 * average month length of 365.25/12.0 days.  Not too
                 * accurate, but plenty good enough for our purposes.
                 *
                 * This also works for infinite intervals, which just have all
                 * fields set to INT_MIN/INT_MAX, and so will produce a result
                 * smaller/larger than any finite interval.
                 */
                return interval->time + interval->day * (double) USECS_PER_DAY +
                    interval->month * ((DAYS_PER_YEAR / (double) MONTHS_PER_YEAR) * USECS_PER_DAY);
            }
        case TIMEOID:
            return DatumGetTimeADT(value);
        case TIMETZOID:
            {
                TimeTzADT  *timetz = DatumGetTimeTzADTP(value);
 
                /* use GMT-equivalent time */
                return (double) (timetz->time + (timetz->zone * 1000000.0));
            }
    }
 
    *failure = true;
    return 0;
}

References date2timestamp_no_overflow(), DatumGetDateADT(), DatumGetIntervalP(), DatumGetTimeADT(), DatumGetTimestamp(), DatumGetTimestampTz(), DatumGetTimeTzADTP(), DAYS_PER_YEAR, interval::month, MONTHS_PER_YEAR, TimeTzADT::time, interval::time, USECS_PER_DAY, value, and TimeTzADT::zone.

Referenced by convert_to_scalar().

convert_to_scalar()

static bool convert_to_scalar ( Datum  value, Oid  valuetypid, Oid  collid, double *  scaledvalue, Datum  lobound, Datum  hibound, Oid  boundstypid, double *  scaledlobound, double *  scaledhibound  )

static

Definition at line 4306 of file selfuncs.c.

{
    bool        failure = false;
 
    /*
     * Both the valuetypid and the boundstypid should exactly match the
     * declared input type(s) of the operator we are invoked for.  However,
     * extensions might try to use scalarineqsel as estimator for operators
     * with input type(s) we don't handle here; in such cases, we want to
     * return false, not fail.  In any case, we mustn't assume that valuetypid
     * and boundstypid are identical.
     *
     * XXX The histogram we are interpolating between points of could belong
     * to a column that's only binary-compatible with the declared type. In
     * essence we are assuming that the semantics of binary-compatible types
     * are enough alike that we can use a histogram generated with one type's
     * operators to estimate selectivity for the other's.  This is outright
     * wrong in some cases --- in particular signed versus unsigned
     * interpretation could trip us up.  But it's useful enough in the
     * majority of cases that we do it anyway.  Should think about more
     * rigorous ways to do it.
     */
    switch (valuetypid)
    {
            /*
             * Built-in numeric types
             */
        case BOOLOID:
        case INT2OID:
        case INT4OID:
        case INT8OID:
        case FLOAT4OID:
        case FLOAT8OID:
        case NUMERICOID:
        case OIDOID:
        case REGPROCOID:
        case REGPROCEDUREOID:
        case REGOPEROID:
        case REGOPERATOROID:
        case REGCLASSOID:
        case REGTYPEOID:
        case REGCOLLATIONOID:
        case REGCONFIGOID:
        case REGDICTIONARYOID:
        case REGROLEOID:
        case REGNAMESPACEOID:
            *scaledvalue = convert_numeric_to_scalar(value, valuetypid,
                                                     &failure);
            *scaledlobound = convert_numeric_to_scalar(lobound, boundstypid,
                                                       &failure);
            *scaledhibound = convert_numeric_to_scalar(hibound, boundstypid,
                                                       &failure);
            return !failure;
 
            /*
             * Built-in string types
             */
        case CHAROID:
        case BPCHAROID:
        case VARCHAROID:
        case TEXTOID:
        case NAMEOID:
            {
                char       *valstr = convert_string_datum(value, valuetypid,
                                                          collid, &failure);
                char       *lostr = convert_string_datum(lobound, boundstypid,
                                                         collid, &failure);
                char       *histr = convert_string_datum(hibound, boundstypid,
                                                         collid, &failure);
 
                /*
                 * Bail out if any of the values is not of string type.  We
                 * might leak converted strings for the other value(s), but
                 * that's not worth troubling over.
                 */
                if (failure)
                    return false;
 
                convert_string_to_scalar(valstr, scaledvalue,
                                         lostr, scaledlobound,
                                         histr, scaledhibound);
                pfree(valstr);
                pfree(lostr);
                pfree(histr);
                return true;
            }
 
            /*
             * Built-in bytea type
             */
        case BYTEAOID:
            {
                /* We only support bytea vs bytea comparison */
                if (boundstypid != BYTEAOID)
                    return false;
                convert_bytea_to_scalar(value, scaledvalue,
                                        lobound, scaledlobound,
                                        hibound, scaledhibound);
                return true;
            }
 
            /*
             * Built-in time types
             */
        case TIMESTAMPOID:
        case TIMESTAMPTZOID:
        case DATEOID:
        case INTERVALOID:
        case TIMEOID:
        case TIMETZOID:
            *scaledvalue = convert_timevalue_to_scalar(value, valuetypid,
                                                       &failure);
            *scaledlobound = convert_timevalue_to_scalar(lobound, boundstypid,
                                                         &failure);
            *scaledhibound = convert_timevalue_to_scalar(hibound, boundstypid,
                                                         &failure);
            return !failure;
 
            /*
             * Built-in network types
             */
        case INETOID:
        case CIDROID:
        case MACADDROID:
        case MACADDR8OID:
            *scaledvalue = convert_network_to_scalar(value, valuetypid,
                                                     &failure);
            *scaledlobound = convert_network_to_scalar(lobound, boundstypid,
                                                       &failure);
            *scaledhibound = convert_network_to_scalar(hibound, boundstypid,
                                                       &failure);
            return !failure;
    }
    /* Don't know how to convert */
    *scaledvalue = *scaledlobound = *scaledhibound = 0;
    return false;
}

References collid, convert_bytea_to_scalar(), convert_network_to_scalar(), convert_numeric_to_scalar(), convert_string_datum(), convert_string_to_scalar(), convert_timevalue_to_scalar(), pfree(), and value.

Referenced by ineq_histogram_selectivity().

eqjoinsel()

Datum eqjoinsel

(

PG_FUNCTION_ARGS

)

Definition at line 2269 of file selfuncs.c.

{
    PlannerInfo *root = (PlannerInfo *) PG_GETARG_POINTER(0);
    Oid         operator = PG_GETARG_OID(1);
    List       *args = (List *) PG_GETARG_POINTER(2);
 
#ifdef NOT_USED
    JoinType    jointype = (JoinType) PG_GETARG_INT16(3);
#endif
    SpecialJoinInfo *sjinfo = (SpecialJoinInfo *) PG_GETARG_POINTER(4);
    Oid         collation = PG_GET_COLLATION();
    double      selec;
    double      selec_inner;
    VariableStatData vardata1;
    VariableStatData vardata2;
    double      nd1;
    double      nd2;
    bool        isdefault1;
    bool        isdefault2;
    Oid         opfuncoid;
    AttStatsSlot sslot1;
    AttStatsSlot sslot2;
    Form_pg_statistic stats1 = NULL;
    Form_pg_statistic stats2 = NULL;
    bool        have_mcvs1 = false;
    bool        have_mcvs2 = false;
    bool        get_mcv_stats;
    bool        join_is_reversed;
    RelOptInfo *inner_rel;
 
    get_join_variables(root, args, sjinfo,
                       &vardata1, &vardata2, &join_is_reversed);
 
    nd1 = get_variable_numdistinct(&vardata1, &isdefault1);
    nd2 = get_variable_numdistinct(&vardata2, &isdefault2);
 
    opfuncoid = get_opcode(operator);
 
    memset(&sslot1, 0, sizeof(sslot1));
    memset(&sslot2, 0, sizeof(sslot2));
 
    /*
     * There is no use in fetching one side's MCVs if we lack MCVs for the
     * other side, so do a quick check to verify that both stats exist.
     */
    get_mcv_stats = (HeapTupleIsValid(vardata1.statsTuple) &&
                     HeapTupleIsValid(vardata2.statsTuple) &&
                     get_attstatsslot(&sslot1, vardata1.statsTuple,
                                      STATISTIC_KIND_MCV, InvalidOid,
                                      0) &&
                     get_attstatsslot(&sslot2, vardata2.statsTuple,
                                      STATISTIC_KIND_MCV, InvalidOid,
                                      0));
 
    if (HeapTupleIsValid(vardata1.statsTuple))
    {
        /* note we allow use of nullfrac regardless of security check */
        stats1 = (Form_pg_statistic) GETSTRUCT(vardata1.statsTuple);
        if (get_mcv_stats &&
            statistic_proc_security_check(&vardata1, opfuncoid))
            have_mcvs1 = get_attstatsslot(&sslot1, vardata1.statsTuple,
                                          STATISTIC_KIND_MCV, InvalidOid,
                                          ATTSTATSSLOT_VALUES | ATTSTATSSLOT_NUMBERS);
    }
 
    if (HeapTupleIsValid(vardata2.statsTuple))
    {
        /* note we allow use of nullfrac regardless of security check */
        stats2 = (Form_pg_statistic) GETSTRUCT(vardata2.statsTuple);
        if (get_mcv_stats &&
            statistic_proc_security_check(&vardata2, opfuncoid))
            have_mcvs2 = get_attstatsslot(&sslot2, vardata2.statsTuple,
                                          STATISTIC_KIND_MCV, InvalidOid,
                                          ATTSTATSSLOT_VALUES | ATTSTATSSLOT_NUMBERS);
    }
 
    /* We need to compute the inner-join selectivity in all cases */
    selec_inner = eqjoinsel_inner(opfuncoid, collation,
                                  &vardata1, &vardata2,
                                  nd1, nd2,
                                  isdefault1, isdefault2,
                                  &sslot1, &sslot2,
                                  stats1, stats2,
                                  have_mcvs1, have_mcvs2);
 
    switch (sjinfo->jointype)
    {
        case JOIN_INNER:
        case JOIN_LEFT:
        case JOIN_FULL:
            selec = selec_inner;
            break;
        case JOIN_SEMI:
        case JOIN_ANTI:
 
            /*
             * Look up the join's inner relation.  min_righthand is sufficient
             * information because neither SEMI nor ANTI joins permit any
             * reassociation into or out of their RHS, so the righthand will
             * always be exactly that set of rels.
             */
            inner_rel = find_join_input_rel(root, sjinfo->min_righthand);
 
            if (!join_is_reversed)
                selec = eqjoinsel_semi(opfuncoid, collation,
                                       &vardata1, &vardata2,
                                       nd1, nd2,
                                       isdefault1, isdefault2,
                                       &sslot1, &sslot2,
                                       stats1, stats2,
                                       have_mcvs1, have_mcvs2,
                                       inner_rel);
            else
            {
                Oid         commop = get_commutator(operator);
                Oid         commopfuncoid = OidIsValid(commop) ? get_opcode(commop) : InvalidOid;
 
                selec = eqjoinsel_semi(commopfuncoid, collation,
                                       &vardata2, &vardata1,
                                       nd2, nd1,
                                       isdefault2, isdefault1,
                                       &sslot2, &sslot1,
                                       stats2, stats1,
                                       have_mcvs2, have_mcvs1,
                                       inner_rel);
            }
 
            /*
             * We should never estimate the output of a semijoin to be more
             * rows than we estimate for an inner join with the same input
             * rels and join condition; it's obviously impossible for that to
             * happen.  The former estimate is N1 * Ssemi while the latter is
             * N1 * N2 * Sinner, so we may clamp Ssemi <= N2 * Sinner.  Doing
             * this is worthwhile because of the shakier estimation rules we
             * use in eqjoinsel_semi, particularly in cases where it has to
             * punt entirely.
             */
            selec = Min(selec, inner_rel->rows * selec_inner);
            break;
        default:
            /* other values not expected here */
            elog(ERROR, "unrecognized join type: %d",
                 (int) sjinfo->jointype);
            selec = 0;          /* keep compiler quiet */
            break;
    }
 
    free_attstatsslot(&sslot1);
    free_attstatsslot(&sslot2);
 
    ReleaseVariableStats(vardata1);
    ReleaseVariableStats(vardata2);
 
    CLAMP_PROBABILITY(selec);
 
    PG_RETURN_FLOAT8((float8) selec);
}

References generate_unaccent_rules::args, ATTSTATSSLOT_NUMBERS, ATTSTATSSLOT_VALUES, CLAMP_PROBABILITY, elog, eqjoinsel_inner(), eqjoinsel_semi(), ERROR, find_join_input_rel(), free_attstatsslot(), get_attstatsslot(), get_commutator(), get_join_variables(), get_opcode(), get_variable_numdistinct(), GETSTRUCT, HeapTupleIsValid, InvalidOid, JOIN_ANTI, JOIN_FULL, JOIN_INNER, JOIN_LEFT, JOIN_SEMI, SpecialJoinInfo::jointype, Min, SpecialJoinInfo::min_righthand, OidIsValid, PG_GET_COLLATION, PG_GETARG_INT16, PG_GETARG_OID, PG_GETARG_POINTER, PG_RETURN_FLOAT8, ReleaseVariableStats, root, RelOptInfo::rows, statistic_proc_security_check(), and VariableStatData::statsTuple.

Referenced by neqjoinsel().

eqjoinsel_inner()

static double eqjoinsel_inner ( Oid  opfuncoid, Oid  collation, VariableStatData *  vardata1, VariableStatData *  vardata2, double  nd1, double  nd2, bool  isdefault1, bool  isdefault2, AttStatsSlot *  sslot1, AttStatsSlot *  sslot2, Form_pg_statistic  stats1, Form_pg_statistic  stats2, bool  have_mcvs1, bool  have_mcvs2  )

static

Definition at line 2434 of file selfuncs.c.

{
    double      selec;
 
    if (have_mcvs1 && have_mcvs2)
    {
        /*
         * We have most-common-value lists for both relations.  Run through
         * the lists to see which MCVs actually join to each other with the
         * given operator.  This allows us to determine the exact join
         * selectivity for the portion of the relations represented by the MCV
         * lists.  We still have to estimate for the remaining population, but
         * in a skewed distribution this gives us a big leg up in accuracy.
         * For motivation see the analysis in Y. Ioannidis and S.
         * Christodoulakis, "On the propagation of errors in the size of join
         * results", Technical Report 1018, Computer Science Dept., University
         * of Wisconsin, Madison, March 1991 (available from ftp.cs.wisc.edu).
         */
        LOCAL_FCINFO(fcinfo, 2);
        FmgrInfo    eqproc;
        bool       *hasmatch1;
        bool       *hasmatch2;
        double      nullfrac1 = stats1->stanullfrac;
        double      nullfrac2 = stats2->stanullfrac;
        double      matchprodfreq,
                    matchfreq1,
                    matchfreq2,
                    unmatchfreq1,
                    unmatchfreq2,
                    otherfreq1,
                    otherfreq2,
                    totalsel1,
                    totalsel2;
        int         i,
                    nmatches;
 
        fmgr_info(opfuncoid, &eqproc);
 
        /*
         * Save a few cycles by setting up the fcinfo struct just once. Using
         * FunctionCallInvoke directly also avoids failure if the eqproc
         * returns NULL, though really equality functions should never do
         * that.
         */
        InitFunctionCallInfoData(*fcinfo, &eqproc, 2, collation,
                                 NULL, NULL);
        fcinfo->args[0].isnull = false;
        fcinfo->args[1].isnull = false;
 
        hasmatch1 = (bool *) palloc0(sslot1->nvalues * sizeof(bool));
        hasmatch2 = (bool *) palloc0(sslot2->nvalues * sizeof(bool));
 
        /*
         * Note we assume that each MCV will match at most one member of the
         * other MCV list.  If the operator isn't really equality, there could
         * be multiple matches --- but we don't look for them, both for speed
         * and because the math wouldn't add up...
         */
        matchprodfreq = 0.0;
        nmatches = 0;
        for (i = 0; i < sslot1->nvalues; i++)
        {
            int         j;
 
            fcinfo->args[0].value = sslot1->values[i];
 
            for (j = 0; j < sslot2->nvalues; j++)
            {
                Datum       fresult;
 
                if (hasmatch2[j])
                    continue;
                fcinfo->args[1].value = sslot2->values[j];
                fcinfo->isnull = false;
                fresult = FunctionCallInvoke(fcinfo);
                if (!fcinfo->isnull && DatumGetBool(fresult))
                {
                    hasmatch1[i] = hasmatch2[j] = true;
                    matchprodfreq += sslot1->numbers[i] * sslot2->numbers[j];
                    nmatches++;
                    break;
                }
            }
        }
        CLAMP_PROBABILITY(matchprodfreq);
        /* Sum up frequencies of matched and unmatched MCVs */
        matchfreq1 = unmatchfreq1 = 0.0;
        for (i = 0; i < sslot1->nvalues; i++)
        {
            if (hasmatch1[i])
                matchfreq1 += sslot1->numbers[i];
            else
                unmatchfreq1 += sslot1->numbers[i];
        }
        CLAMP_PROBABILITY(matchfreq1);
        CLAMP_PROBABILITY(unmatchfreq1);
        matchfreq2 = unmatchfreq2 = 0.0;
        for (i = 0; i < sslot2->nvalues; i++)
        {
            if (hasmatch2[i])
                matchfreq2 += sslot2->numbers[i];
            else
                unmatchfreq2 += sslot2->numbers[i];
        }
        CLAMP_PROBABILITY(matchfreq2);
        CLAMP_PROBABILITY(unmatchfreq2);
        pfree(hasmatch1);
        pfree(hasmatch2);
 
        /*
         * Compute total frequency of non-null values that are not in the MCV
         * lists.
         */
        otherfreq1 = 1.0 - nullfrac1 - matchfreq1 - unmatchfreq1;
        otherfreq2 = 1.0 - nullfrac2 - matchfreq2 - unmatchfreq2;
        CLAMP_PROBABILITY(otherfreq1);
        CLAMP_PROBABILITY(otherfreq2);
 
        /*
         * We can estimate the total selectivity from the point of view of
         * relation 1 as: the known selectivity for matched MCVs, plus
         * unmatched MCVs that are assumed to match against random members of
         * relation 2's non-MCV population, plus non-MCV values that are
         * assumed to match against random members of relation 2's unmatched
         * MCVs plus non-MCV values.
         */
        totalsel1 = matchprodfreq;
        if (nd2 > sslot2->nvalues)
            totalsel1 += unmatchfreq1 * otherfreq2 / (nd2 - sslot2->nvalues);
        if (nd2 > nmatches)
            totalsel1 += otherfreq1 * (otherfreq2 + unmatchfreq2) /
                (nd2 - nmatches);
        /* Same estimate from the point of view of relation 2. */
        totalsel2 = matchprodfreq;
        if (nd1 > sslot1->nvalues)
            totalsel2 += unmatchfreq2 * otherfreq1 / (nd1 - sslot1->nvalues);
        if (nd1 > nmatches)
            totalsel2 += otherfreq2 * (otherfreq1 + unmatchfreq1) /
                (nd1 - nmatches);
 
        /*
         * Use the smaller of the two estimates.  This can be justified in
         * essentially the same terms as given below for the no-stats case: to
         * a first approximation, we are estimating from the point of view of
         * the relation with smaller nd.
         */
        selec = (totalsel1 < totalsel2) ? totalsel1 : totalsel2;
    }
    else
    {
        /*
         * We do not have MCV lists for both sides.  Estimate the join
         * selectivity as MIN(1/nd1,1/nd2)*(1-nullfrac1)*(1-nullfrac2). This
         * is plausible if we assume that the join operator is strict and the
         * non-null values are about equally distributed: a given non-null
         * tuple of rel1 will join to either zero or N2*(1-nullfrac2)/nd2 rows
         * of rel2, so total join rows are at most
         * N1*(1-nullfrac1)*N2*(1-nullfrac2)/nd2 giving a join selectivity of
         * not more than (1-nullfrac1)*(1-nullfrac2)/nd2. By the same logic it
         * is not more than (1-nullfrac1)*(1-nullfrac2)/nd1, so the expression
         * with MIN() is an upper bound.  Using the MIN() means we estimate
         * from the point of view of the relation with smaller nd (since the
         * larger nd is determining the MIN).  It is reasonable to assume that
         * most tuples in this rel will have join partners, so the bound is
         * probably reasonably tight and should be taken as-is.
         *
         * XXX Can we be smarter if we have an MCV list for just one side? It
         * seems that if we assume equal distribution for the other side, we
         * end up with the same answer anyway.
         */
        double      nullfrac1 = stats1 ? stats1->stanullfrac : 0.0;
        double      nullfrac2 = stats2 ? stats2->stanullfrac : 0.0;
 
        selec = (1.0 - nullfrac1) * (1.0 - nullfrac2);
        if (nd1 > nd2)
            selec /= nd1;
        else
            selec /= nd2;
    }
 
    return selec;
}

References CLAMP_PROBABILITY, DatumGetBool(), fmgr_info(), FunctionCallInvoke, i, InitFunctionCallInfoData, j, LOCAL_FCINFO, AttStatsSlot::numbers, AttStatsSlot::nvalues, palloc0(), pfree(), and AttStatsSlot::values.

Referenced by eqjoinsel().

eqjoinsel_semi()

static double eqjoinsel_semi ( Oid  opfuncoid, Oid  collation, VariableStatData *  vardata1, VariableStatData *  vardata2, double  nd1, double  nd2, bool  isdefault1, bool  isdefault2, AttStatsSlot *  sslot1, AttStatsSlot *  sslot2, Form_pg_statistic  stats1, Form_pg_statistic  stats2, bool  have_mcvs1, bool  have_mcvs2, RelOptInfo *  inner_rel  )

static

Definition at line 2631 of file selfuncs.c.

{
    double      selec;
 
    /*
     * We clamp nd2 to be not more than what we estimate the inner relation's
     * size to be.  This is intuitively somewhat reasonable since obviously
     * there can't be more than that many distinct values coming from the
     * inner rel.  The reason for the asymmetry (ie, that we don't clamp nd1
     * likewise) is that this is the only pathway by which restriction clauses
     * applied to the inner rel will affect the join result size estimate,
     * since set_joinrel_size_estimates will multiply SEMI/ANTI selectivity by
     * only the outer rel's size.  If we clamped nd1 we'd be double-counting
     * the selectivity of outer-rel restrictions.
     *
     * We can apply this clamping both with respect to the base relation from
     * which the join variable comes (if there is just one), and to the
     * immediate inner input relation of the current join.
     *
     * If we clamp, we can treat nd2 as being a non-default estimate; it's not
     * great, maybe, but it didn't come out of nowhere either.  This is most
     * helpful when the inner relation is empty and consequently has no stats.
     */
    if (vardata2->rel)
    {
        if (nd2 >= vardata2->rel->rows)
        {
            nd2 = vardata2->rel->rows;
            isdefault2 = false;
        }
    }
    if (nd2 >= inner_rel->rows)
    {
        nd2 = inner_rel->rows;
        isdefault2 = false;
    }
 
    if (have_mcvs1 && have_mcvs2 && OidIsValid(opfuncoid))
    {
        /*
         * We have most-common-value lists for both relations.  Run through
         * the lists to see which MCVs actually join to each other with the
         * given operator.  This allows us to determine the exact join
         * selectivity for the portion of the relations represented by the MCV
         * lists.  We still have to estimate for the remaining population, but
         * in a skewed distribution this gives us a big leg up in accuracy.
         */
        LOCAL_FCINFO(fcinfo, 2);
        FmgrInfo    eqproc;
        bool       *hasmatch1;
        bool       *hasmatch2;
        double      nullfrac1 = stats1->stanullfrac;
        double      matchfreq1,
                    uncertainfrac,
                    uncertain;
        int         i,
                    nmatches,
                    clamped_nvalues2;
 
        /*
         * The clamping above could have resulted in nd2 being less than
         * sslot2->nvalues; in which case, we assume that precisely the nd2
         * most common values in the relation will appear in the join input,
         * and so compare to only the first nd2 members of the MCV list.  Of
         * course this is frequently wrong, but it's the best bet we can make.
         */
        clamped_nvalues2 = Min(sslot2->nvalues, nd2);
 
        fmgr_info(opfuncoid, &eqproc);
 
        /*
         * Save a few cycles by setting up the fcinfo struct just once. Using
         * FunctionCallInvoke directly also avoids failure if the eqproc
         * returns NULL, though really equality functions should never do
         * that.
         */
        InitFunctionCallInfoData(*fcinfo, &eqproc, 2, collation,
                                 NULL, NULL);
        fcinfo->args[0].isnull = false;
        fcinfo->args[1].isnull = false;
 
        hasmatch1 = (bool *) palloc0(sslot1->nvalues * sizeof(bool));
        hasmatch2 = (bool *) palloc0(clamped_nvalues2 * sizeof(bool));
 
        /*
         * Note we assume that each MCV will match at most one member of the
         * other MCV list.  If the operator isn't really equality, there could
         * be multiple matches --- but we don't look for them, both for speed
         * and because the math wouldn't add up...
         */
        nmatches = 0;
        for (i = 0; i < sslot1->nvalues; i++)
        {
            int         j;
 
            fcinfo->args[0].value = sslot1->values[i];
 
            for (j = 0; j < clamped_nvalues2; j++)
            {
                Datum       fresult;
 
                if (hasmatch2[j])
                    continue;
                fcinfo->args[1].value = sslot2->values[j];
                fcinfo->isnull = false;
                fresult = FunctionCallInvoke(fcinfo);
                if (!fcinfo->isnull && DatumGetBool(fresult))
                {
                    hasmatch1[i] = hasmatch2[j] = true;
                    nmatches++;
                    break;
                }
            }
        }
        /* Sum up frequencies of matched MCVs */
        matchfreq1 = 0.0;
        for (i = 0; i < sslot1->nvalues; i++)
        {
            if (hasmatch1[i])
                matchfreq1 += sslot1->numbers[i];
        }
        CLAMP_PROBABILITY(matchfreq1);
        pfree(hasmatch1);
        pfree(hasmatch2);
 
        /*
         * Now we need to estimate the fraction of relation 1 that has at
         * least one join partner.  We know for certain that the matched MCVs
         * do, so that gives us a lower bound, but we're really in the dark
         * about everything else.  Our crude approach is: if nd1 <= nd2 then
         * assume all non-null rel1 rows have join partners, else assume for
         * the uncertain rows that a fraction nd2/nd1 have join partners. We
         * can discount the known-matched MCVs from the distinct-values counts
         * before doing the division.
         *
         * Crude as the above is, it's completely useless if we don't have
         * reliable ndistinct values for both sides.  Hence, if either nd1 or
         * nd2 is default, punt and assume half of the uncertain rows have
         * join partners.
         */
        if (!isdefault1 && !isdefault2)
        {
            nd1 -= nmatches;
            nd2 -= nmatches;
            if (nd1 <= nd2 || nd2 < 0)
                uncertainfrac = 1.0;
            else
                uncertainfrac = nd2 / nd1;
        }
        else
            uncertainfrac = 0.5;
        uncertain = 1.0 - matchfreq1 - nullfrac1;
        CLAMP_PROBABILITY(uncertain);
        selec = matchfreq1 + uncertainfrac * uncertain;
    }
    else
    {
        /*
         * Without MCV lists for both sides, we can only use the heuristic
         * about nd1 vs nd2.
         */
        double      nullfrac1 = stats1 ? stats1->stanullfrac : 0.0;
 
        if (!isdefault1 && !isdefault2)
        {
            if (nd1 <= nd2 || nd2 < 0)
                selec = 1.0 - nullfrac1;
            else
                selec = (nd2 / nd1) * (1.0 - nullfrac1);
        }
        else
            selec = 0.5 * (1.0 - nullfrac1);
    }
 
    return selec;
}

References CLAMP_PROBABILITY, DatumGetBool(), fmgr_info(), FunctionCallInvoke, i, InitFunctionCallInfoData, j, LOCAL_FCINFO, Min, AttStatsSlot::numbers, AttStatsSlot::nvalues, OidIsValid, palloc0(), pfree(), VariableStatData::rel, RelOptInfo::rows, and AttStatsSlot::values.

Referenced by eqjoinsel().

eqsel()

Datum eqsel

(

PG_FUNCTION_ARGS

)

Definition at line 227 of file selfuncs.c.

{
    PG_RETURN_FLOAT8((float8) eqsel_internal(fcinfo, false));
}

References eqsel_internal(), and PG_RETURN_FLOAT8.

eqsel_internal()

static double eqsel_internal ( PG_FUNCTION_ARGS  , bool  negate  )

static

Definition at line 236 of file selfuncs.c.

{
    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);
    Oid         collation = PG_GET_COLLATION();
    VariableStatData vardata;
    Node       *other;
    bool        varonleft;
    double      selec;
 
    /*
     * When asked about <>, we do the estimation using the corresponding =
     * operator, then convert to <> via "1.0 - eq_selectivity - nullfrac".
     */
    if (negate)
    {
        operator = get_negator(operator);
        if (!OidIsValid(operator))
        {
            /* Use default selectivity (should we raise an error instead?) */
            return 1.0 - DEFAULT_EQ_SEL;
        }
    }
 
    /*
     * If expression is not variable = something or something = variable, then
     * punt and return a default estimate.
     */
    if (!get_restriction_variable(root, args, varRelid,
                                  &vardata, &other, &varonleft))
        return negate ? (1.0 - DEFAULT_EQ_SEL) : DEFAULT_EQ_SEL;
 
    /*
     * We can do a lot better if the something is a constant.  (Note: the
     * Const might result from estimation rather than being a simple constant
     * in the query.)
     */
    if (IsA(other, Const))
        selec = var_eq_const(&vardata, operator, collation,
                             ((Const *) other)->constvalue,
                             ((Const *) other)->constisnull,
                             varonleft, negate);
    else
        selec = var_eq_non_const(&vardata, operator, collation, other,
                                 varonleft, negate);
 
    ReleaseVariableStats(vardata);
 
    return selec;
}

References generate_unaccent_rules::args, DEFAULT_EQ_SEL, get_negator(), get_restriction_variable(), IsA, OidIsValid, PG_GET_COLLATION, PG_GETARG_INT32, PG_GETARG_OID, PG_GETARG_POINTER, ReleaseVariableStats, root, var_eq_const(), and var_eq_non_const().

Referenced by eqsel(), and neqsel().

estimate_array_length()

double estimate_array_length	(	PlannerInfo *	root,
		Node *	arrayexpr
	)

Definition at line 2136 of file selfuncs.c.

{
    /* look through any binary-compatible relabeling of arrayexpr */
    arrayexpr = strip_array_coercion(arrayexpr);
 
    if (arrayexpr && IsA(arrayexpr, Const))
    {
        Datum       arraydatum = ((Const *) arrayexpr)->constvalue;
        bool        arrayisnull = ((Const *) arrayexpr)->constisnull;
        ArrayType  *arrayval;
 
        if (arrayisnull)
            return 0;
        arrayval = DatumGetArrayTypeP(arraydatum);
        return ArrayGetNItems(ARR_NDIM(arrayval), ARR_DIMS(arrayval));
    }
    else if (arrayexpr && IsA(arrayexpr, ArrayExpr) &&
             !((ArrayExpr *) arrayexpr)->multidims)
    {
        return list_length(((ArrayExpr *) arrayexpr)->elements);
    }
    else if (arrayexpr)
    {
        /* See if we can find any statistics about it */
        VariableStatData vardata;
        AttStatsSlot sslot;
        double      nelem = 0;
 
        examine_variable(root, arrayexpr, 0, &vardata);
        if (HeapTupleIsValid(vardata.statsTuple))
        {
            /*
             * Found stats, so use the average element count, which is stored
             * in the last stanumbers element of the DECHIST statistics.
             * Actually that is the average count of *distinct* elements;
             * perhaps we should scale it up somewhat?
             */
            if (get_attstatsslot(&sslot, vardata.statsTuple,
                                 STATISTIC_KIND_DECHIST, InvalidOid,
                                 ATTSTATSSLOT_NUMBERS))
            {
                if (sslot.nnumbers > 0)
                    nelem = clamp_row_est(sslot.numbers[sslot.nnumbers - 1]);
                free_attstatsslot(&sslot);
            }
        }
        ReleaseVariableStats(vardata);
 
        if (nelem > 0)
            return nelem;
    }
 
    /* Else use a default guess --- this should match scalararraysel */
    return 10;
}

References ARR_DIMS, ARR_NDIM, ArrayGetNItems(), ATTSTATSSLOT_NUMBERS, clamp_row_est(), DatumGetArrayTypeP, examine_variable(), free_attstatsslot(), get_attstatsslot(), HeapTupleIsValid, InvalidOid, IsA, list_length(), AttStatsSlot::nnumbers, AttStatsSlot::numbers, ReleaseVariableStats, root, VariableStatData::statsTuple, and strip_array_coercion().

Referenced by array_unnest_support(), btcostestimate(), cost_qual_eval_walker(), cost_tidscan(), genericcostestimate(), and gincost_scalararrayopexpr().

estimate_hash_bucket_stats()

void estimate_hash_bucket_stats	(	PlannerInfo *	root,
		Node *	hashkey,
		double	nbuckets,
		Selectivity *	mcv_freq,
		Selectivity *	bucketsize_frac
	)

Definition at line 3799 of file selfuncs.c.

{
    VariableStatData vardata;
    double      estfract,
                ndistinct,
                stanullfrac,
                avgfreq;
    bool        isdefault;
    AttStatsSlot sslot;
 
    examine_variable(root, hashkey, 0, &vardata);
 
    /* Look up the frequency of the most common value, if available */
    *mcv_freq = 0.0;
 
    if (HeapTupleIsValid(vardata.statsTuple))
    {
        if (get_attstatsslot(&sslot, vardata.statsTuple,
                             STATISTIC_KIND_MCV, InvalidOid,
                             ATTSTATSSLOT_NUMBERS))
        {
            /*
             * The first MCV stat is for the most common value.
             */
            if (sslot.nnumbers > 0)
                *mcv_freq = sslot.numbers[0];
            free_attstatsslot(&sslot);
        }
    }
 
    /* Get number of distinct values */
    ndistinct = get_variable_numdistinct(&vardata, &isdefault);
 
    /*
     * If ndistinct isn't real, punt.  We normally return 0.1, but if the
     * mcv_freq is known to be even higher than that, use it instead.
     */
    if (isdefault)
    {
        *bucketsize_frac = (Selectivity) Max(0.1, *mcv_freq);
        ReleaseVariableStats(vardata);
        return;
    }
 
    /* Get fraction that are null */
    if (HeapTupleIsValid(vardata.statsTuple))
    {
        Form_pg_statistic stats;
 
        stats = (Form_pg_statistic) GETSTRUCT(vardata.statsTuple);
        stanullfrac = stats->stanullfrac;
    }
    else
        stanullfrac = 0.0;
 
    /* Compute avg freq of all distinct data values in raw relation */
    avgfreq = (1.0 - stanullfrac) / ndistinct;
 
    /*
     * Adjust ndistinct to account for restriction clauses.  Observe we are
     * assuming that the data distribution is affected uniformly by the
     * restriction clauses!
     *
     * XXX Possibly better way, but much more expensive: multiply by
     * selectivity of rel's restriction clauses that mention the target Var.
     */
    if (vardata.rel && vardata.rel->tuples > 0)
    {
        ndistinct *= vardata.rel->rows / vardata.rel->tuples;
        ndistinct = clamp_row_est(ndistinct);
    }
 
    /*
     * Initial estimate of bucketsize fraction is 1/nbuckets as long as the
     * number of buckets is less than the expected number of distinct values;
     * otherwise it is 1/ndistinct.
     */
    if (ndistinct > nbuckets)
        estfract = 1.0 / nbuckets;
    else
        estfract = 1.0 / ndistinct;
 
    /*
     * Adjust estimated bucketsize upward to account for skewed distribution.
     */
    if (avgfreq > 0.0 && *mcv_freq > avgfreq)
        estfract *= *mcv_freq / avgfreq;
 
    /*
     * Clamp bucketsize to sane range (the above adjustment could easily
     * produce an out-of-range result).  We set the lower bound a little above
     * zero, since zero isn't a very sane result.
     */
    if (estfract < 1.0e-6)
        estfract = 1.0e-6;
    else if (estfract > 1.0)
        estfract = 1.0;
 
    *bucketsize_frac = (Selectivity) estfract;
 
    ReleaseVariableStats(vardata);
}

References ATTSTATSSLOT_NUMBERS, clamp_row_est(), examine_variable(), free_attstatsslot(), get_attstatsslot(), get_variable_numdistinct(), GETSTRUCT, HeapTupleIsValid, InvalidOid, Max, AttStatsSlot::nnumbers, AttStatsSlot::numbers, VariableStatData::rel, ReleaseVariableStats, root, RelOptInfo::rows, VariableStatData::statsTuple, and RelOptInfo::tuples.

Referenced by final_cost_hashjoin().

estimate_hashagg_tablesize()

double estimate_hashagg_tablesize	(	PlannerInfo *	root,
		Path *	path,
		const AggClauseCosts *	agg_costs,
		double	dNumGroups
	)

Definition at line 3918 of file selfuncs.c.

{
    Size        hashentrysize;
 
    hashentrysize = hash_agg_entry_size(list_length(root->aggtransinfos),
                                        path->pathtarget->width,
                                        agg_costs->transitionSpace);
 
    /*
     * Note that this disregards the effect of fill-factor and growth policy
     * of the hash table.  That's probably ok, given that the default
     * fill-factor is relatively high.  It'd be hard to meaningfully factor in
     * "double-in-size" growth policies here.
     */
    return hashentrysize * dNumGroups;
}

References hash_agg_entry_size(), list_length(), root, and AggClauseCosts::transitionSpace.

Referenced by consider_groupingsets_paths().

estimate_multivariate_ndistinct()

static bool estimate_multivariate_ndistinct ( PlannerInfo *  root, RelOptInfo *  rel, List **  varinfos, double *  ndistinct  )

static

Definition at line 3955 of file selfuncs.c.

{
    ListCell   *lc;
    int         nmatches_vars;
    int         nmatches_exprs;
    Oid         statOid = InvalidOid;
    MVNDistinct *stats;
    StatisticExtInfo *matched_info = NULL;
    RangeTblEntry *rte = planner_rt_fetch(rel->relid, root);
 
    /* bail out immediately if the table has no extended statistics */
    if (!rel->statlist)
        return false;
 
    /* look for the ndistinct statistics object matching the most vars */
    nmatches_vars = 0;          /* we require at least two matches */
    nmatches_exprs = 0;
    foreach(lc, rel->statlist)
    {
        ListCell   *lc2;
        StatisticExtInfo *info = (StatisticExtInfo *) lfirst(lc);
        int         nshared_vars = 0;
        int         nshared_exprs = 0;
 
        /* skip statistics of other kinds */
        if (info->kind != STATS_EXT_NDISTINCT)
            continue;
 
        /* skip statistics with mismatching stxdinherit value */
        if (info->inherit != rte->inh)
            continue;
 
        /*
         * Determine how many expressions (and variables in non-matched
         * expressions) match. We'll then use these numbers to pick the
         * statistics object that best matches the clauses.
         */
        foreach(lc2, *varinfos)
        {
            ListCell   *lc3;
            GroupVarInfo *varinfo = (GroupVarInfo *) lfirst(lc2);
            AttrNumber  attnum;
 
            Assert(varinfo->rel == rel);
 
            /* simple Var, search in statistics keys directly */
            if (IsA(varinfo->var, Var))
            {
                attnum = ((Var *) varinfo->var)->varattno;
 
                /*
                 * Ignore system attributes - we don't support statistics on
                 * them, so can't match them (and it'd fail as the values are
                 * negative).
                 */
                if (!AttrNumberIsForUserDefinedAttr(attnum))
                    continue;
 
                if (bms_is_member(attnum, info->keys))
                    nshared_vars++;
 
                continue;
            }
 
            /* expression - see if it's in the statistics object */
            foreach(lc3, info->exprs)
            {
                Node       *expr = (Node *) lfirst(lc3);
 
                if (equal(varinfo->var, expr))
                {
                    nshared_exprs++;
                    break;
                }
            }
        }
 
        if (nshared_vars + nshared_exprs < 2)
            continue;
 
        /*
         * Does this statistics object match more columns than the currently
         * best object?  If so, use this one instead.
         *
         * XXX This should break ties using name of the object, or something
         * like that, to make the outcome stable.
         */
        if ((nshared_exprs > nmatches_exprs) ||
            (((nshared_exprs == nmatches_exprs)) && (nshared_vars > nmatches_vars)))
        {
            statOid = info->statOid;
            nmatches_vars = nshared_vars;
            nmatches_exprs = nshared_exprs;
            matched_info = info;
        }
    }
 
    /* No match? */
    if (statOid == InvalidOid)
        return false;
 
    Assert(nmatches_vars + nmatches_exprs > 1);
 
    stats = statext_ndistinct_load(statOid, rte->inh);
 
    /*
     * If we have a match, search it for the specific item that matches (there
     * must be one), and construct the output values.
     */
    if (stats)
    {
        int         i;
        List       *newlist = NIL;
        MVNDistinctItem *item = NULL;
        ListCell   *lc2;
        Bitmapset  *matched = NULL;
        AttrNumber  attnum_offset;
 
        /*
         * How much we need to offset the attnums? If there are no
         * expressions, no offset is needed. Otherwise offset enough to move
         * the lowest one (which is equal to number of expressions) to 1.
         */
        if (matched_info->exprs)
            attnum_offset = (list_length(matched_info->exprs) + 1);
        else
            attnum_offset = 0;
 
        /* see what actually matched */
        foreach(lc2, *varinfos)
        {
            ListCell   *lc3;
            int         idx;
            bool        found = false;
 
            GroupVarInfo *varinfo = (GroupVarInfo *) lfirst(lc2);
 
            /*
             * Process a simple Var expression, by matching it to keys
             * directly. If there's a matching expression, we'll try matching
             * it later.
             */
            if (IsA(varinfo->var, Var))
            {
                AttrNumber  attnum = ((Var *) varinfo->var)->varattno;
 
                /*
                 * Ignore expressions on system attributes. Can't rely on the
                 * bms check for negative values.
                 */
                if (!AttrNumberIsForUserDefinedAttr(attnum))
                    continue;
 
                /* Is the variable covered by the statistics object? */
                if (!bms_is_member(attnum, matched_info->keys))
                    continue;
 
                attnum = attnum + attnum_offset;
 
                /* ensure sufficient offset */
                Assert(AttrNumberIsForUserDefinedAttr(attnum));
 
                matched = bms_add_member(matched, attnum);
 
                found = true;
            }
 
            /*
             * XXX Maybe we should allow searching the expressions even if we
             * found an attribute matching the expression? That would handle
             * trivial expressions like "(a)" but it seems fairly useless.
             */
            if (found)
                continue;
 
            /* expression - see if it's in the statistics object */
            idx = 0;
            foreach(lc3, matched_info->exprs)
            {
                Node       *expr = (Node *) lfirst(lc3);
 
                if (equal(varinfo->var, expr))
                {
                    AttrNumber  attnum = -(idx + 1);
 
                    attnum = attnum + attnum_offset;
 
                    /* ensure sufficient offset */
                    Assert(AttrNumberIsForUserDefinedAttr(attnum));
 
                    matched = bms_add_member(matched, attnum);
 
                    /* there should be just one matching expression */
                    break;
                }
 
                idx++;
            }
        }
 
        /* Find the specific item that exactly matches the combination */
        for (i = 0; i < stats->nitems; i++)
        {
            int         j;
            MVNDistinctItem *tmpitem = &stats->items[i];
 
            if (tmpitem->nattributes != bms_num_members(matched))
                continue;
 
            /* assume it's the right item */
            item = tmpitem;
 
            /* check that all item attributes/expressions fit the match */
            for (j = 0; j < tmpitem->nattributes; j++)
            {
                AttrNumber  attnum = tmpitem->attributes[j];
 
                /*
                 * Thanks to how we constructed the matched bitmap above, we
                 * can just offset all attnums the same way.
                 */
                attnum = attnum + attnum_offset;
 
                if (!bms_is_member(attnum, matched))
                {
                    /* nah, it's not this item */
                    item = NULL;
                    break;
                }
            }
 
            /*
             * If the item has all the matched attributes, we know it's the
             * right one - there can't be a better one. matching more.
             */
            if (item)
                break;
        }
 
        /*
         * Make sure we found an item. There has to be one, because ndistinct
         * statistics includes all combinations of attributes.
         */
        if (!item)
            elog(ERROR, "corrupt MVNDistinct entry");
 
        /* Form the output varinfo list, keeping only unmatched ones */
        foreach(lc, *varinfos)
        {
            GroupVarInfo *varinfo = (GroupVarInfo *) lfirst(lc);
            ListCell   *lc3;
            bool        found = false;
 
            /*
             * Let's look at plain variables first, because it's the most
             * common case and the check is quite cheap. We can simply get the
             * attnum and check (with an offset) matched bitmap.
             */
            if (IsA(varinfo->var, Var))
            {
                AttrNumber  attnum = ((Var *) varinfo->var)->varattno;
 
                /*
                 * If it's a system attribute, we're done. We don't support
                 * extended statistics on system attributes, so it's clearly
                 * not matched. Just keep the expression and continue.
                 */
                if (!AttrNumberIsForUserDefinedAttr(attnum))
                {
                    newlist = lappend(newlist, varinfo);
                    continue;
                }
 
                /* apply the same offset as above */
                attnum += attnum_offset;
 
                /* if it's not matched, keep the varinfo */
                if (!bms_is_member(attnum, matched))
                    newlist = lappend(newlist, varinfo);
 
                /* The rest of the loop deals with complex expressions. */
                continue;
            }
 
            /*
             * Process complex expressions, not just simple Vars.
             *
             * First, we search for an exact match of an expression. If we
             * find one, we can just discard the whole GroupVarInfo, with all
             * the variables we extracted from it.
             *
             * Otherwise we inspect the individual vars, and try matching it
             * to variables in the item.
             */
            foreach(lc3, matched_info->exprs)
            {
                Node       *expr = (Node *) lfirst(lc3);
 
                if (equal(varinfo->var, expr))
                {
                    found = true;
                    break;
                }
            }
 
            /* found exact match, skip */
            if (found)
                continue;
 
            newlist = lappend(newlist, varinfo);
        }
 
        *varinfos = newlist;
        *ndistinct = item->ndistinct;
        return true;
    }
 
    return false;
}

References Assert, attnum, MVNDistinctItem::attributes, AttrNumberIsForUserDefinedAttr, bms_add_member(), bms_is_member(), bms_num_members(), elog, equal(), ERROR, StatisticExtInfo::exprs, i, idx(), RangeTblEntry::inh, StatisticExtInfo::inherit, InvalidOid, IsA, MVNDistinct::items, j, StatisticExtInfo::keys, StatisticExtInfo::kind, lappend(), lfirst, list_length(), MVNDistinctItem::nattributes, MVNDistinctItem::ndistinct, NIL, MVNDistinct::nitems, planner_rt_fetch, GroupVarInfo::rel, RelOptInfo::relid, root, statext_ndistinct_load(), RelOptInfo::statlist, StatisticExtInfo::statOid, and GroupVarInfo::var.

Referenced by estimate_num_groups().

estimate_num_groups()

double estimate_num_groups	(	PlannerInfo *	root,
		List *	groupExprs,
		double	input_rows,
		List **	pgset,
		EstimationInfo *	estinfo
	)

Definition at line 3417 of file selfuncs.c.

{
    List       *varinfos = NIL;
    double      srf_multiplier = 1.0;
    double      numdistinct;
    ListCell   *l;
    int         i;
 
    /* Zero the estinfo output parameter, if non-NULL */
    if (estinfo != NULL)
        memset(estinfo, 0, sizeof(EstimationInfo));
 
    /*
     * We don't ever want to return an estimate of zero groups, as that tends
     * to lead to division-by-zero and other unpleasantness.  The input_rows
     * estimate is usually already at least 1, but clamp it just in case it
     * isn't.
     */
    input_rows = clamp_row_est(input_rows);
 
    /*
     * If no grouping columns, there's exactly one group.  (This can't happen
     * for normal cases with GROUP BY or DISTINCT, but it is possible for
     * corner cases with set operations.)
     */
    if (groupExprs == NIL || (pgset && *pgset == NIL))
        return 1.0;
 
    /*
     * Count groups derived from boolean grouping expressions.  For other
     * expressions, find the unique Vars used, treating an expression as a Var
     * if we can find stats for it.  For each one, record the statistical
     * estimate of number of distinct values (total in its table, without
     * regard for filtering).
     */
    numdistinct = 1.0;
 
    i = 0;
    foreach(l, groupExprs)
    {
        Node       *groupexpr = (Node *) lfirst(l);
        double      this_srf_multiplier;
        VariableStatData vardata;
        List       *varshere;
        ListCell   *l2;
 
        /* is expression in this grouping set? */
        if (pgset && !list_member_int(*pgset, i++))
            continue;
 
        /*
         * Set-returning functions in grouping columns are a bit problematic.
         * The code below will effectively ignore their SRF nature and come up
         * with a numdistinct estimate as though they were scalar functions.
         * We compensate by scaling up the end result by the largest SRF
         * rowcount estimate.  (This will be an overestimate if the SRF
         * produces multiple copies of any output value, but it seems best to
         * assume the SRF's outputs are distinct.  In any case, it's probably
         * pointless to worry too much about this without much better
         * estimates for SRF output rowcounts than we have today.)
         */
        this_srf_multiplier = expression_returns_set_rows(root, groupexpr);
        if (srf_multiplier < this_srf_multiplier)
            srf_multiplier = this_srf_multiplier;
 
        /* Short-circuit for expressions returning boolean */
        if (exprType(groupexpr) == BOOLOID)
        {
            numdistinct *= 2.0;
            continue;
        }
 
        /*
         * If examine_variable is able to deduce anything about the GROUP BY
         * expression, treat it as a single variable even if it's really more
         * complicated.
         *
         * XXX This has the consequence that if there's a statistics object on
         * the expression, we don't split it into individual Vars. This
         * affects our selection of statistics in
         * estimate_multivariate_ndistinct, because it's probably better to
         * use more accurate estimate for each expression and treat them as
         * independent, than to combine estimates for the extracted variables
         * when we don't know how that relates to the expressions.
         */
        examine_variable(root, groupexpr, 0, &vardata);
        if (HeapTupleIsValid(vardata.statsTuple) || vardata.isunique)
        {
            varinfos = add_unique_group_var(root, varinfos,
                                            groupexpr, &vardata);
            ReleaseVariableStats(vardata);
            continue;
        }
        ReleaseVariableStats(vardata);
 
        /*
         * Else pull out the component Vars.  Handle PlaceHolderVars by
         * recursing into their arguments (effectively assuming that the
         * PlaceHolderVar doesn't change the number of groups, which boils
         * down to ignoring the possible addition of nulls to the result set).
         */
        varshere = pull_var_clause(groupexpr,
                                   PVC_RECURSE_AGGREGATES |
                                   PVC_RECURSE_WINDOWFUNCS |
                                   PVC_RECURSE_PLACEHOLDERS);
 
        /*
         * If we find any variable-free GROUP BY item, then either it is a
         * constant (and we can ignore it) or it contains a volatile function;
         * in the latter case we punt and assume that each input row will
         * yield a distinct group.
         */
        if (varshere == NIL)
        {
            if (contain_volatile_functions(groupexpr))
                return input_rows;
            continue;
        }
 
        /*
         * Else add variables to varinfos list
         */
        foreach(l2, varshere)
        {
            Node       *var = (Node *) lfirst(l2);
 
            examine_variable(root, var, 0, &vardata);
            varinfos = add_unique_group_var(root, varinfos, var, &vardata);
            ReleaseVariableStats(vardata);
        }
    }
 
    /*
     * If now no Vars, we must have an all-constant or all-boolean GROUP BY
     * list.
     */
    if (varinfos == NIL)
    {
        /* Apply SRF multiplier as we would do in the long path */
        numdistinct *= srf_multiplier;
        /* Round off */
        numdistinct = ceil(numdistinct);
        /* Guard against out-of-range answers */
        if (numdistinct > input_rows)
            numdistinct = input_rows;
        if (numdistinct < 1.0)
            numdistinct = 1.0;
        return numdistinct;
    }
 
    /*
     * Group Vars by relation and estimate total numdistinct.
     *
     * For each iteration of the outer loop, we process the frontmost Var in
     * varinfos, plus all other Vars in the same relation.  We remove these
     * Vars from the newvarinfos list for the next iteration. This is the
     * easiest way to group Vars of same rel together.
     */
    do
    {
        GroupVarInfo *varinfo1 = (GroupVarInfo *) linitial(varinfos);
        RelOptInfo *rel = varinfo1->rel;
        double      reldistinct = 1;
        double      relmaxndistinct = reldistinct;
        int         relvarcount = 0;
        List       *newvarinfos = NIL;
        List       *relvarinfos = NIL;
 
        /*
         * Split the list of varinfos in two - one for the current rel, one
         * for remaining Vars on other rels.
         */
        relvarinfos = lappend(relvarinfos, varinfo1);
        for_each_from(l, varinfos, 1)
        {
            GroupVarInfo *varinfo2 = (GroupVarInfo *) lfirst(l);
 
            if (varinfo2->rel == varinfo1->rel)
            {
                /* varinfos on current rel */
                relvarinfos = lappend(relvarinfos, varinfo2);
            }
            else
            {
                /* not time to process varinfo2 yet */
                newvarinfos = lappend(newvarinfos, varinfo2);
            }
        }
 
        /*
         * Get the numdistinct estimate for the Vars of this rel.  We
         * iteratively search for multivariate n-distinct with maximum number
         * of vars; assuming that each var group is independent of the others,
         * we multiply them together.  Any remaining relvarinfos after no more
         * multivariate matches are found are assumed independent too, so
         * their individual ndistinct estimates are multiplied also.
         *
         * While iterating, count how many separate numdistinct values we
         * apply.  We apply a fudge factor below, but only if we multiplied
         * more than one such values.
         */
        while (relvarinfos)
        {
            double      mvndistinct;
 
            if (estimate_multivariate_ndistinct(root, rel, &relvarinfos,
                                                &mvndistinct))
            {
                reldistinct *= mvndistinct;
                if (relmaxndistinct < mvndistinct)
                    relmaxndistinct = mvndistinct;
                relvarcount++;
            }
            else
            {
                foreach(l, relvarinfos)
                {
                    GroupVarInfo *varinfo2 = (GroupVarInfo *) lfirst(l);
 
                    reldistinct *= varinfo2->ndistinct;
                    if (relmaxndistinct < varinfo2->ndistinct)
                        relmaxndistinct = varinfo2->ndistinct;
                    relvarcount++;
 
                    /*
                     * When varinfo2's isdefault is set then we'd better set
                     * the SELFLAG_USED_DEFAULT bit in the EstimationInfo.
                     */
                    if (estinfo != NULL && varinfo2->isdefault)
                        estinfo->flags |= SELFLAG_USED_DEFAULT;
                }
 
                /* we're done with this relation */
                relvarinfos = NIL;
            }
        }
 
        /*
         * Sanity check --- don't divide by zero if empty relation.
         */
        Assert(IS_SIMPLE_REL(rel));
        if (rel->tuples > 0)
        {
            /*
             * Clamp to size of rel, or size of rel / 10 if multiple Vars. The
             * fudge factor is because the Vars are probably correlated but we
             * don't know by how much.  We should never clamp to less than the
             * largest ndistinct value for any of the Vars, though, since
             * there will surely be at least that many groups.
             */
            double      clamp = rel->tuples;
 
            if (relvarcount > 1)
            {
                clamp *= 0.1;
                if (clamp < relmaxndistinct)
                {
                    clamp = relmaxndistinct;
                    /* for sanity in case some ndistinct is too large: */
                    if (clamp > rel->tuples)
                        clamp = rel->tuples;
                }
            }
            if (reldistinct > clamp)
                reldistinct = clamp;
 
            /*
             * Update the estimate based on the restriction selectivity,
             * guarding against division by zero when reldistinct is zero.
             * Also skip this if we know that we are returning all rows.
             */
            if (reldistinct > 0 && rel->rows < rel->tuples)
            {
                /*
                 * Given a table containing N rows with n distinct values in a
                 * uniform distribution, if we select p rows at random then
                 * the expected number of distinct values selected is
                 *
                 * n * (1 - product((N-N/n-i)/(N-i), i=0..p-1))
                 *
                 * = n * (1 - (N-N/n)! / (N-N/n-p)! * (N-p)! / N!)
                 *
                 * See "Approximating block accesses in database
                 * organizations", S. B. Yao, Communications of the ACM,
                 * Volume 20 Issue 4, April 1977 Pages 260-261.
                 *
                 * Alternatively, re-arranging the terms from the factorials,
                 * this may be written as
                 *
                 * n * (1 - product((N-p-i)/(N-i), i=0..N/n-1))
                 *
                 * This form of the formula is more efficient to compute in
                 * the common case where p is larger than N/n.  Additionally,
                 * as pointed out by Dell'Era, if i << N for all terms in the
                 * product, it can be approximated by
                 *
                 * n * (1 - ((N-p)/N)^(N/n))
                 *
                 * See "Expected distinct values when selecting from a bag
                 * without replacement", Alberto Dell'Era,
                 * http://www.adellera.it/investigations/distinct_balls/.
                 *
                 * The condition i << N is equivalent to n >> 1, so this is a
                 * good approximation when the number of distinct values in
                 * the table is large.  It turns out that this formula also
                 * works well even when n is small.
                 */
                reldistinct *=
                    (1 - pow((rel->tuples - rel->rows) / rel->tuples,
                             rel->tuples / reldistinct));
            }
            reldistinct = clamp_row_est(reldistinct);
 
            /*
             * Update estimate of total distinct groups.
             */
            numdistinct *= reldistinct;
        }
 
        varinfos = newvarinfos;
    } while (varinfos != NIL);
 
    /* Now we can account for the effects of any SRFs */
    numdistinct *= srf_multiplier;
 
    /* Round off */
    numdistinct = ceil(numdistinct);
 
    /* Guard against out-of-range answers */
    if (numdistinct > input_rows)
        numdistinct = input_rows;
    if (numdistinct < 1.0)
        numdistinct = 1.0;
 
    return numdistinct;
}

References add_unique_group_var(), Assert, clamp_row_est(), contain_volatile_functions(), estimate_multivariate_ndistinct(), examine_variable(), expression_returns_set_rows(), exprType(), EstimationInfo::flags, for_each_from, HeapTupleIsValid, i, IS_SIMPLE_REL, GroupVarInfo::isdefault, VariableStatData::isunique, lappend(), lfirst, linitial, list_member_int(), GroupVarInfo::ndistinct, NIL, pull_var_clause(), PVC_RECURSE_AGGREGATES, PVC_RECURSE_PLACEHOLDERS, PVC_RECURSE_WINDOWFUNCS, GroupVarInfo::rel, ReleaseVariableStats, root, RelOptInfo::rows, SELFLAG_USED_DEFAULT, VariableStatData::statsTuple, and RelOptInfo::tuples.

Referenced by adjust_rowcount_for_semijoins(), build_setop_child_paths(), cost_incremental_sort(), cost_memoize_rescan(), create_final_distinct_paths(), create_partial_distinct_paths(), create_unique_path(), estimate_path_cost_size(), get_number_of_groups(), and get_windowclause_startup_tuples().

examine_simple_variable()

static void examine_simple_variable ( PlannerInfo *  root, Var *  var, VariableStatData *  vardata  )

static

Definition at line 5414 of file selfuncs.c.

{
    RangeTblEntry *rte = root->simple_rte_array[var->varno];
 
    Assert(IsA(rte, RangeTblEntry));
 
    if (get_relation_stats_hook &&
        (*get_relation_stats_hook) (root, rte, var->varattno, vardata))
    {
        /*
         * The hook took control of acquiring a stats tuple.  If it did supply
         * a tuple, it'd better have supplied a freefunc.
         */
        if (HeapTupleIsValid(vardata->statsTuple) &&
            !vardata->freefunc)
            elog(ERROR, "no function provided to release variable stats with");
    }
    else if (rte->rtekind == RTE_RELATION)
    {
        /*
         * Plain table or parent of an inheritance appendrel, so look up the
         * column in pg_statistic
         */
        vardata->statsTuple = SearchSysCache3(STATRELATTINH,
                                              ObjectIdGetDatum(rte->relid),
                                              Int16GetDatum(var->varattno),
                                              BoolGetDatum(rte->inh));
        vardata->freefunc = ReleaseSysCache;
 
        if (HeapTupleIsValid(vardata->statsTuple))
        {
            RelOptInfo *onerel = find_base_rel_noerr(root, var->varno);
            Oid         userid;
 
            /*
             * Check if user has permission to read this column.  We require
             * all rows to be accessible, so there must be no securityQuals
             * from security barrier views or RLS policies.
             *
             * Normally the Var will have an associated RelOptInfo from which
             * we can find out which userid to do the check as; but it might
             * not if it's a RETURNING Var for an INSERT target relation.  In
             * that case use the RTEPermissionInfo associated with the RTE.
             */
            if (onerel)
                userid = onerel->userid;
            else
            {
                RTEPermissionInfo *perminfo;
 
                perminfo = getRTEPermissionInfo(root->parse->rteperminfos, rte);
                userid = perminfo->checkAsUser;
            }
            if (!OidIsValid(userid))
                userid = GetUserId();
 
            vardata->acl_ok =
                rte->securityQuals == NIL &&
                ((pg_class_aclcheck(rte->relid, userid,
                                    ACL_SELECT) == ACLCHECK_OK) ||
                 (pg_attribute_aclcheck(rte->relid, var->varattno, userid,
                                        ACL_SELECT) == ACLCHECK_OK));
 
            /*
             * If the user doesn't have permissions to access an inheritance
             * child relation or specifically this attribute, check the
             * permissions of the table/column actually mentioned in the
             * query, since most likely the user does have that permission
             * (else the query will fail at runtime), and if the user can read
             * the column there then he can get the values of the child table
             * too.  To do that, we must find out which of the root parent's
             * attributes the child relation's attribute corresponds to.
             */
            if (!vardata->acl_ok && var->varattno > 0 &&
                root->append_rel_array != NULL)
            {
                AppendRelInfo *appinfo;
                Index       varno = var->varno;
                int         varattno = var->varattno;
                bool        found = false;
 
                appinfo = root->append_rel_array[varno];
 
                /*
                 * Partitions are mapped to their immediate parent, not the
                 * root parent, so must be ready to walk up multiple
                 * AppendRelInfos.  But stop if we hit a parent that is not
                 * RTE_RELATION --- that's a flattened UNION ALL subquery, not
                 * an inheritance parent.
                 */
                while (appinfo &&
                       planner_rt_fetch(appinfo->parent_relid,
                                        root)->rtekind == RTE_RELATION)
                {
                    int         parent_varattno;
 
                    found = false;
                    if (varattno <= 0 || varattno > appinfo->num_child_cols)
                        break;  /* safety check */
                    parent_varattno = appinfo->parent_colnos[varattno - 1];
                    if (parent_varattno == 0)
                        break;  /* Var is local to child */
 
                    varno = appinfo->parent_relid;
                    varattno = parent_varattno;
                    found = true;
 
                    /* If the parent is itself a child, continue up. */
                    appinfo = root->append_rel_array[varno];
                }
 
                /*
                 * In rare cases, the Var may be local to the child table, in
                 * which case, we've got to live with having no access to this
                 * column's stats.
                 */
                if (!found)
                    return;
 
                /* Repeat the access check on this parent rel & column */
                rte = planner_rt_fetch(varno, root);
                Assert(rte->rtekind == RTE_RELATION);
 
                /*
                 * Fine to use the same userid as it's the same in all
                 * relations of a given inheritance tree.
                 */
                vardata->acl_ok =
                    rte->securityQuals == NIL &&
                    ((pg_class_aclcheck(rte->relid, userid,
                                        ACL_SELECT) == ACLCHECK_OK) ||
                     (pg_attribute_aclcheck(rte->relid, varattno, userid,
                                            ACL_SELECT) == ACLCHECK_OK));
            }
        }
        else
        {
            /* suppress any possible leakproofness checks later */
            vardata->acl_ok = true;
        }
    }
    else if ((rte->rtekind == RTE_SUBQUERY && !rte->inh) ||
             (rte->rtekind == RTE_CTE && !rte->self_reference))
    {
        /*
         * Plain subquery (not one that was converted to an appendrel) or
         * non-recursive CTE.  In either case, we can try to find out what the
         * Var refers to within the subquery.  We skip this for appendrel and
         * recursive-CTE cases because any column stats we did find would
         * likely not be very relevant.
         */
        PlannerInfo *subroot;
        Query      *subquery;
        List       *subtlist;
        TargetEntry *ste;
 
        /*
         * Punt if it's a whole-row var rather than a plain column reference.
         */
        if (var->varattno == InvalidAttrNumber)
            return;
 
        /*
         * Otherwise, find the subquery's planner subroot.
         */
        if (rte->rtekind == RTE_SUBQUERY)
        {
            RelOptInfo *rel;
 
            /*
             * Fetch RelOptInfo for subquery.  Note that we don't change the
             * rel returned in vardata, since caller expects it to be a rel of
             * the caller's query level.  Because we might already be
             * recursing, we can't use that rel pointer either, but have to
             * look up the Var's rel afresh.
             */
            rel = find_base_rel(root, var->varno);
 
            subroot = rel->subroot;
        }
        else
        {
            /* CTE case is more difficult */
            PlannerInfo *cteroot;
            Index       levelsup;
            int         ndx;
            int         plan_id;
            ListCell   *lc;
 
            /*
             * Find the referenced CTE, and locate the subroot previously made
             * for it.
             */
            levelsup = rte->ctelevelsup;
            cteroot = root;
            while (levelsup-- > 0)
            {
                cteroot = cteroot->parent_root;
                if (!cteroot)   /* shouldn't happen */
                    elog(ERROR, "bad levelsup for CTE \"%s\"", rte->ctename);
            }
 
            /*
             * Note: cte_plan_ids can be shorter than cteList, if we are still
             * working on planning the CTEs (ie, this is a side-reference from
             * another CTE).  So we mustn't use forboth here.
             */
            ndx = 0;
            foreach(lc, cteroot->parse->cteList)
            {
                CommonTableExpr *cte = (CommonTableExpr *) lfirst(lc);
 
                if (strcmp(cte->ctename, rte->ctename) == 0)
                    break;
                ndx++;
            }
            if (lc == NULL)     /* shouldn't happen */
                elog(ERROR, "could not find CTE \"%s\"", rte->ctename);
            if (ndx >= list_length(cteroot->cte_plan_ids))
                elog(ERROR, "could not find plan for CTE \"%s\"", rte->ctename);
            plan_id = list_nth_int(cteroot->cte_plan_ids, ndx);
            if (plan_id <= 0)
                elog(ERROR, "no plan was made for CTE \"%s\"", rte->ctename);
            subroot = list_nth(root->glob->subroots, plan_id - 1);
        }
 
        /* If the subquery hasn't been planned yet, we have to punt */
        if (subroot == NULL)
            return;
        Assert(IsA(subroot, PlannerInfo));
 
        /*
         * We must use the subquery parsetree as mangled by the planner, not
         * the raw version from the RTE, because we need a Var that will refer
         * to the subroot's live RelOptInfos.  For instance, if any subquery
         * pullup happened during planning, Vars in the targetlist might have
         * gotten replaced, and we need to see the replacement expressions.
         */
        subquery = subroot->parse;
        Assert(IsA(subquery, Query));
 
        /*
         * Punt if subquery uses set operations or GROUP BY, as these will
         * mash underlying columns' stats beyond recognition.  (Set ops are
         * particularly nasty; if we forged ahead, we would return stats
         * relevant to only the leftmost subselect...)  DISTINCT is also
         * problematic, but we check that later because there is a possibility
         * of learning something even with it.
         */
        if (subquery->setOperations ||
            subquery->groupClause ||
            subquery->groupingSets)
            return;
 
        /* Get the subquery output expression referenced by the upper Var */
        if (subquery->returningList)
            subtlist = subquery->returningList;
        else
            subtlist = subquery->targetList;
        ste = get_tle_by_resno(subtlist, var->varattno);
        if (ste == NULL || ste->resjunk)
            elog(ERROR, "subquery %s does not have attribute %d",
                 rte->eref->aliasname, var->varattno);
        var = (Var *) ste->expr;
 
        /*
         * If subquery uses DISTINCT, we can't make use of any stats for the
         * variable ... but, if it's the only DISTINCT column, we are entitled
         * to consider it unique.  We do the test this way so that it works
         * for cases involving DISTINCT ON.
         */
        if (subquery->distinctClause)
        {
            if (list_length(subquery->distinctClause) == 1 &&
                targetIsInSortList(ste, InvalidOid, subquery->distinctClause))
                vardata->isunique = true;
            /* cannot go further */
            return;
        }
 
        /*
         * If the sub-query originated from a view with the security_barrier
         * attribute, we must not look at the variable's statistics, though it
         * seems all right to notice the existence of a DISTINCT clause. So
         * stop here.
         *
         * This is probably a harsher restriction than necessary; it's
         * certainly OK for the selectivity estimator (which is a C function,
         * and therefore omnipotent anyway) to look at the statistics.  But
         * many selectivity estimators will happily *invoke the operator
         * function* to try to work out a good estimate - and that's not OK.
         * So for now, don't dig down for stats.
         */
        if (rte->security_barrier)
            return;
 
        /* Can only handle a simple Var of subquery's query level */
        if (var && IsA(var, Var) &&
            var->varlevelsup == 0)
        {
            /*
             * OK, recurse into the subquery.  Note that the original setting
             * of vardata->isunique (which will surely be false) is left
             * unchanged in this situation.  That's what we want, since even
             * if the underlying column is unique, the subquery may have
             * joined to other tables in a way that creates duplicates.
             */
            examine_simple_variable(subroot, var, vardata);
        }
    }
    else
    {
        /*
         * Otherwise, the Var comes from a FUNCTION or VALUES RTE.  (We won't
         * see RTE_JOIN here because join alias Vars have already been
         * flattened.)  There's not much we can do with function outputs, but
         * maybe someday try to be smarter about VALUES.
         */
    }
}

References VariableStatData::acl_ok, ACL_SELECT, ACLCHECK_OK, Assert, BoolGetDatum(), RTEPermissionInfo::checkAsUser, PlannerInfo::cte_plan_ids, RangeTblEntry::ctelevelsup, Query::cteList, RangeTblEntry::ctename, CommonTableExpr::ctename, Query::distinctClause, elog, ERROR, TargetEntry::expr, find_base_rel(), find_base_rel_noerr(), VariableStatData::freefunc, get_relation_stats_hook, get_tle_by_resno(), getRTEPermissionInfo(), GetUserId(), Query::groupClause, Query::groupingSets, HeapTupleIsValid, if(), RangeTblEntry::inh, Int16GetDatum(), InvalidAttrNumber, InvalidOid, IsA, VariableStatData::isunique, lfirst, list_length(), list_nth(), list_nth_int(), NIL, AppendRelInfo::num_child_cols, ObjectIdGetDatum(), OidIsValid, AppendRelInfo::parent_relid, PlannerInfo::parse, pg_attribute_aclcheck(), pg_class_aclcheck(), planner_rt_fetch, ReleaseSysCache(), RangeTblEntry::relid, Query::returningList, root, RTE_CTE, RTE_RELATION, RTE_SUBQUERY, RangeTblEntry::rtekind, SearchSysCache3(), Query::setOperations, VariableStatData::statsTuple, RelOptInfo::subroot, targetIsInSortList(), Query::targetList, RelOptInfo::userid, Var::varattno, Var::varlevelsup, and Var::varno.

Referenced by examine_variable().

examine_variable()

void examine_variable	(	PlannerInfo *	root,
		Node *	node,
		int	varRelid,
		VariableStatData *	vardata
	)

Definition at line 5020 of file selfuncs.c.

{
    Node       *basenode;
    Relids      varnos;
    RelOptInfo *onerel;
 
    /* Make sure we don't return dangling pointers in vardata */
    MemSet(vardata, 0, sizeof(VariableStatData));
 
    /* Save the exposed type of the expression */
    vardata->vartype = exprType(node);
 
    /* Look inside any binary-compatible relabeling */
 
    if (IsA(node, RelabelType))
        basenode = (Node *) ((RelabelType *) node)->arg;
    else
        basenode = node;
 
    /* Fast path for a simple Var */
 
    if (IsA(basenode, Var) &&
        (varRelid == 0 || varRelid == ((Var *) basenode)->varno))
    {
        Var        *var = (Var *) basenode;
 
        /* Set up result fields other than the stats tuple */
        vardata->var = basenode;    /* return Var without relabeling */
        vardata->rel = find_base_rel(root, var->varno);
        vardata->atttype = var->vartype;
        vardata->atttypmod = var->vartypmod;
        vardata->isunique = has_unique_index(vardata->rel, var->varattno);
 
        /* Try to locate some stats */
        examine_simple_variable(root, var, vardata);
 
        return;
    }
 
    /*
     * Okay, it's a more complicated expression.  Determine variable
     * membership.  Note that when varRelid isn't zero, only vars of that
     * relation are considered "real" vars.
     */
    varnos = pull_varnos(root, basenode);
 
    onerel = NULL;
 
    if (bms_is_empty(varnos))
    {
        /* No Vars at all ... must be pseudo-constant clause */
    }
    else
    {
        int         relid;
 
        if (bms_get_singleton_member(varnos, &relid))
        {
            if (varRelid == 0 || varRelid == relid)
            {
                onerel = find_base_rel(root, relid);
                vardata->rel = onerel;
                node = basenode;    /* strip any relabeling */
            }
            /* else treat it as a constant */
        }
        else
        {
            /* varnos has multiple relids */
            if (varRelid == 0)
            {
                /* treat it as a variable of a join relation */
                vardata->rel = find_join_rel(root, varnos);
                node = basenode;    /* strip any relabeling */
            }
            else if (bms_is_member(varRelid, varnos))
            {
                /* ignore the vars belonging to other relations */
                vardata->rel = find_base_rel(root, varRelid);
                node = basenode;    /* strip any relabeling */
                /* note: no point in expressional-index search here */
            }
            /* else treat it as a constant */
        }
    }
 
    bms_free(varnos);
 
    vardata->var = node;
    vardata->atttype = exprType(node);
    vardata->atttypmod = exprTypmod(node);
 
    if (onerel)
    {
        /*
         * We have an expression in vars of a single relation.  Try to match
         * it to expressional index columns, in hopes of finding some
         * statistics.
         *
         * Note that we consider all index columns including INCLUDE columns,
         * since there could be stats for such columns.  But the test for
         * uniqueness needs to be warier.
         *
         * XXX it's conceivable that there are multiple matches with different
         * index opfamilies; if so, we need to pick one that matches the
         * operator we are estimating for.  FIXME later.
         */
        ListCell   *ilist;
        ListCell   *slist;
        Oid         userid;
 
        /*
         * Determine the user ID to use for privilege checks: either
         * onerel->userid if it's set (e.g., in case we're accessing the table
         * via a view), or the current user otherwise.
         *
         * If we drill down to child relations, we keep using the same userid:
         * it's going to be the same anyway, due to how we set up the relation
         * tree (q.v. build_simple_rel).
         */
        userid = OidIsValid(onerel->userid) ? onerel->userid : GetUserId();
 
        foreach(ilist, onerel->indexlist)
        {
            IndexOptInfo *index = (IndexOptInfo *) lfirst(ilist);
            ListCell   *indexpr_item;
            int         pos;
 
            indexpr_item = list_head(index->indexprs);
            if (indexpr_item == NULL)
                continue;       /* no expressions here... */
 
            for (pos = 0; pos < index->ncolumns; pos++)
            {
                if (index->indexkeys[pos] == 0)
                {
                    Node       *indexkey;
 
                    if (indexpr_item == NULL)
                        elog(ERROR, "too few entries in indexprs list");
                    indexkey = (Node *) lfirst(indexpr_item);
                    if (indexkey && IsA(indexkey, RelabelType))
                        indexkey = (Node *) ((RelabelType *) indexkey)->arg;
                    if (equal(node, indexkey))
                    {
                        /*
                         * Found a match ... is it a unique index? Tests here
                         * should match has_unique_index().
                         */
                        if (index->unique &&
                            index->nkeycolumns == 1 &&
                            pos == 0 &&
                            (index->indpred == NIL || index->predOK))
                            vardata->isunique = true;
 
                        /*
                         * Has it got stats?  We only consider stats for
                         * non-partial indexes, since partial indexes probably
                         * don't reflect whole-relation statistics; the above
                         * check for uniqueness is the only info we take from
                         * a partial index.
                         *
                         * An index stats hook, however, must make its own
                         * decisions about what to do with partial indexes.
                         */
                        if (get_index_stats_hook &&
                            (*get_index_stats_hook) (root, index->indexoid,
                                                     pos + 1, vardata))
                        {
                            /*
                             * The hook took control of acquiring a stats
                             * tuple.  If it did supply a tuple, it'd better
                             * have supplied a freefunc.
                             */
                            if (HeapTupleIsValid(vardata->statsTuple) &&
                                !vardata->freefunc)
                                elog(ERROR, "no function provided to release variable stats with");
                        }
                        else if (index->indpred == NIL)
                        {
                            vardata->statsTuple =
                                SearchSysCache3(STATRELATTINH,
                                                ObjectIdGetDatum(index->indexoid),
                                                Int16GetDatum(pos + 1),
                                                BoolGetDatum(false));
                            vardata->freefunc = ReleaseSysCache;
 
                            if (HeapTupleIsValid(vardata->statsTuple))
                            {
                                /* Get index's table for permission check */
                                RangeTblEntry *rte;
 
                                rte = planner_rt_fetch(index->rel->relid, root);
                                Assert(rte->rtekind == RTE_RELATION);
 
                                /*
                                 * For simplicity, we insist on the whole
                                 * table being selectable, rather than trying
                                 * to identify which column(s) the index
                                 * depends on.  Also require all rows to be
                                 * selectable --- there must be no
                                 * securityQuals from security barrier views
                                 * or RLS policies.
                                 */
                                vardata->acl_ok =
                                    rte->securityQuals == NIL &&
                                    (pg_class_aclcheck(rte->relid, userid,
                                                       ACL_SELECT) == ACLCHECK_OK);
 
                                /*
                                 * If the user doesn't have permissions to
                                 * access an inheritance child relation, check
                                 * the permissions of the table actually
                                 * mentioned in the query, since most likely
                                 * the user does have that permission.  Note
                                 * that whole-table select privilege on the
                                 * parent doesn't quite guarantee that the
                                 * user could read all columns of the child.
                                 * But in practice it's unlikely that any
                                 * interesting security violation could result
                                 * from allowing access to the expression
                                 * index's stats, so we allow it anyway.  See
                                 * similar code in examine_simple_variable()
                                 * for additional comments.
                                 */
                                if (!vardata->acl_ok &&
                                    root->append_rel_array != NULL)
                                {
                                    AppendRelInfo *appinfo;
                                    Index       varno = index->rel->relid;
 
                                    appinfo = root->append_rel_array[varno];
                                    while (appinfo &&
                                           planner_rt_fetch(appinfo->parent_relid,
                                                            root)->rtekind == RTE_RELATION)
                                    {
                                        varno = appinfo->parent_relid;
                                        appinfo = root->append_rel_array[varno];
                                    }
                                    if (varno != index->rel->relid)
                                    {
                                        /* Repeat access check on this rel */
                                        rte = planner_rt_fetch(varno, root);
                                        Assert(rte->rtekind == RTE_RELATION);
 
                                        vardata->acl_ok =
                                            rte->securityQuals == NIL &&
                                            (pg_class_aclcheck(rte->relid,
                                                               userid,
                                                               ACL_SELECT) == ACLCHECK_OK);
                                    }
                                }
                            }
                            else
                            {
                                /* suppress leakproofness checks later */
                                vardata->acl_ok = true;
                            }
                        }
                        if (vardata->statsTuple)
                            break;
                    }
                    indexpr_item = lnext(index->indexprs, indexpr_item);
                }
            }
            if (vardata->statsTuple)
                break;
        }
 
        /*
         * Search extended statistics for one with a matching expression.
         * There might be multiple ones, so just grab the first one. In the
         * future, we might consider the statistics target (and pick the most
         * accurate statistics) and maybe some other parameters.
         */
        foreach(slist, onerel->statlist)
        {
            StatisticExtInfo *info = (StatisticExtInfo *) lfirst(slist);
            RangeTblEntry *rte = planner_rt_fetch(onerel->relid, root);
            ListCell   *expr_item;
            int         pos;
 
            /*
             * Stop once we've found statistics for the expression (either
             * from extended stats, or for an index in the preceding loop).
             */
            if (vardata->statsTuple)
                break;
 
            /* skip stats without per-expression stats */
            if (info->kind != STATS_EXT_EXPRESSIONS)
                continue;
 
            /* skip stats with mismatching stxdinherit value */
            if (info->inherit != rte->inh)
                continue;
 
            pos = 0;
            foreach(expr_item, info->exprs)
            {
                Node       *expr = (Node *) lfirst(expr_item);
 
                Assert(expr);
 
                /* strip RelabelType before comparing it */
                if (expr && IsA(expr, RelabelType))
                    expr = (Node *) ((RelabelType *) expr)->arg;
 
                /* found a match, see if we can extract pg_statistic row */
                if (equal(node, expr))
                {
                    /*
                     * XXX Not sure if we should cache the tuple somewhere.
                     * Now we just create a new copy every time.
                     */
                    vardata->statsTuple =
                        statext_expressions_load(info->statOid, rte->inh, pos);
 
                    vardata->freefunc = ReleaseDummy;
 
                    /*
                     * For simplicity, we insist on the whole table being
                     * selectable, rather than trying to identify which
                     * column(s) the statistics object depends on.  Also
                     * require all rows to be selectable --- there must be no
                     * securityQuals from security barrier views or RLS
                     * policies.
                     */
                    vardata->acl_ok =
                        rte->securityQuals == NIL &&
                        (pg_class_aclcheck(rte->relid, userid,
                                           ACL_SELECT) == ACLCHECK_OK);
 
                    /*
                     * If the user doesn't have permissions to access an
                     * inheritance child relation, check the permissions of
                     * the table actually mentioned in the query, since most
                     * likely the user does have that permission.  Note that
                     * whole-table select privilege on the parent doesn't
                     * quite guarantee that the user could read all columns of
                     * the child. But in practice it's unlikely that any
                     * interesting security violation could result from
                     * allowing access to the expression stats, so we allow it
                     * anyway.  See similar code in examine_simple_variable()
                     * for additional comments.
                     */
                    if (!vardata->acl_ok &&
                        root->append_rel_array != NULL)
                    {
                        AppendRelInfo *appinfo;
                        Index       varno = onerel->relid;
 
                        appinfo = root->append_rel_array[varno];
                        while (appinfo &&
                               planner_rt_fetch(appinfo->parent_relid,
                                                root)->rtekind == RTE_RELATION)
                        {
                            varno = appinfo->parent_relid;
                            appinfo = root->append_rel_array[varno];
                        }
                        if (varno != onerel->relid)
                        {
                            /* Repeat access check on this rel */
                            rte = planner_rt_fetch(varno, root);
                            Assert(rte->rtekind == RTE_RELATION);
 
                            vardata->acl_ok =
                                rte->securityQuals == NIL &&
                                (pg_class_aclcheck(rte->relid,
                                                   userid,
                                                   ACL_SELECT) == ACLCHECK_OK);
                        }
                    }
 
                    break;
                }
 
                pos++;
            }
        }
    }
}

References VariableStatData::acl_ok, ACL_SELECT, ACLCHECK_OK, arg, Assert, VariableStatData::atttype, VariableStatData::atttypmod, bms_free(), bms_get_singleton_member(), bms_is_empty, bms_is_member(), BoolGetDatum(), elog, equal(), ERROR, examine_simple_variable(), StatisticExtInfo::exprs, exprType(), exprTypmod(), find_base_rel(), find_join_rel(), VariableStatData::freefunc, get_index_stats_hook, GetUserId(), has_unique_index(), HeapTupleIsValid, if(), RelOptInfo::indexlist, RangeTblEntry::inh, StatisticExtInfo::inherit, Int16GetDatum(), IsA, VariableStatData::isunique, StatisticExtInfo::kind, lfirst, list_head(), lnext(), MemSet, NIL, ObjectIdGetDatum(), OidIsValid, AppendRelInfo::parent_relid, pg_class_aclcheck(), planner_rt_fetch, pull_varnos(), VariableStatData::rel, ReleaseDummy(), ReleaseSysCache(), RangeTblEntry::relid, RelOptInfo::relid, root, RTE_RELATION, RangeTblEntry::rtekind, SearchSysCache3(), statext_expressions_load(), RelOptInfo::statlist, StatisticExtInfo::statOid, VariableStatData::statsTuple, RelOptInfo::userid, VariableStatData::var, Var::varattno, Var::varno, and VariableStatData::vartype.

Referenced by booltestsel(), boolvarsel(), estimate_array_length(), estimate_hash_bucket_stats(), estimate_num_groups(), get_join_variables(), get_restriction_variable(), mergejoinscansel(), nulltestsel(), and scalararraysel_containment().

find_join_input_rel()

static RelOptInfo * find_join_input_rel ( PlannerInfo *  root, Relids  relids  )

static

Definition at line 6436 of file selfuncs.c.

{
    RelOptInfo *rel = NULL;
 
    if (!bms_is_empty(relids))
    {
        int         relid;
 
        if (bms_get_singleton_member(relids, &relid))
            rel = find_base_rel(root, relid);
        else
            rel = find_join_rel(root, relids);
    }
 
    if (rel == NULL)
        elog(ERROR, "could not find RelOptInfo for given relids");
 
    return rel;
}

References bms_get_singleton_member(), bms_is_empty, elog, ERROR, find_base_rel(), find_join_rel(), and root.

Referenced by eqjoinsel().

generic_restriction_selectivity()

double generic_restriction_selectivity	(	PlannerInfo *	root,
		Oid	oproid,
		Oid	collation,
		List *	args,
		int	varRelid,
		double	default_selectivity
	)

Definition at line 914 of file selfuncs.c.

{
    double      selec;
    VariableStatData vardata;
    Node       *other;
    bool        varonleft;
 
    /*
     * If expression is not variable OP something or something OP variable,
     * then punt and return the default estimate.
     */
    if (!get_restriction_variable(root, args, varRelid,
                                  &vardata, &other, &varonleft))
        return default_selectivity;
 
    /*
     * If the something is a NULL constant, assume operator is strict and
     * return zero, ie, operator will never return TRUE.
     */
    if (IsA(other, Const) &&
        ((Const *) other)->constisnull)
    {
        ReleaseVariableStats(vardata);
        return 0.0;
    }
 
    if (IsA(other, Const))
    {
        /* Variable is being compared to a known non-null constant */
        Datum       constval = ((Const *) other)->constvalue;
        FmgrInfo    opproc;
        double      mcvsum;
        double      mcvsel;
        double      nullfrac;
        int         hist_size;
 
        fmgr_info(get_opcode(oproid), &opproc);
 
        /*
         * Calculate the selectivity for the column's most common values.
         */
        mcvsel = mcv_selectivity(&vardata, &opproc, collation,
                                 constval, varonleft,
                                 &mcvsum);
 
        /*
         * If the histogram is large enough, see what fraction of it matches
         * the query, and assume that's representative of the non-MCV
         * population.  Otherwise use the default selectivity for the non-MCV
         * population.
         */
        selec = histogram_selectivity(&vardata, &opproc, collation,
                                      constval, varonleft,
                                      10, 1, &hist_size);
        if (selec < 0)
        {
            /* Nope, fall back on default */
            selec = default_selectivity;
        }
        else if (hist_size < 100)
        {
            /*
             * For histogram sizes from 10 to 100, we combine the histogram
             * and default selectivities, putting increasingly more trust in
             * the histogram for larger sizes.
             */
            double      hist_weight = hist_size / 100.0;
 
            selec = selec * hist_weight +
                default_selectivity * (1.0 - hist_weight);
        }
 
        /* In any case, don't believe extremely small or large estimates. */
        if (selec < 0.0001)
            selec = 0.0001;
        else if (selec > 0.9999)
            selec = 0.9999;
 
        /* Don't forget to account for nulls. */
        if (HeapTupleIsValid(vardata.statsTuple))
            nullfrac = ((Form_pg_statistic) GETSTRUCT(vardata.statsTuple))->stanullfrac;
        else
            nullfrac = 0.0;
 
        /*
         * 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 - nullfrac - mcvsum;
        selec += mcvsel;
    }
    else
    {
        /* Comparison value is not constant, so we can't do anything */
        selec = default_selectivity;
    }
 
    ReleaseVariableStats(vardata);
 
    /* result should be in range, but make sure... */
    CLAMP_PROBABILITY(selec);
 
    return selec;
}

References generate_unaccent_rules::args, CLAMP_PROBABILITY, fmgr_info(), get_opcode(), get_restriction_variable(), GETSTRUCT, HeapTupleIsValid, histogram_selectivity(), IsA, mcv_selectivity(), ReleaseVariableStats, root, and VariableStatData::statsTuple.

Referenced by ltreeparentsel(), and matchingsel().

genericcostestimate()

void genericcostestimate	(	PlannerInfo *	root,
		IndexPath *	path,
		double	loop_count,
		GenericCosts *	costs
	)

Definition at line 6552 of file selfuncs.c.

{
    IndexOptInfo *index = path->indexinfo;
    List       *indexQuals = get_quals_from_indexclauses(path->indexclauses);
    List       *indexOrderBys = path->indexorderbys;
    Cost        indexStartupCost;
    Cost        indexTotalCost;
    Selectivity indexSelectivity;
    double      indexCorrelation;
    double      numIndexPages;
    double      numIndexTuples;
    double      spc_random_page_cost;
    double      num_sa_scans;
    double      num_outer_scans;
    double      num_scans;
    double      qual_op_cost;
    double      qual_arg_cost;
    List       *selectivityQuals;
    ListCell   *l;
 
    /*
     * If the index is partial, AND the index predicate with the explicitly
     * given indexquals to produce a more accurate idea of the index
     * selectivity.
     */
    selectivityQuals = add_predicate_to_index_quals(index, indexQuals);
 
    /*
     * If caller didn't give us an estimate for ScalarArrayOpExpr index scans,
     * just assume that the number of index descents is the number of distinct
     * combinations of array elements from all of the scan's SAOP clauses.
     */
    num_sa_scans = costs->num_sa_scans;
    if (num_sa_scans < 1)
    {
        num_sa_scans = 1;
        foreach(l, indexQuals)
        {
            RestrictInfo *rinfo = (RestrictInfo *) lfirst(l);
 
            if (IsA(rinfo->clause, ScalarArrayOpExpr))
            {
                ScalarArrayOpExpr *saop = (ScalarArrayOpExpr *) rinfo->clause;
                double      alength = estimate_array_length(root, lsecond(saop->args));
 
                if (alength > 1)
                    num_sa_scans *= alength;
            }
        }
    }
 
    /* Estimate the fraction of main-table tuples that will be visited */
    indexSelectivity = clauselist_selectivity(root, selectivityQuals,
                                              index->rel->relid,
                                              JOIN_INNER,
                                              NULL);
 
    /*
     * If caller didn't give us an estimate, estimate the number of index
     * tuples that will be visited.  We do it in this rather peculiar-looking
     * way in order to get the right answer for partial indexes.
     */
    numIndexTuples = costs->numIndexTuples;
    if (numIndexTuples <= 0.0)
    {
        numIndexTuples = indexSelectivity * index->rel->tuples;
 
        /*
         * The above calculation counts all the tuples visited across all
         * scans induced by ScalarArrayOpExpr nodes.  We want to consider the
         * average per-indexscan number, so adjust.  This is a handy place to
         * round to integer, too.  (If caller supplied tuple estimate, it's
         * responsible for handling these considerations.)
         */
        numIndexTuples = rint(numIndexTuples / num_sa_scans);
    }
 
    /*
     * We can bound the number of tuples by the index size in any case. Also,
     * always estimate at least one tuple is touched, even when
     * indexSelectivity estimate is tiny.
     */
    if (numIndexTuples > index->tuples)
        numIndexTuples = index->tuples;
    if (numIndexTuples < 1.0)
        numIndexTuples = 1.0;
 
    /*
     * Estimate the number of index pages that will be retrieved.
     *
     * We use the simplistic method of taking a pro-rata fraction of the total
     * number of index pages.  In effect, this counts only leaf pages and not
     * any overhead such as index metapage or upper tree levels.
     *
     * In practice access to upper index levels is often nearly free because
     * those tend to stay in cache under load; moreover, the cost involved is
     * highly dependent on index type.  We therefore ignore such costs here
     * and leave it to the caller to add a suitable charge if needed.
     */
    if (index->pages > 1 && index->tuples > 1)
        numIndexPages = ceil(numIndexTuples * index->pages / index->tuples);
    else
        numIndexPages = 1.0;
 
    /* fetch estimated page cost for tablespace containing index */
    get_tablespace_page_costs(index->reltablespace,
                              &spc_random_page_cost,
                              NULL);
 
    /*
     * Now compute the disk access costs.
     *
     * The above calculations are all per-index-scan.  However, if we are in a
     * nestloop inner scan, we can expect the scan to be repeated (with
     * different search keys) for each row of the outer relation.  Likewise,
     * ScalarArrayOpExpr quals result in multiple index scans.  This creates
     * the potential for cache effects to reduce the number of disk page
     * fetches needed.  We want to estimate the average per-scan I/O cost in
     * the presence of caching.
     *
     * We use the Mackert-Lohman formula (see costsize.c for details) to
     * estimate the total number of page fetches that occur.  While this
     * wasn't what it was designed for, it seems a reasonable model anyway.
     * Note that we are counting pages not tuples anymore, so we take N = T =
     * index size, as if there were one "tuple" per page.
     */
    num_outer_scans = loop_count;
    num_scans = num_sa_scans * num_outer_scans;
 
    if (num_scans > 1)
    {
        double      pages_fetched;
 
        /* total page fetches ignoring cache effects */
        pages_fetched = numIndexPages * num_scans;
 
        /* use Mackert and Lohman formula to adjust for cache effects */
        pages_fetched = index_pages_fetched(pages_fetched,
                                            index->pages,
                                            (double) index->pages,
                                            root);
 
        /*
         * Now compute the total disk access cost, and then report a pro-rated
         * share for each outer scan.  (Don't pro-rate for ScalarArrayOpExpr,
         * since that's internal to the indexscan.)
         */
        indexTotalCost = (pages_fetched * spc_random_page_cost)
            / num_outer_scans;
    }
    else
    {
        /*
         * For a single index scan, we just charge spc_random_page_cost per
         * page touched.
         */
        indexTotalCost = numIndexPages * spc_random_page_cost;
    }
 
    /*
     * CPU cost: any complex expressions in the indexquals will need to be
     * evaluated once at the start of the scan to reduce them to runtime keys
     * to pass to the index AM (see nodeIndexscan.c).  We model the per-tuple
     * CPU costs as cpu_index_tuple_cost plus one cpu_operator_cost per
     * indexqual operator.  Because we have numIndexTuples as a per-scan
     * number, we have to multiply by num_sa_scans to get the correct result
     * for ScalarArrayOpExpr cases.  Similarly add in costs for any index
     * ORDER BY expressions.
     *
     * Note: this neglects the possible costs of rechecking lossy operators.
     * Detecting that that might be needed seems more expensive than it's
     * worth, though, considering all the other inaccuracies here ...
     */
    qual_arg_cost = index_other_operands_eval_cost(root, indexQuals) +
        index_other_operands_eval_cost(root, indexOrderBys);
    qual_op_cost = cpu_operator_cost *
        (list_length(indexQuals) + list_length(indexOrderBys));
 
    indexStartupCost = qual_arg_cost;
    indexTotalCost += qual_arg_cost;
    indexTotalCost += numIndexTuples * num_sa_scans * (cpu_index_tuple_cost + qual_op_cost);
 
    /*
     * Generic assumption about index correlation: there isn't any.
     */
    indexCorrelation = 0.0;
 
    /*
     * Return everything to caller.
     */
    costs->indexStartupCost = indexStartupCost;
    costs->indexTotalCost = indexTotalCost;
    costs->indexSelectivity = indexSelectivity;
    costs->indexCorrelation = indexCorrelation;
    costs->numIndexPages = numIndexPages;
    costs->numIndexTuples = numIndexTuples;
    costs->spc_random_page_cost = spc_random_page_cost;
    costs->num_sa_scans = num_sa_scans;
}

References add_predicate_to_index_quals(), ScalarArrayOpExpr::args, RestrictInfo::clause, clauselist_selectivity(), cpu_index_tuple_cost, cpu_operator_cost, estimate_array_length(), get_quals_from_indexclauses(), get_tablespace_page_costs(), index_other_operands_eval_cost(), index_pages_fetched(), IndexPath::indexclauses, GenericCosts::indexCorrelation, IndexPath::indexinfo, IndexPath::indexorderbys, GenericCosts::indexSelectivity, GenericCosts::indexStartupCost, GenericCosts::indexTotalCost, IsA, JOIN_INNER, lfirst, list_length(), lsecond, GenericCosts::num_sa_scans, GenericCosts::numIndexPages, GenericCosts::numIndexTuples, root, and GenericCosts::spc_random_page_cost.

Referenced by blcostestimate(), btcostestimate(), gistcostestimate(), hashcostestimate(), and spgcostestimate().

get_actual_variable_endpoint()

static bool get_actual_variable_endpoint ( Relation  heapRel, Relation  indexRel, ScanDirection  indexscandir, ScanKey  scankeys, int16  typLen, bool  typByVal, TupleTableSlot *  tableslot, MemoryContext  outercontext, Datum *  endpointDatum  )

static

Definition at line 6275 of file selfuncs.c.

{
    bool        have_data = false;
    SnapshotData SnapshotNonVacuumable;
    IndexScanDesc index_scan;
    Buffer      vmbuffer = InvalidBuffer;
    BlockNumber last_heap_block = InvalidBlockNumber;
    int         n_visited_heap_pages = 0;
    ItemPointer tid;
    Datum       values[INDEX_MAX_KEYS];
    bool        isnull[INDEX_MAX_KEYS];
    MemoryContext oldcontext;
 
    /*
     * We use the index-only-scan machinery for this.  With mostly-static
     * tables that's a win because it avoids a heap visit.  It's also a win
     * for dynamic data, but the reason is less obvious; read on for details.
     *
     * In principle, we should scan the index with our current active
     * snapshot, which is the best approximation we've got to what the query
     * will see when executed.  But that won't be exact if a new snap is taken
     * before running the query, and it can be very expensive if a lot of
     * recently-dead or uncommitted rows exist at the beginning or end of the
     * index (because we'll laboriously fetch each one and reject it).
     * Instead, we use SnapshotNonVacuumable.  That will accept recently-dead
     * and uncommitted rows as well as normal visible rows.  On the other
     * hand, it will reject known-dead rows, and thus not give a bogus answer
     * when the extreme value has been deleted (unless the deletion was quite
     * recent); that case motivates not using SnapshotAny here.
     *
     * A crucial point here is that SnapshotNonVacuumable, with
     * GlobalVisTestFor(heapRel) as horizon, yields the inverse of the
     * condition that the indexscan will use to decide that index entries are
     * killable (see heap_hot_search_buffer()).  Therefore, if the snapshot
     * rejects a tuple (or more precisely, all tuples of a HOT chain) and we
     * have to continue scanning past it, we know that the indexscan will mark
     * that index entry killed.  That means that the next
     * get_actual_variable_endpoint() call will not have to re-consider that
     * index entry.  In this way we avoid repetitive work when this function
     * is used a lot during planning.
     *
     * But using SnapshotNonVacuumable creates a hazard of its own.  In a
     * recently-created index, some index entries may point at "broken" HOT
     * chains in which not all the tuple versions contain data matching the
     * index entry.  The live tuple version(s) certainly do match the index,
     * but SnapshotNonVacuumable can accept recently-dead tuple versions that
     * don't match.  Hence, if we took data from the selected heap tuple, we
     * might get a bogus answer that's not close to the index extremal value,
     * or could even be NULL.  We avoid this hazard because we take the data
     * from the index entry not the heap.
     *
     * Despite all this care, there are situations where we might find many
     * non-visible tuples near the end of the index.  We don't want to expend
     * a huge amount of time here, so we give up once we've read too many heap
     * pages.  When we fail for that reason, the caller will end up using
     * whatever extremal value is recorded in pg_statistic.
     */
    InitNonVacuumableSnapshot(SnapshotNonVacuumable,
                              GlobalVisTestFor(heapRel));
 
    index_scan = index_beginscan(heapRel, indexRel,
                                 &SnapshotNonVacuumable,
                                 1, 0);
    /* Set it up for index-only scan */
    index_scan->xs_want_itup = true;
    index_rescan(index_scan, scankeys, 1, NULL, 0);
 
    /* Fetch first/next tuple in specified direction */
    while ((tid = index_getnext_tid(index_scan, indexscandir)) != NULL)
    {
        BlockNumber block = ItemPointerGetBlockNumber(tid);
 
        if (!VM_ALL_VISIBLE(heapRel,
                            block,
                            &vmbuffer))
        {
            /* Rats, we have to visit the heap to check visibility */
            if (!index_fetch_heap(index_scan, tableslot))
            {
                /*
                 * No visible tuple for this index entry, so we need to
                 * advance to the next entry.  Before doing so, count heap
                 * page fetches and give up if we've done too many.
                 *
                 * We don't charge a page fetch if this is the same heap page
                 * as the previous tuple.  This is on the conservative side,
                 * since other recently-accessed pages are probably still in
                 * buffers too; but it's good enough for this heuristic.
                 */
#define VISITED_PAGES_LIMIT 100
 
                if (block != last_heap_block)
                {
                    last_heap_block = block;
                    n_visited_heap_pages++;
                    if (n_visited_heap_pages > VISITED_PAGES_LIMIT)
                        break;
                }
 
                continue;       /* no visible tuple, try next index entry */
            }
 
            /* We don't actually need the heap tuple for anything */
            ExecClearTuple(tableslot);
 
            /*
             * We don't care whether there's more than one visible tuple in
             * the HOT chain; if any are visible, that's good enough.
             */
        }
 
        /*
         * We expect that btree will return data in IndexTuple not HeapTuple
         * format.  It's not lossy either.
         */
        if (!index_scan->xs_itup)
            elog(ERROR, "no data returned for index-only scan");
        if (index_scan->xs_recheck)
            elog(ERROR, "unexpected recheck indication from btree");
 
        /* OK to deconstruct the index tuple */
        index_deform_tuple(index_scan->xs_itup,
                           index_scan->xs_itupdesc,
                           values, isnull);
 
        /* Shouldn't have got a null, but be careful */
        if (isnull[0])
            elog(ERROR, "found unexpected null value in index \"%s\"",
                 RelationGetRelationName(indexRel));
 
        /* Copy the index column value out to caller's context */
        oldcontext = MemoryContextSwitchTo(outercontext);
        *endpointDatum = datumCopy(values[0], typByVal, typLen);
        MemoryContextSwitchTo(oldcontext);
        have_data = true;
        break;
    }
 
    if (vmbuffer != InvalidBuffer)
        ReleaseBuffer(vmbuffer);
    index_endscan(index_scan);
 
    return have_data;
}

References datumCopy(), elog, ERROR, ExecClearTuple(), GlobalVisTestFor(), index_beginscan(), index_deform_tuple(), index_endscan(), index_fetch_heap(), index_getnext_tid(), INDEX_MAX_KEYS, index_rescan(), InitNonVacuumableSnapshot, InvalidBlockNumber, InvalidBuffer, ItemPointerGetBlockNumber(), MemoryContextSwitchTo(), RelationGetRelationName, ReleaseBuffer(), values, VISITED_PAGES_LIMIT, VM_ALL_VISIBLE, IndexScanDescData::xs_itup, IndexScanDescData::xs_itupdesc, IndexScanDescData::xs_recheck, and IndexScanDescData::xs_want_itup.

Referenced by get_actual_variable_range().

get_actual_variable_range()

static bool get_actual_variable_range ( PlannerInfo *  root, VariableStatData *  vardata, Oid  sortop, Oid  collation, Datum *  min, Datum *  max  )

static

Definition at line 6095 of file selfuncs.c.

{
    bool        have_data = false;
    RelOptInfo *rel = vardata->rel;
    RangeTblEntry *rte;
    ListCell   *lc;
 
    /* No hope if no relation or it doesn't have indexes */
    if (rel == NULL || rel->indexlist == NIL)
        return false;
    /* If it has indexes it must be a plain relation */
    rte = root->simple_rte_array[rel->relid];
    Assert(rte->rtekind == RTE_RELATION);
 
    /* ignore partitioned tables.  Any indexes here are not real indexes */
    if (rte->relkind == RELKIND_PARTITIONED_TABLE)
        return false;
 
    /* Search through the indexes to see if any match our problem */
    foreach(lc, rel->indexlist)
    {
        IndexOptInfo *index = (IndexOptInfo *) lfirst(lc);
        ScanDirection indexscandir;
 
        /* Ignore non-btree indexes */
        if (index->relam != BTREE_AM_OID)
            continue;
 
        /*
         * Ignore partial indexes --- we only want stats that cover the entire
         * relation.
         */
        if (index->indpred != NIL)
            continue;
 
        /*
         * The index list might include hypothetical indexes inserted by a
         * get_relation_info hook --- don't try to access them.
         */
        if (index->hypothetical)
            continue;
 
        /*
         * The first index column must match the desired variable, sortop, and
         * collation --- but we can use a descending-order index.
         */
        if (collation != index->indexcollations[0])
            continue;           /* test first 'cause it's cheapest */
        if (!match_index_to_operand(vardata->var, 0, index))
            continue;
        switch (get_op_opfamily_strategy(sortop, index->sortopfamily[0]))
        {
            case BTLessStrategyNumber:
                if (index->reverse_sort[0])
                    indexscandir = BackwardScanDirection;
                else
                    indexscandir = ForwardScanDirection;
                break;
            case BTGreaterStrategyNumber:
                if (index->reverse_sort[0])
                    indexscandir = ForwardScanDirection;
                else
                    indexscandir = BackwardScanDirection;
                break;
            default:
                /* index doesn't match the sortop */
                continue;
        }
 
        /*
         * Found a suitable index to extract data from.  Set up some data that
         * can be used by both invocations of get_actual_variable_endpoint.
         */
        {
            MemoryContext tmpcontext;
            MemoryContext oldcontext;
            Relation    heapRel;
            Relation    indexRel;
            TupleTableSlot *slot;
            int16       typLen;
            bool        typByVal;
            ScanKeyData scankeys[1];
 
            /* Make sure any cruft gets recycled when we're done */
            tmpcontext = AllocSetContextCreate(CurrentMemoryContext,
                                               "get_actual_variable_range workspace",
                                               ALLOCSET_DEFAULT_SIZES);
            oldcontext = MemoryContextSwitchTo(tmpcontext);
 
            /*
             * Open the table and index so we can read from them.  We should
             * already have some type of lock on each.
             */
            heapRel = table_open(rte->relid, NoLock);
            indexRel = index_open(index->indexoid, NoLock);
 
            /* build some stuff needed for indexscan execution */
            slot = table_slot_create(heapRel, NULL);
            get_typlenbyval(vardata->atttype, &typLen, &typByVal);
 
            /* set up an IS NOT NULL scan key so that we ignore nulls */
            ScanKeyEntryInitialize(&scankeys[0],
                                   SK_ISNULL | SK_SEARCHNOTNULL,
                                   1,   /* index col to scan */
                                   InvalidStrategy, /* no strategy */
                                   InvalidOid,  /* no strategy subtype */
                                   InvalidOid,  /* no collation */
                                   InvalidOid,  /* no reg proc for this */
                                   (Datum) 0);  /* constant */
 
            /* If min is requested ... */
            if (min)
            {
                have_data = get_actual_variable_endpoint(heapRel,
                                                         indexRel,
                                                         indexscandir,
                                                         scankeys,
                                                         typLen,
                                                         typByVal,
                                                         slot,
                                                         oldcontext,
                                                         min);
            }
            else
            {
                /* If min not requested, still want to fetch max */
                have_data = true;
            }
 
            /* If max is requested, and we didn't already fail ... */
            if (max && have_data)
            {
                /* scan in the opposite direction; all else is the same */
                have_data = get_actual_variable_endpoint(heapRel,
                                                         indexRel,
                                                         -indexscandir,
                                                         scankeys,
                                                         typLen,
                                                         typByVal,
                                                         slot,
                                                         oldcontext,
                                                         max);
            }
 
            /* Clean everything up */
            ExecDropSingleTupleTableSlot(slot);
 
            index_close(indexRel, NoLock);
            table_close(heapRel, NoLock);
 
            MemoryContextSwitchTo(oldcontext);
            MemoryContextDelete(tmpcontext);
 
            /* And we're done */
            break;
        }
    }
 
    return have_data;
}

References ALLOCSET_DEFAULT_SIZES, AllocSetContextCreate, Assert, VariableStatData::atttype, BackwardScanDirection, BTGreaterStrategyNumber, BTLessStrategyNumber, CurrentMemoryContext, ExecDropSingleTupleTableSlot(), ForwardScanDirection, get_actual_variable_endpoint(), get_op_opfamily_strategy(), get_typlenbyval(), index_close(), index_open(), RelOptInfo::indexlist, InvalidOid, InvalidStrategy, lfirst, match_index_to_operand(), MemoryContextDelete(), MemoryContextSwitchTo(), NIL, NoLock, VariableStatData::rel, RangeTblEntry::relid, RelOptInfo::relid, root, RTE_RELATION, RangeTblEntry::rtekind, ScanKeyEntryInitialize(), SK_ISNULL, SK_SEARCHNOTNULL, table_close(), table_open(), table_slot_create(), and VariableStatData::var.

Referenced by get_variable_range(), and ineq_histogram_selectivity().

get_join_variables()

void get_join_variables	(	PlannerInfo *	root,
		List *	args,
		SpecialJoinInfo *	sjinfo,
		VariableStatData *	vardata1,
		VariableStatData *	vardata2,
		bool *	join_is_reversed
	)

Definition at line 4951 of file selfuncs.c.

{
    Node       *left,
               *right;
 
    if (list_length(args) != 2)
        elog(ERROR, "join operator should take two arguments");
 
    left = (Node *) linitial(args);
    right = (Node *) lsecond(args);
 
    examine_variable(root, left, 0, vardata1);
    examine_variable(root, right, 0, vardata2);
 
    if (vardata1->rel &&
        bms_is_subset(vardata1->rel->relids, sjinfo->syn_righthand))
        *join_is_reversed = true;   /* var1 is on RHS */
    else if (vardata2->rel &&
             bms_is_subset(vardata2->rel->relids, sjinfo->syn_lefthand))
        *join_is_reversed = true;   /* var2 is on LHS */
    else
        *join_is_reversed = false;
}

References generate_unaccent_rules::args, bms_is_subset(), elog, ERROR, examine_variable(), linitial, list_length(), lsecond, VariableStatData::rel, RelOptInfo::relids, root, SpecialJoinInfo::syn_lefthand, and SpecialJoinInfo::syn_righthand.

Referenced by eqjoinsel(), neqjoinsel(), and networkjoinsel().

get_quals_from_indexclauses()

List* get_quals_from_indexclauses

(

List *

indexclauses

)

Definition at line 6468 of file selfuncs.c.

{
    List       *result = NIL;
    ListCell   *lc;
 
    foreach(lc, indexclauses)
    {
        IndexClause *iclause = lfirst_node(IndexClause, lc);
        ListCell   *lc2;
 
        foreach(lc2, iclause->indexquals)
        {
            RestrictInfo *rinfo = lfirst_node(RestrictInfo, lc2);
 
            result = lappend(result, rinfo);
        }
    }
    return result;
}

References IndexClause::indexquals, lappend(), lfirst_node, and NIL.

Referenced by brincostestimate(), genericcostestimate(), and gincostestimate().

get_restriction_variable()

bool get_restriction_variable	(	PlannerInfo *	root,
		List *	args,
		int	varRelid,
		VariableStatData *	vardata,
		Node **	other,
		bool *	varonleft
	)

Definition at line 4891 of file selfuncs.c.

{
    Node       *left,
               *right;
    VariableStatData rdata;
 
    /* Fail if not a binary opclause (probably shouldn't happen) */
    if (list_length(args) != 2)
        return false;
 
    left = (Node *) linitial(args);
    right = (Node *) lsecond(args);
 
    /*
     * Examine both sides.  Note that when varRelid is nonzero, Vars of other
     * relations will be treated as pseudoconstants.
     */
    examine_variable(root, left, varRelid, vardata);
    examine_variable(root, right, varRelid, &rdata);
 
    /*
     * If one side is a variable and the other not, we win.
     */
    if (vardata->rel && rdata.rel == NULL)
    {
        *varonleft = true;
        *other = estimate_expression_value(root, rdata.var);
        /* Assume we need no ReleaseVariableStats(rdata) here */
        return true;
    }
 
    if (vardata->rel == NULL && rdata.rel)
    {
        *varonleft = false;
        *other = estimate_expression_value(root, vardata->var);
        /* Assume we need no ReleaseVariableStats(*vardata) here */
        *vardata = rdata;
        return true;
    }
 
    /* Oops, clause has wrong structure (probably var op var) */
    ReleaseVariableStats(*vardata);
    ReleaseVariableStats(rdata);
 
    return false;
}

References generate_unaccent_rules::args, estimate_expression_value(), examine_variable(), linitial, list_length(), lsecond, VariableStatData::rel, ReleaseVariableStats, root, and VariableStatData::var.

Referenced by _int_matchsel(), arraycontsel(), eqsel_internal(), generic_restriction_selectivity(), multirangesel(), networksel(), patternsel_common(), rangesel(), scalarineqsel_wrapper(), and tsmatchsel().

get_stats_slot_range()

static void get_stats_slot_range ( AttStatsSlot *  sslot, Oid  opfuncoid, FmgrInfo *  opproc, Oid  collation, int16  typLen, bool  typByVal, Datum *  min, Datum *  max, bool *  p_have_data  )

static

Definition at line 6032 of file selfuncs.c.

{
    Datum       tmin = *min;
    Datum       tmax = *max;
    bool        have_data = *p_have_data;
    bool        found_tmin = false;
    bool        found_tmax = false;
 
    /* Look up the comparison function, if we didn't already do so */
    if (opproc->fn_oid != opfuncoid)
        fmgr_info(opfuncoid, opproc);
 
    /* Scan all the slot's values */
    for (int i = 0; i < sslot->nvalues; i++)
    {
        if (!have_data)
        {
            tmin = tmax = sslot->values[i];
            found_tmin = found_tmax = true;