rangetypes_typanalyze.c File Reference

#include "postgres.h"
#include "catalog/pg_operator.h"
#include "commands/vacuum.h"
#include "utils/float.h"
#include "utils/fmgrprotos.h"
#include "utils/lsyscache.h"
#include "utils/rangetypes.h"
#include "utils/multirangetypes.h"
#include "varatt.h"

Include dependency graph for rangetypes_typanalyze.c:

Go to the source code of this file.

Functions
static int	float8_qsort_cmp (const void *a1, const void *a2, void *arg)
static int	range_bound_qsort_cmp (const void *a1, const void *a2, void *arg)
static void	compute_range_stats (VacAttrStats *stats, AnalyzeAttrFetchFunc fetchfunc, int samplerows, double totalrows)
Datum	range_typanalyze (PG_FUNCTION_ARGS)
Datum	multirange_typanalyze (PG_FUNCTION_ARGS)

Function Documentation

compute_range_stats()

static void compute_range_stats ( VacAttrStats *  stats, AnalyzeAttrFetchFunc  fetchfunc, int  samplerows, double  totalrows  )

static

Definition at line 125 of file rangetypes_typanalyze.c.

{
    TypeCacheEntry *typcache = (TypeCacheEntry *) stats->extra_data;
    TypeCacheEntry *mltrng_typcache = NULL;
    bool        has_subdiff;
    int         null_cnt = 0;
    int         non_null_cnt = 0;
    int         non_empty_cnt = 0;
    int         empty_cnt = 0;
    int         range_no;
    int         slot_idx;
    int         num_bins = stats->attstattarget;
    int         num_hist;
    float8     *lengths;
    RangeBound *lowers,
               *uppers;
    double      total_width = 0;
 
    if (typcache->typtype == TYPTYPE_MULTIRANGE)
    {
        mltrng_typcache = typcache;
        typcache = typcache->rngtype;
    }
    else
        Assert(typcache->typtype == TYPTYPE_RANGE);
    has_subdiff = OidIsValid(typcache->rng_subdiff_finfo.fn_oid);
 
    /* Allocate memory to hold range bounds and lengths of the sample ranges. */
    lowers = (RangeBound *) palloc(sizeof(RangeBound) * samplerows);
    uppers = (RangeBound *) palloc(sizeof(RangeBound) * samplerows);
    lengths = (float8 *) palloc(sizeof(float8) * samplerows);
 
    /* Loop over the sample ranges. */
    for (range_no = 0; range_no < samplerows; range_no++)
    {
        Datum       value;
        bool        isnull,
                    empty;
        MultirangeType *multirange;
        RangeType  *range;
        RangeBound  lower,
                    upper;
        float8      length;
 
        vacuum_delay_point();
 
        value = fetchfunc(stats, range_no, &isnull);
        if (isnull)
        {
            /* range is null, just count that */
            null_cnt++;
            continue;
        }
 
        /*
         * XXX: should we ignore wide values, like std_typanalyze does, to
         * avoid bloating the statistics table?
         */
        total_width += VARSIZE_ANY(DatumGetPointer(value));
 
        /* Get range and deserialize it for further analysis. */
        if (mltrng_typcache != NULL)
        {
            /* Treat multiranges like a big range without gaps. */
            multirange = DatumGetMultirangeTypeP(value);
            if (!MultirangeIsEmpty(multirange))
            {
                RangeBound  tmp;
 
                multirange_get_bounds(typcache, multirange, 0,
                                      &lower, &tmp);
                multirange_get_bounds(typcache, multirange,
                                      multirange->rangeCount - 1,
                                      &tmp, &upper);
                empty = false;
            }
            else
            {
                empty = true;
            }
        }
        else
        {
            range = DatumGetRangeTypeP(value);
            range_deserialize(typcache, range, &lower, &upper, &empty);
        }
 
        if (!empty)
        {
            /* Remember bounds and length for further usage in histograms */
            lowers[non_empty_cnt] = lower;
            uppers[non_empty_cnt] = upper;
 
            if (lower.infinite || upper.infinite)
            {
                /* Length of any kind of an infinite range is infinite */
                length = get_float8_infinity();
            }
            else if (has_subdiff)
            {
                /*
                 * For an ordinary range, use subdiff function between upper
                 * and lower bound values.
                 */
                length = DatumGetFloat8(FunctionCall2Coll(&typcache->rng_subdiff_finfo,
                                                          typcache->rng_collation,
                                                          upper.val, lower.val));
            }
            else
            {
                /* Use default value of 1.0 if no subdiff is available. */
                length = 1.0;
            }
            lengths[non_empty_cnt] = length;
 
            non_empty_cnt++;
        }
        else
            empty_cnt++;
 
        non_null_cnt++;
    }
 
    slot_idx = 0;
 
    /* We can only compute real stats if we found some non-null values. */
    if (non_null_cnt > 0)
    {
        Datum      *bound_hist_values;
        Datum      *length_hist_values;
        int         pos,
                    posfrac,
                    delta,
                    deltafrac,
                    i;
        MemoryContext old_cxt;
        float4     *emptyfrac;
 
        stats->stats_valid = true;
        /* Do the simple null-frac and width stats */
        stats->stanullfrac = (double) null_cnt / (double) samplerows;
        stats->stawidth = total_width / (double) non_null_cnt;
 
        /* Estimate that non-null values are unique */
        stats->stadistinct = -1.0 * (1.0 - stats->stanullfrac);
 
        /* Must copy the target values into anl_context */
        old_cxt = MemoryContextSwitchTo(stats->anl_context);
 
        /*
         * Generate a bounds histogram slot entry if there are at least two
         * values.
         */
        if (non_empty_cnt >= 2)
        {
            /* Sort bound values */
            qsort_interruptible(lowers, non_empty_cnt, sizeof(RangeBound),
                                range_bound_qsort_cmp, typcache);
            qsort_interruptible(uppers, non_empty_cnt, sizeof(RangeBound),
                                range_bound_qsort_cmp, typcache);
 
            num_hist = non_empty_cnt;
            if (num_hist > num_bins)
                num_hist = num_bins + 1;
 
            bound_hist_values = (Datum *) palloc(num_hist * sizeof(Datum));
 
            /*
             * The object of this loop is to construct ranges from first and
             * last entries in lowers[] and uppers[] along with evenly-spaced
             * values in between. So the i'th value is a range of lowers[(i *
             * (nvals - 1)) / (num_hist - 1)] and uppers[(i * (nvals - 1)) /
             * (num_hist - 1)]. But computing that subscript directly risks
             * integer overflow when the stats target is more than a couple
             * thousand.  Instead we add (nvals - 1) / (num_hist - 1) to pos
             * at each step, tracking the integral and fractional parts of the
             * sum separately.
             */
            delta = (non_empty_cnt - 1) / (num_hist - 1);
            deltafrac = (non_empty_cnt - 1) % (num_hist - 1);
            pos = posfrac = 0;
 
            for (i = 0; i < num_hist; i++)
            {
                bound_hist_values[i] = PointerGetDatum(range_serialize(typcache,
                                                                       &lowers[pos],
                                                                       &uppers[pos],
                                                                       false,
                                                                       NULL));
                pos += delta;
                posfrac += deltafrac;
                if (posfrac >= (num_hist - 1))
                {
                    /* fractional part exceeds 1, carry to integer part */
                    pos++;
                    posfrac -= (num_hist - 1);
                }
            }
 
            stats->stakind[slot_idx] = STATISTIC_KIND_BOUNDS_HISTOGRAM;
            stats->stavalues[slot_idx] = bound_hist_values;
            stats->numvalues[slot_idx] = num_hist;
 
            /* Store ranges even if we're analyzing a multirange column */
            stats->statypid[slot_idx] = typcache->type_id;
            stats->statyplen[slot_idx] = typcache->typlen;
            stats->statypbyval[slot_idx] = typcache->typbyval;
            stats->statypalign[slot_idx] = typcache->typalign;
 
            slot_idx++;
        }
 
        /*
         * Generate a length histogram slot entry if there are at least two
         * values.
         */
        if (non_empty_cnt >= 2)
        {
            /*
             * Ascending sort of range lengths for further filling of
             * histogram
             */
            qsort_interruptible(lengths, non_empty_cnt, sizeof(float8),
                                float8_qsort_cmp, NULL);
 
            num_hist = non_empty_cnt;
            if (num_hist > num_bins)
                num_hist = num_bins + 1;
 
            length_hist_values = (Datum *) palloc(num_hist * sizeof(Datum));
 
            /*
             * The object of this loop is to copy the first and last lengths[]
             * entries along with evenly-spaced values in between. So the i'th
             * value is lengths[(i * (nvals - 1)) / (num_hist - 1)]. But
             * computing that subscript directly risks integer overflow when
             * the stats target is more than a couple thousand.  Instead we
             * add (nvals - 1) / (num_hist - 1) to pos at each step, tracking
             * the integral and fractional parts of the sum separately.
             */
            delta = (non_empty_cnt - 1) / (num_hist - 1);
            deltafrac = (non_empty_cnt - 1) % (num_hist - 1);
            pos = posfrac = 0;
 
            for (i = 0; i < num_hist; i++)
            {
                length_hist_values[i] = Float8GetDatum(lengths[pos]);
                pos += delta;
                posfrac += deltafrac;
                if (posfrac >= (num_hist - 1))
                {
                    /* fractional part exceeds 1, carry to integer part */
                    pos++;
                    posfrac -= (num_hist - 1);
                }
            }
        }
        else
        {
            /*
             * Even when we don't create the histogram, store an empty array
             * to mean "no histogram". We can't just leave stavalues NULL,
             * because get_attstatsslot() errors if you ask for stavalues, and
             * it's NULL. We'll still store the empty fraction in stanumbers.
             */
            length_hist_values = palloc(0);
            num_hist = 0;
        }
        stats->staop[slot_idx] = Float8LessOperator;
        stats->stacoll[slot_idx] = InvalidOid;
        stats->stavalues[slot_idx] = length_hist_values;
        stats->numvalues[slot_idx] = num_hist;
        stats->statypid[slot_idx] = FLOAT8OID;
        stats->statyplen[slot_idx] = sizeof(float8);
        stats->statypbyval[slot_idx] = FLOAT8PASSBYVAL;
        stats->statypalign[slot_idx] = 'd';
 
        /* Store the fraction of empty ranges */
        emptyfrac = (float4 *) palloc(sizeof(float4));
        *emptyfrac = ((double) empty_cnt) / ((double) non_null_cnt);
        stats->stanumbers[slot_idx] = emptyfrac;
        stats->numnumbers[slot_idx] = 1;
 
        stats->stakind[slot_idx] = STATISTIC_KIND_RANGE_LENGTH_HISTOGRAM;
        slot_idx++;
 
        MemoryContextSwitchTo(old_cxt);
    }
    else if (null_cnt > 0)
    {
        /* We found only nulls; assume the column is entirely null */
        stats->stats_valid = true;
        stats->stanullfrac = 1.0;
        stats->stawidth = 0;    /* "unknown" */
        stats->stadistinct = 0.0;   /* "unknown" */
    }
 
    /*
     * We don't need to bother cleaning up any of our temporary palloc's. The
     * hashtable should also go away, as it used a child memory context.
     */
}

References VacAttrStats::anl_context, Assert(), VacAttrStats::attstattarget, DatumGetFloat8(), DatumGetMultirangeTypeP(), DatumGetPointer(), DatumGetRangeTypeP(), VacAttrStats::extra_data, float8_qsort_cmp(), Float8GetDatum(), FLOAT8PASSBYVAL, FmgrInfo::fn_oid, FunctionCall2Coll(), get_float8_infinity(), i, if(), InvalidOid, lower(), MemoryContextSwitchTo(), multirange_get_bounds(), MultirangeIsEmpty, VacAttrStats::numnumbers, VacAttrStats::numvalues, OidIsValid, palloc(), PointerGetDatum(), qsort_interruptible(), range(), range_bound_qsort_cmp(), range_deserialize(), range_serialize(), MultirangeType::rangeCount, TypeCacheEntry::rng_collation, TypeCacheEntry::rng_subdiff_finfo, TypeCacheEntry::rngtype, VacAttrStats::stacoll, VacAttrStats::stadistinct, VacAttrStats::stakind, VacAttrStats::stanullfrac, VacAttrStats::stanumbers, VacAttrStats::staop, VacAttrStats::stats_valid, VacAttrStats::statypalign, VacAttrStats::statypbyval, VacAttrStats::statypid, VacAttrStats::statyplen, VacAttrStats::stavalues, VacAttrStats::stawidth, TypeCacheEntry::typalign, TypeCacheEntry::typbyval, TypeCacheEntry::type_id, TypeCacheEntry::typlen, TypeCacheEntry::typtype, upper(), vacuum_delay_point(), value, and VARSIZE_ANY.

Referenced by multirange_typanalyze(), and range_typanalyze().

float8_qsort_cmp()

static int float8_qsort_cmp ( const void *  a1, const void *  a2, void *  arg  )

static

Definition at line 95 of file rangetypes_typanalyze.c.

{
    const float8 *f1 = (const float8 *) a1;
    const float8 *f2 = (const float8 *) a2;
 
    if (*f1 < *f2)
        return -1;
    else if (*f1 == *f2)
        return 0;
    else
        return 1;
}

References a1, a2, f1, and f2.

Referenced by compute_range_stats().

multirange_typanalyze()

Datum multirange_typanalyze

(

PG_FUNCTION_ARGS

)

Definition at line 72 of file rangetypes_typanalyze.c.

{
    VacAttrStats *stats = (VacAttrStats *) PG_GETARG_POINTER(0);
    TypeCacheEntry *typcache;
 
    /* Get information about multirange type; note column might be a domain */
    typcache = multirange_get_typcache(fcinfo, getBaseType(stats->attrtypid));
 
    if (stats->attstattarget < 0)
        stats->attstattarget = default_statistics_target;
 
    stats->compute_stats = compute_range_stats;
    stats->extra_data = typcache;
    /* same as in std_typanalyze */
    stats->minrows = 300 * stats->attstattarget;
 
    PG_RETURN_BOOL(true);
}

References VacAttrStats::attrtypid, VacAttrStats::attstattarget, compute_range_stats(), VacAttrStats::compute_stats, default_statistics_target, VacAttrStats::extra_data, getBaseType(), VacAttrStats::minrows, multirange_get_typcache(), PG_GETARG_POINTER, and PG_RETURN_BOOL.

range_bound_qsort_cmp()

static int range_bound_qsort_cmp ( const void *  a1, const void *  a2, void *  arg  )

static

Definition at line 112 of file rangetypes_typanalyze.c.

{
    RangeBound *b1 = (RangeBound *) a1;
    RangeBound *b2 = (RangeBound *) a2;
    TypeCacheEntry *typcache = (TypeCacheEntry *) arg;
 
    return range_cmp_bounds(typcache, b1, b2);
}

References a1, a2, arg, and range_cmp_bounds().

Referenced by compute_range_stats().

range_typanalyze()

Datum range_typanalyze

(

PG_FUNCTION_ARGS

)

Definition at line 46 of file rangetypes_typanalyze.c.

{
    VacAttrStats *stats = (VacAttrStats *) PG_GETARG_POINTER(0);
    TypeCacheEntry *typcache;
 
    /* Get information about range type; note column might be a domain */
    typcache = range_get_typcache(fcinfo, getBaseType(stats->attrtypid));
 
    if (stats->attstattarget < 0)
        stats->attstattarget = default_statistics_target;
 
    stats->compute_stats = compute_range_stats;
    stats->extra_data = typcache;
    /* same as in std_typanalyze */
    stats->minrows = 300 * stats->attstattarget;
 
    PG_RETURN_BOOL(true);
}

References VacAttrStats::attrtypid, VacAttrStats::attstattarget, compute_range_stats(), VacAttrStats::compute_stats, default_statistics_target, VacAttrStats::extra_data, getBaseType(), VacAttrStats::minrows, PG_GETARG_POINTER, PG_RETURN_BOOL, and range_get_typcache().