ts_selfuncs.c

Go to the documentation of this file.

/*-------------------------------------------------------------------------
 *
 * ts_selfuncs.c
 *    Selectivity estimation functions for text search operators.
 *
 * Portions Copyright (c) 1996-2013, PostgreSQL Global Development Group
 *
 *
 * IDENTIFICATION
 *    src/backend/tsearch/ts_selfuncs.c
 *
 *-------------------------------------------------------------------------
 */
#include "postgres.h"

#include "access/htup_details.h"
#include "catalog/pg_statistic.h"
#include "catalog/pg_type.h"
#include "miscadmin.h"
#include "nodes/nodes.h"
#include "tsearch/ts_type.h"
#include "utils/lsyscache.h"
#include "utils/selfuncs.h"
#include "utils/syscache.h"


/*
 * The default text search selectivity is chosen to be small enough to
 * encourage indexscans for typical table densities.  See selfuncs.h and
 * DEFAULT_EQ_SEL for details.
 */
#define DEFAULT_TS_MATCH_SEL 0.005

/* lookup table type for binary searching through MCELEMs */
typedef struct
{
    text       *element;
    float4      frequency;
} TextFreq;

/* type of keys for bsearch'ing through an array of TextFreqs */
typedef struct
{
    char       *lexeme;
    int         length;
} LexemeKey;

static Selectivity tsquerysel(VariableStatData *vardata, Datum constval);
static Selectivity mcelem_tsquery_selec(TSQuery query,
                     Datum *mcelem, int nmcelem,
                     float4 *numbers, int nnumbers);
static Selectivity tsquery_opr_selec(QueryItem *item, char *operand,
                  TextFreq *lookup, int length, float4 minfreq);
static int  compare_lexeme_textfreq(const void *e1, const void *e2);

#define tsquery_opr_selec_no_stats(query) \
    tsquery_opr_selec(GETQUERY(query), GETOPERAND(query), NULL, 0, 0)


/*
 *  tsmatchsel -- Selectivity of "@@"
 *
 * restriction selectivity function for tsvector @@ tsquery and
 * tsquery @@ tsvector
 */
Datum
tsmatchsel(PG_FUNCTION_ARGS)
{
    PlannerInfo *root = (PlannerInfo *) PG_GETARG_POINTER(0);

#ifdef NOT_USED
    Oid         operator = PG_GETARG_OID(1);
#endif
    List       *args = (List *) PG_GETARG_POINTER(2);
    int         varRelid = PG_GETARG_INT32(3);
    VariableStatData vardata;
    Node       *other;
    bool        varonleft;
    Selectivity selec;

    /*
     * 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))
        PG_RETURN_FLOAT8(DEFAULT_TS_MATCH_SEL);

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

    /*
     * The "@@" operator is strict, so we can cope with NULL right away
     */
    if (((Const *) other)->constisnull)
    {
        ReleaseVariableStats(vardata);
        PG_RETURN_FLOAT8(0.0);
    }

    /*
     * OK, there's a Var and a Const we're dealing with here.  We need the
     * Const to be a TSQuery, else we can't do anything useful.  We have to
     * check this because the Var might be the TSQuery not the TSVector.
     */
    if (((Const *) other)->consttype == TSQUERYOID)
    {
        /* tsvector @@ tsquery or the other way around */
        Assert(vardata.vartype == TSVECTOROID);

        selec = tsquerysel(&vardata, ((Const *) other)->constvalue);
    }
    else
    {
        /* If we can't see the query structure, must punt */
        selec = DEFAULT_TS_MATCH_SEL;
    }

    ReleaseVariableStats(vardata);

    CLAMP_PROBABILITY(selec);

    PG_RETURN_FLOAT8((float8) selec);
}


/*
 *  tsmatchjoinsel -- join selectivity of "@@"
 *
 * join selectivity function for tsvector @@ tsquery and tsquery @@ tsvector
 */
Datum
tsmatchjoinsel(PG_FUNCTION_ARGS)
{
    /* for the moment we just punt */
    PG_RETURN_FLOAT8(DEFAULT_TS_MATCH_SEL);
}


/*
 * @@ selectivity for tsvector var vs tsquery constant
 */
static Selectivity
tsquerysel(VariableStatData *vardata, Datum constval)
{
    Selectivity selec;
    TSQuery     query;

    /* The caller made sure the const is a TSQuery, so get it now */
    query = DatumGetTSQuery(constval);

    /* Empty query matches nothing */
    if (query->size == 0)
        return (Selectivity) 0.0;

    if (HeapTupleIsValid(vardata->statsTuple))
    {
        Form_pg_statistic stats;
        Datum      *values;
        int         nvalues;
        float4     *numbers;
        int         nnumbers;

        stats = (Form_pg_statistic) GETSTRUCT(vardata->statsTuple);

        /* MCELEM will be an array of TEXT elements for a tsvector column */
        if (get_attstatsslot(vardata->statsTuple,
                             TEXTOID, -1,
                             STATISTIC_KIND_MCELEM, InvalidOid,
                             NULL,
                             &values, &nvalues,
                             &numbers, &nnumbers))
        {
            /*
             * There is a most-common-elements slot for the tsvector Var, so
             * use that.
             */
            selec = mcelem_tsquery_selec(query, values, nvalues,
                                         numbers, nnumbers);
            free_attstatsslot(TEXTOID, values, nvalues, numbers, nnumbers);
        }
        else
        {
            /* No most-common-elements info, so do without */
            selec = tsquery_opr_selec_no_stats(query);
        }

        /*
         * MCE stats count only non-null rows, so adjust for null rows.
         */
        selec *= (1.0 - stats->stanullfrac);
    }
    else
    {
        /* No stats at all, so do without */
        selec = tsquery_opr_selec_no_stats(query);
        /* we assume no nulls here, so no stanullfrac correction */
    }

    return selec;
}

/*
 * Extract data from the pg_statistic arrays into useful format.
 */
static Selectivity
mcelem_tsquery_selec(TSQuery query, Datum *mcelem, int nmcelem,
                     float4 *numbers, int nnumbers)
{
    float4      minfreq;
    TextFreq   *lookup;
    Selectivity selec;
    int         i;

    /*
     * There should be two more Numbers than Values, because the last two
     * cells are taken for minimal and maximal frequency.  Punt if not.
     *
     * (Note: the MCELEM statistics slot definition allows for a third extra
     * number containing the frequency of nulls, but we're not expecting that
     * to appear for a tsvector column.)
     */
    if (nnumbers != nmcelem + 2)
        return tsquery_opr_selec_no_stats(query);

    /*
     * Transpose the data into a single array so we can use bsearch().
     */
    lookup = (TextFreq *) palloc(sizeof(TextFreq) * nmcelem);
    for (i = 0; i < nmcelem; i++)
    {
        /*
         * The text Datums came from an array, so it cannot be compressed or
         * stored out-of-line -- it's safe to use VARSIZE_ANY*.
         */
        Assert(!VARATT_IS_COMPRESSED(mcelem[i]) && !VARATT_IS_EXTERNAL(mcelem[i]));
        lookup[i].element = (text *) DatumGetPointer(mcelem[i]);
        lookup[i].frequency = numbers[i];
    }

    /*
     * Grab the lowest frequency. compute_tsvector_stats() stored it for us in
     * the one before the last cell of the Numbers array. See ts_typanalyze.c
     */
    minfreq = numbers[nnumbers - 2];

    selec = tsquery_opr_selec(GETQUERY(query), GETOPERAND(query), lookup,
                              nmcelem, minfreq);

    pfree(lookup);

    return selec;
}

/*
 * Traverse the tsquery in preorder, calculating selectivity as:
 *
 *   selec(left_oper) * selec(right_oper) in AND nodes,
 *
 *   selec(left_oper) + selec(right_oper) -
 *      selec(left_oper) * selec(right_oper) in OR nodes,
 *
 *   1 - select(oper) in NOT nodes
 *
 *   histogram-based estimation in prefix VAL nodes
 *
 *   freq[val] in exact VAL nodes, if the value is in MCELEM
 *   min(freq[MCELEM]) / 2 in VAL nodes, if it is not
 *
 * The MCELEM array is already sorted (see ts_typanalyze.c), so we can use
 * binary search for determining freq[MCELEM].
 *
 * If we don't have stats for the tsvector, we still use this logic,
 * except we use default estimates for VAL nodes.  This case is signaled
 * by lookup == NULL.
 */
static Selectivity
tsquery_opr_selec(QueryItem *item, char *operand,
                  TextFreq *lookup, int length, float4 minfreq)
{
    Selectivity selec;

    /* since this function recurses, it could be driven to stack overflow */
    check_stack_depth();

    if (item->type == QI_VAL)
    {
        QueryOperand *oper = (QueryOperand *) item;
        LexemeKey   key;

        /*
         * Prepare the key for bsearch().
         */
        key.lexeme = operand + oper->distance;
        key.length = oper->length;

        if (oper->prefix)
        {
            /* Prefix match, ie the query item is lexeme:* */
            Selectivity matched,
                        allmces;
            int         i,
                        n_matched;

            /*
             * Our strategy is to scan through the MCELEM list and combine the
             * frequencies of the ones that match the prefix.  We then
             * extrapolate the fraction of matching MCELEMs to the remaining
             * rows, assuming that the MCELEMs are representative of the whole
             * lexeme population in this respect.  (Compare
             * histogram_selectivity().)  Note that these are most common
             * elements not most common values, so they're not mutually
             * exclusive.  We treat occurrences as independent events.
             *
             * This is only a good plan if we have a pretty fair number of
             * MCELEMs available; we set the threshold at 100.  If no stats or
             * insufficient stats, arbitrarily use DEFAULT_TS_MATCH_SEL*4.
             */
            if (lookup == NULL || length < 100)
                return (Selectivity) (DEFAULT_TS_MATCH_SEL * 4);

            matched = allmces = 0;
            n_matched = 0;
            for (i = 0; i < length; i++)
            {
                TextFreq   *t = lookup + i;
                int         tlen = VARSIZE_ANY_EXHDR(t->element);

                if (tlen >= key.length &&
                    strncmp(key.lexeme, VARDATA_ANY(t->element),
                            key.length) == 0)
                {
                    matched += t->frequency - matched * t->frequency;
                    n_matched++;
                }
                allmces += t->frequency - allmces * t->frequency;
            }

            /* Clamp to ensure sanity in the face of roundoff error */
            CLAMP_PROBABILITY(matched);
            CLAMP_PROBABILITY(allmces);

            selec = matched + (1.0 - allmces) * ((double) n_matched / length);

            /*
             * In any case, never believe that a prefix match has selectivity
             * less than we would assign for a non-MCELEM lexeme.  This
             * preserves the property that "word:*" should be estimated to
             * match at least as many rows as "word" would be.
             */
            selec = Max(Min(DEFAULT_TS_MATCH_SEL, minfreq / 2), selec);
        }
        else
        {
            /* Regular exact lexeme match */
            TextFreq   *searchres;

            /* If no stats for the variable, use DEFAULT_TS_MATCH_SEL */
            if (lookup == NULL)
                return (Selectivity) DEFAULT_TS_MATCH_SEL;

            searchres = (TextFreq *) bsearch(&key, lookup, length,
                                             sizeof(TextFreq),
                                             compare_lexeme_textfreq);

            if (searchres)
            {
                /*
                 * The element is in MCELEM.  Return precise selectivity (or
                 * at least as precise as ANALYZE could find out).
                 */
                selec = searchres->frequency;
            }
            else
            {
                /*
                 * The element is not in MCELEM.  Punt, but assume that the
                 * selectivity cannot be more than minfreq / 2.
                 */
                selec = Min(DEFAULT_TS_MATCH_SEL, minfreq / 2);
            }
        }
    }
    else
    {
        /* Current TSQuery node is an operator */
        Selectivity s1,
                    s2;

        switch (item->qoperator.oper)
        {
            case OP_NOT:
                selec = 1.0 - tsquery_opr_selec(item + 1, operand,
                                                lookup, length, minfreq);
                break;

            case OP_AND:
                s1 = tsquery_opr_selec(item + 1, operand,
                                       lookup, length, minfreq);
                s2 = tsquery_opr_selec(item + item->qoperator.left, operand,
                                       lookup, length, minfreq);
                selec = s1 * s2;
                break;

            case OP_OR:
                s1 = tsquery_opr_selec(item + 1, operand,
                                       lookup, length, minfreq);
                s2 = tsquery_opr_selec(item + item->qoperator.left, operand,
                                       lookup, length, minfreq);
                selec = s1 + s2 - s1 * s2;
                break;

            default:
                elog(ERROR, "unrecognized operator: %d", item->qoperator.oper);
                selec = 0;      /* keep compiler quiet */
                break;
        }
    }

    /* Clamp intermediate results to stay sane despite roundoff error */
    CLAMP_PROBABILITY(selec);

    return selec;
}

/*
 * bsearch() comparator for a lexeme (non-NULL terminated string with length)
 * and a TextFreq. Use length, then byte-for-byte comparison, because that's
 * how ANALYZE code sorted data before storing it in a statistic tuple.
 * See ts_typanalyze.c for details.
 */
static int
compare_lexeme_textfreq(const void *e1, const void *e2)
{
    const LexemeKey *key = (const LexemeKey *) e1;
    const TextFreq *t = (const TextFreq *) e2;
    int         len1,
                len2;

    len1 = key->length;
    len2 = VARSIZE_ANY_EXHDR(t->element);

    /* Compare lengths first, possibly avoiding a strncmp call */
    if (len1 > len2)
        return 1;
    else if (len1 < len2)
        return -1;

    /* Fall back on byte-for-byte comparison */
    return strncmp(key->lexeme, VARDATA_ANY(t->element), len1);
}