mirror of
https://git.postgresql.org/git/postgresql.git
synced 2024-12-27 08:39:28 +08:00
1265a9c8f8
We can adjust the not-yet-released cube--1.4--1.5.sql upgrade rather than making a whole new version. KaiGai Kohei Discussion: https://postgr.es/m/CAOP8fzZO4y60QPTK=RGDXeVeVHV9tLHKOsh7voUOoUouVCPV8A@mail.gmail.com
1909 lines
41 KiB
C
1909 lines
41 KiB
C
/******************************************************************************
|
|
contrib/cube/cube.c
|
|
|
|
This file contains routines that can be bound to a Postgres backend and
|
|
called by the backend in the process of processing queries. The calling
|
|
format for these routines is dictated by Postgres architecture.
|
|
******************************************************************************/
|
|
|
|
#include "postgres.h"
|
|
|
|
#include <math.h>
|
|
|
|
#include "access/gist.h"
|
|
#include "access/stratnum.h"
|
|
#include "cubedata.h"
|
|
#include "libpq/pqformat.h"
|
|
#include "utils/array.h"
|
|
#include "utils/float.h"
|
|
|
|
PG_MODULE_MAGIC;
|
|
|
|
/*
|
|
* Taken from the intarray contrib header
|
|
*/
|
|
#define ARRPTR(x) ( (double *) ARR_DATA_PTR(x) )
|
|
#define ARRNELEMS(x) ArrayGetNItems( ARR_NDIM(x), ARR_DIMS(x))
|
|
|
|
/*
|
|
** Input/Output routines
|
|
*/
|
|
PG_FUNCTION_INFO_V1(cube_in);
|
|
PG_FUNCTION_INFO_V1(cube_a_f8_f8);
|
|
PG_FUNCTION_INFO_V1(cube_a_f8);
|
|
PG_FUNCTION_INFO_V1(cube_out);
|
|
PG_FUNCTION_INFO_V1(cube_send);
|
|
PG_FUNCTION_INFO_V1(cube_recv);
|
|
PG_FUNCTION_INFO_V1(cube_f8);
|
|
PG_FUNCTION_INFO_V1(cube_f8_f8);
|
|
PG_FUNCTION_INFO_V1(cube_c_f8);
|
|
PG_FUNCTION_INFO_V1(cube_c_f8_f8);
|
|
PG_FUNCTION_INFO_V1(cube_dim);
|
|
PG_FUNCTION_INFO_V1(cube_ll_coord);
|
|
PG_FUNCTION_INFO_V1(cube_ur_coord);
|
|
PG_FUNCTION_INFO_V1(cube_coord);
|
|
PG_FUNCTION_INFO_V1(cube_coord_llur);
|
|
PG_FUNCTION_INFO_V1(cube_subset);
|
|
|
|
/*
|
|
** GiST support methods
|
|
*/
|
|
|
|
PG_FUNCTION_INFO_V1(g_cube_consistent);
|
|
PG_FUNCTION_INFO_V1(g_cube_compress);
|
|
PG_FUNCTION_INFO_V1(g_cube_decompress);
|
|
PG_FUNCTION_INFO_V1(g_cube_penalty);
|
|
PG_FUNCTION_INFO_V1(g_cube_picksplit);
|
|
PG_FUNCTION_INFO_V1(g_cube_union);
|
|
PG_FUNCTION_INFO_V1(g_cube_same);
|
|
PG_FUNCTION_INFO_V1(g_cube_distance);
|
|
|
|
/*
|
|
** B-tree support functions
|
|
*/
|
|
PG_FUNCTION_INFO_V1(cube_eq);
|
|
PG_FUNCTION_INFO_V1(cube_ne);
|
|
PG_FUNCTION_INFO_V1(cube_lt);
|
|
PG_FUNCTION_INFO_V1(cube_gt);
|
|
PG_FUNCTION_INFO_V1(cube_le);
|
|
PG_FUNCTION_INFO_V1(cube_ge);
|
|
PG_FUNCTION_INFO_V1(cube_cmp);
|
|
|
|
/*
|
|
** R-tree support functions
|
|
*/
|
|
|
|
PG_FUNCTION_INFO_V1(cube_contains);
|
|
PG_FUNCTION_INFO_V1(cube_contained);
|
|
PG_FUNCTION_INFO_V1(cube_overlap);
|
|
PG_FUNCTION_INFO_V1(cube_union);
|
|
PG_FUNCTION_INFO_V1(cube_inter);
|
|
PG_FUNCTION_INFO_V1(cube_size);
|
|
|
|
/*
|
|
** miscellaneous
|
|
*/
|
|
PG_FUNCTION_INFO_V1(distance_taxicab);
|
|
PG_FUNCTION_INFO_V1(cube_distance);
|
|
PG_FUNCTION_INFO_V1(distance_chebyshev);
|
|
PG_FUNCTION_INFO_V1(cube_is_point);
|
|
PG_FUNCTION_INFO_V1(cube_enlarge);
|
|
|
|
/*
|
|
** For internal use only
|
|
*/
|
|
int32 cube_cmp_v0(NDBOX *a, NDBOX *b);
|
|
bool cube_contains_v0(NDBOX *a, NDBOX *b);
|
|
bool cube_overlap_v0(NDBOX *a, NDBOX *b);
|
|
NDBOX *cube_union_v0(NDBOX *a, NDBOX *b);
|
|
void rt_cube_size(NDBOX *a, double *sz);
|
|
NDBOX *g_cube_binary_union(NDBOX *r1, NDBOX *r2, int *sizep);
|
|
bool g_cube_leaf_consistent(NDBOX *key, NDBOX *query, StrategyNumber strategy);
|
|
bool g_cube_internal_consistent(NDBOX *key, NDBOX *query, StrategyNumber strategy);
|
|
|
|
/*
|
|
** Auxiliary functions
|
|
*/
|
|
static double distance_1D(double a1, double a2, double b1, double b2);
|
|
static bool cube_is_point_internal(NDBOX *cube);
|
|
|
|
|
|
/*****************************************************************************
|
|
* Input/Output functions
|
|
*****************************************************************************/
|
|
|
|
/* NdBox = [(lowerleft),(upperright)] */
|
|
/* [(xLL(1)...xLL(N)),(xUR(1)...xUR(n))] */
|
|
Datum
|
|
cube_in(PG_FUNCTION_ARGS)
|
|
{
|
|
char *str = PG_GETARG_CSTRING(0);
|
|
NDBOX *result;
|
|
|
|
cube_scanner_init(str);
|
|
|
|
if (cube_yyparse(&result) != 0)
|
|
cube_yyerror(&result, "cube parser failed");
|
|
|
|
cube_scanner_finish();
|
|
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|
|
|
|
|
|
/*
|
|
** Allows the construction of a cube from 2 float[]'s
|
|
*/
|
|
Datum
|
|
cube_a_f8_f8(PG_FUNCTION_ARGS)
|
|
{
|
|
ArrayType *ur = PG_GETARG_ARRAYTYPE_P(0);
|
|
ArrayType *ll = PG_GETARG_ARRAYTYPE_P(1);
|
|
NDBOX *result;
|
|
int i;
|
|
int dim;
|
|
int size;
|
|
bool point;
|
|
double *dur,
|
|
*dll;
|
|
|
|
if (array_contains_nulls(ur) || array_contains_nulls(ll))
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_ARRAY_ELEMENT_ERROR),
|
|
errmsg("cannot work with arrays containing NULLs")));
|
|
|
|
dim = ARRNELEMS(ur);
|
|
if (dim > CUBE_MAX_DIM)
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
|
|
errmsg("can't extend cube"),
|
|
errdetail("A cube cannot have more than %d dimensions.",
|
|
CUBE_MAX_DIM)));
|
|
|
|
if (ARRNELEMS(ll) != dim)
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_ARRAY_ELEMENT_ERROR),
|
|
errmsg("UR and LL arrays must be of same length")));
|
|
|
|
dur = ARRPTR(ur);
|
|
dll = ARRPTR(ll);
|
|
|
|
/* Check if it's a point */
|
|
point = true;
|
|
for (i = 0; i < dim; i++)
|
|
{
|
|
if (dur[i] != dll[i])
|
|
{
|
|
point = false;
|
|
break;
|
|
}
|
|
}
|
|
|
|
size = point ? POINT_SIZE(dim) : CUBE_SIZE(dim);
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, dim);
|
|
|
|
for (i = 0; i < dim; i++)
|
|
result->x[i] = dur[i];
|
|
|
|
if (!point)
|
|
{
|
|
for (i = 0; i < dim; i++)
|
|
result->x[i + dim] = dll[i];
|
|
}
|
|
else
|
|
SET_POINT_BIT(result);
|
|
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|
|
|
|
/*
|
|
** Allows the construction of a zero-volume cube from a float[]
|
|
*/
|
|
Datum
|
|
cube_a_f8(PG_FUNCTION_ARGS)
|
|
{
|
|
ArrayType *ur = PG_GETARG_ARRAYTYPE_P(0);
|
|
NDBOX *result;
|
|
int i;
|
|
int dim;
|
|
int size;
|
|
double *dur;
|
|
|
|
if (array_contains_nulls(ur))
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_ARRAY_ELEMENT_ERROR),
|
|
errmsg("cannot work with arrays containing NULLs")));
|
|
|
|
dim = ARRNELEMS(ur);
|
|
if (dim > CUBE_MAX_DIM)
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
|
|
errmsg("array is too long"),
|
|
errdetail("A cube cannot have more than %d dimensions.",
|
|
CUBE_MAX_DIM)));
|
|
|
|
dur = ARRPTR(ur);
|
|
|
|
size = POINT_SIZE(dim);
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, dim);
|
|
SET_POINT_BIT(result);
|
|
|
|
for (i = 0; i < dim; i++)
|
|
result->x[i] = dur[i];
|
|
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|
|
|
|
Datum
|
|
cube_subset(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *c = PG_GETARG_NDBOX_P(0);
|
|
ArrayType *idx = PG_GETARG_ARRAYTYPE_P(1);
|
|
NDBOX *result;
|
|
int size,
|
|
dim,
|
|
i;
|
|
int *dx;
|
|
|
|
if (array_contains_nulls(idx))
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_ARRAY_ELEMENT_ERROR),
|
|
errmsg("cannot work with arrays containing NULLs")));
|
|
|
|
dx = (int32 *) ARR_DATA_PTR(idx);
|
|
|
|
dim = ARRNELEMS(idx);
|
|
if (dim > CUBE_MAX_DIM)
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
|
|
errmsg("array is too long"),
|
|
errdetail("A cube cannot have more than %d dimensions.",
|
|
CUBE_MAX_DIM)));
|
|
|
|
size = IS_POINT(c) ? POINT_SIZE(dim) : CUBE_SIZE(dim);
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, dim);
|
|
|
|
if (IS_POINT(c))
|
|
SET_POINT_BIT(result);
|
|
|
|
for (i = 0; i < dim; i++)
|
|
{
|
|
if ((dx[i] <= 0) || (dx[i] > DIM(c)))
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_ARRAY_ELEMENT_ERROR),
|
|
errmsg("Index out of bounds")));
|
|
result->x[i] = c->x[dx[i] - 1];
|
|
if (!IS_POINT(c))
|
|
result->x[i + dim] = c->x[dx[i] + DIM(c) - 1];
|
|
}
|
|
|
|
PG_FREE_IF_COPY(c, 0);
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|
|
|
|
Datum
|
|
cube_out(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *cube = PG_GETARG_NDBOX_P(0);
|
|
StringInfoData buf;
|
|
int dim = DIM(cube);
|
|
int i;
|
|
|
|
initStringInfo(&buf);
|
|
|
|
appendStringInfoChar(&buf, '(');
|
|
for (i = 0; i < dim; i++)
|
|
{
|
|
if (i > 0)
|
|
appendStringInfoString(&buf, ", ");
|
|
appendStringInfoString(&buf, float8out_internal(LL_COORD(cube, i)));
|
|
}
|
|
appendStringInfoChar(&buf, ')');
|
|
|
|
if (!cube_is_point_internal(cube))
|
|
{
|
|
appendStringInfoString(&buf, ",(");
|
|
for (i = 0; i < dim; i++)
|
|
{
|
|
if (i > 0)
|
|
appendStringInfoString(&buf, ", ");
|
|
appendStringInfoString(&buf, float8out_internal(UR_COORD(cube, i)));
|
|
}
|
|
appendStringInfoChar(&buf, ')');
|
|
}
|
|
|
|
PG_FREE_IF_COPY(cube, 0);
|
|
PG_RETURN_CSTRING(buf.data);
|
|
}
|
|
|
|
/*
|
|
* cube_send - a binary output handler for cube type
|
|
*/
|
|
Datum
|
|
cube_send(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *cube = PG_GETARG_NDBOX_P(0);
|
|
StringInfoData buf;
|
|
int32 i,
|
|
nitems = DIM(cube);
|
|
|
|
pq_begintypsend(&buf);
|
|
pq_sendint32(&buf, cube->header);
|
|
if (!IS_POINT(cube))
|
|
nitems += nitems;
|
|
/* for symmetry with cube_recv, we don't use LL_COORD/UR_COORD here */
|
|
for (i = 0; i < nitems; i++)
|
|
pq_sendfloat8(&buf, cube->x[i]);
|
|
|
|
PG_RETURN_BYTEA_P(pq_endtypsend(&buf));
|
|
}
|
|
|
|
/*
|
|
* cube_recv - a binary input handler for cube type
|
|
*/
|
|
Datum
|
|
cube_recv(PG_FUNCTION_ARGS)
|
|
{
|
|
StringInfo buf = (StringInfo) PG_GETARG_POINTER(0);
|
|
int32 header;
|
|
int32 i,
|
|
nitems;
|
|
NDBOX *cube;
|
|
|
|
header = pq_getmsgint(buf, sizeof(int32));
|
|
nitems = (header & DIM_MASK);
|
|
if (nitems > CUBE_MAX_DIM)
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
|
|
errmsg("cube dimension is too large"),
|
|
errdetail("A cube cannot have more than %d dimensions.",
|
|
CUBE_MAX_DIM)));
|
|
if ((header & POINT_BIT) == 0)
|
|
nitems += nitems;
|
|
cube = palloc(offsetof(NDBOX, x) + sizeof(double) * nitems);
|
|
SET_VARSIZE(cube, offsetof(NDBOX, x) + sizeof(double) * nitems);
|
|
cube->header = header;
|
|
for (i = 0; i < nitems; i++)
|
|
cube->x[i] = pq_getmsgfloat8(buf);
|
|
|
|
PG_RETURN_NDBOX_P(cube);
|
|
}
|
|
|
|
|
|
/*****************************************************************************
|
|
* GiST functions
|
|
*****************************************************************************/
|
|
|
|
/*
|
|
** The GiST Consistent method for boxes
|
|
** Should return false if for all data items x below entry,
|
|
** the predicate x op query == false, where op is the oper
|
|
** corresponding to strategy in the pg_amop table.
|
|
*/
|
|
Datum
|
|
g_cube_consistent(PG_FUNCTION_ARGS)
|
|
{
|
|
GISTENTRY *entry = (GISTENTRY *) PG_GETARG_POINTER(0);
|
|
NDBOX *query = PG_GETARG_NDBOX_P(1);
|
|
StrategyNumber strategy = (StrategyNumber) PG_GETARG_UINT16(2);
|
|
|
|
/* Oid subtype = PG_GETARG_OID(3); */
|
|
bool *recheck = (bool *) PG_GETARG_POINTER(4);
|
|
bool res;
|
|
|
|
/* All cases served by this function are exact */
|
|
*recheck = false;
|
|
|
|
/*
|
|
* if entry is not leaf, use g_cube_internal_consistent, else use
|
|
* g_cube_leaf_consistent
|
|
*/
|
|
if (GIST_LEAF(entry))
|
|
res = g_cube_leaf_consistent(DatumGetNDBOXP(entry->key),
|
|
query, strategy);
|
|
else
|
|
res = g_cube_internal_consistent(DatumGetNDBOXP(entry->key),
|
|
query, strategy);
|
|
|
|
PG_FREE_IF_COPY(query, 1);
|
|
PG_RETURN_BOOL(res);
|
|
}
|
|
|
|
|
|
/*
|
|
** The GiST Union method for boxes
|
|
** returns the minimal bounding box that encloses all the entries in entryvec
|
|
*/
|
|
Datum
|
|
g_cube_union(PG_FUNCTION_ARGS)
|
|
{
|
|
GistEntryVector *entryvec = (GistEntryVector *) PG_GETARG_POINTER(0);
|
|
int *sizep = (int *) PG_GETARG_POINTER(1);
|
|
NDBOX *out = (NDBOX *) NULL;
|
|
NDBOX *tmp;
|
|
int i;
|
|
|
|
tmp = DatumGetNDBOXP(entryvec->vector[0].key);
|
|
|
|
/*
|
|
* sizep = sizeof(NDBOX); -- NDBOX has variable size
|
|
*/
|
|
*sizep = VARSIZE(tmp);
|
|
|
|
for (i = 1; i < entryvec->n; i++)
|
|
{
|
|
out = g_cube_binary_union(tmp,
|
|
DatumGetNDBOXP(entryvec->vector[i].key),
|
|
sizep);
|
|
tmp = out;
|
|
}
|
|
|
|
PG_RETURN_POINTER(out);
|
|
}
|
|
|
|
/*
|
|
** GiST Compress and Decompress methods for boxes
|
|
** do not do anything.
|
|
*/
|
|
|
|
Datum
|
|
g_cube_compress(PG_FUNCTION_ARGS)
|
|
{
|
|
PG_RETURN_DATUM(PG_GETARG_DATUM(0));
|
|
}
|
|
|
|
Datum
|
|
g_cube_decompress(PG_FUNCTION_ARGS)
|
|
{
|
|
GISTENTRY *entry = (GISTENTRY *) PG_GETARG_POINTER(0);
|
|
NDBOX *key = DatumGetNDBOXP(entry->key);
|
|
|
|
if (key != DatumGetNDBOXP(entry->key))
|
|
{
|
|
GISTENTRY *retval = (GISTENTRY *) palloc(sizeof(GISTENTRY));
|
|
|
|
gistentryinit(*retval, PointerGetDatum(key),
|
|
entry->rel, entry->page,
|
|
entry->offset, false);
|
|
PG_RETURN_POINTER(retval);
|
|
}
|
|
PG_RETURN_POINTER(entry);
|
|
}
|
|
|
|
|
|
/*
|
|
** The GiST Penalty method for boxes
|
|
** As in the R-tree paper, we use change in area as our penalty metric
|
|
*/
|
|
Datum
|
|
g_cube_penalty(PG_FUNCTION_ARGS)
|
|
{
|
|
GISTENTRY *origentry = (GISTENTRY *) PG_GETARG_POINTER(0);
|
|
GISTENTRY *newentry = (GISTENTRY *) PG_GETARG_POINTER(1);
|
|
float *result = (float *) PG_GETARG_POINTER(2);
|
|
NDBOX *ud;
|
|
double tmp1,
|
|
tmp2;
|
|
|
|
ud = cube_union_v0(DatumGetNDBOXP(origentry->key),
|
|
DatumGetNDBOXP(newentry->key));
|
|
rt_cube_size(ud, &tmp1);
|
|
rt_cube_size(DatumGetNDBOXP(origentry->key), &tmp2);
|
|
*result = (float) (tmp1 - tmp2);
|
|
|
|
PG_RETURN_FLOAT8(*result);
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
** The GiST PickSplit method for boxes
|
|
** We use Guttman's poly time split algorithm
|
|
*/
|
|
Datum
|
|
g_cube_picksplit(PG_FUNCTION_ARGS)
|
|
{
|
|
GistEntryVector *entryvec = (GistEntryVector *) PG_GETARG_POINTER(0);
|
|
GIST_SPLITVEC *v = (GIST_SPLITVEC *) PG_GETARG_POINTER(1);
|
|
OffsetNumber i,
|
|
j;
|
|
NDBOX *datum_alpha,
|
|
*datum_beta;
|
|
NDBOX *datum_l,
|
|
*datum_r;
|
|
NDBOX *union_d,
|
|
*union_dl,
|
|
*union_dr;
|
|
NDBOX *inter_d;
|
|
bool firsttime;
|
|
double size_alpha,
|
|
size_beta,
|
|
size_union,
|
|
size_inter;
|
|
double size_waste,
|
|
waste;
|
|
double size_l,
|
|
size_r;
|
|
int nbytes;
|
|
OffsetNumber seed_1 = 1,
|
|
seed_2 = 2;
|
|
OffsetNumber *left,
|
|
*right;
|
|
OffsetNumber maxoff;
|
|
|
|
maxoff = entryvec->n - 2;
|
|
nbytes = (maxoff + 2) * sizeof(OffsetNumber);
|
|
v->spl_left = (OffsetNumber *) palloc(nbytes);
|
|
v->spl_right = (OffsetNumber *) palloc(nbytes);
|
|
|
|
firsttime = true;
|
|
waste = 0.0;
|
|
|
|
for (i = FirstOffsetNumber; i < maxoff; i = OffsetNumberNext(i))
|
|
{
|
|
datum_alpha = DatumGetNDBOXP(entryvec->vector[i].key);
|
|
for (j = OffsetNumberNext(i); j <= maxoff; j = OffsetNumberNext(j))
|
|
{
|
|
datum_beta = DatumGetNDBOXP(entryvec->vector[j].key);
|
|
|
|
/* compute the wasted space by unioning these guys */
|
|
/* size_waste = size_union - size_inter; */
|
|
union_d = cube_union_v0(datum_alpha, datum_beta);
|
|
rt_cube_size(union_d, &size_union);
|
|
inter_d = DatumGetNDBOXP(DirectFunctionCall2(cube_inter,
|
|
entryvec->vector[i].key,
|
|
entryvec->vector[j].key));
|
|
rt_cube_size(inter_d, &size_inter);
|
|
size_waste = size_union - size_inter;
|
|
|
|
/*
|
|
* are these a more promising split than what we've already seen?
|
|
*/
|
|
|
|
if (size_waste > waste || firsttime)
|
|
{
|
|
waste = size_waste;
|
|
seed_1 = i;
|
|
seed_2 = j;
|
|
firsttime = false;
|
|
}
|
|
}
|
|
}
|
|
|
|
left = v->spl_left;
|
|
v->spl_nleft = 0;
|
|
right = v->spl_right;
|
|
v->spl_nright = 0;
|
|
|
|
datum_alpha = DatumGetNDBOXP(entryvec->vector[seed_1].key);
|
|
datum_l = cube_union_v0(datum_alpha, datum_alpha);
|
|
rt_cube_size(datum_l, &size_l);
|
|
datum_beta = DatumGetNDBOXP(entryvec->vector[seed_2].key);
|
|
datum_r = cube_union_v0(datum_beta, datum_beta);
|
|
rt_cube_size(datum_r, &size_r);
|
|
|
|
/*
|
|
* Now split up the regions between the two seeds. An important property
|
|
* of this split algorithm is that the split vector v has the indices of
|
|
* items to be split in order in its left and right vectors. We exploit
|
|
* this property by doing a merge in the code that actually splits the
|
|
* page.
|
|
*
|
|
* For efficiency, we also place the new index tuple in this loop. This is
|
|
* handled at the very end, when we have placed all the existing tuples
|
|
* and i == maxoff + 1.
|
|
*/
|
|
|
|
maxoff = OffsetNumberNext(maxoff);
|
|
for (i = FirstOffsetNumber; i <= maxoff; i = OffsetNumberNext(i))
|
|
{
|
|
/*
|
|
* If we've already decided where to place this item, just put it on
|
|
* the right list. Otherwise, we need to figure out which page needs
|
|
* the least enlargement in order to store the item.
|
|
*/
|
|
|
|
if (i == seed_1)
|
|
{
|
|
*left++ = i;
|
|
v->spl_nleft++;
|
|
continue;
|
|
}
|
|
else if (i == seed_2)
|
|
{
|
|
*right++ = i;
|
|
v->spl_nright++;
|
|
continue;
|
|
}
|
|
|
|
/* okay, which page needs least enlargement? */
|
|
datum_alpha = DatumGetNDBOXP(entryvec->vector[i].key);
|
|
union_dl = cube_union_v0(datum_l, datum_alpha);
|
|
union_dr = cube_union_v0(datum_r, datum_alpha);
|
|
rt_cube_size(union_dl, &size_alpha);
|
|
rt_cube_size(union_dr, &size_beta);
|
|
|
|
/* pick which page to add it to */
|
|
if (size_alpha - size_l < size_beta - size_r)
|
|
{
|
|
datum_l = union_dl;
|
|
size_l = size_alpha;
|
|
*left++ = i;
|
|
v->spl_nleft++;
|
|
}
|
|
else
|
|
{
|
|
datum_r = union_dr;
|
|
size_r = size_beta;
|
|
*right++ = i;
|
|
v->spl_nright++;
|
|
}
|
|
}
|
|
*left = *right = FirstOffsetNumber; /* sentinel value */
|
|
|
|
v->spl_ldatum = PointerGetDatum(datum_l);
|
|
v->spl_rdatum = PointerGetDatum(datum_r);
|
|
|
|
PG_RETURN_POINTER(v);
|
|
}
|
|
|
|
/*
|
|
** Equality method
|
|
*/
|
|
Datum
|
|
g_cube_same(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *b1 = PG_GETARG_NDBOX_P(0);
|
|
NDBOX *b2 = PG_GETARG_NDBOX_P(1);
|
|
bool *result = (bool *) PG_GETARG_POINTER(2);
|
|
|
|
if (cube_cmp_v0(b1, b2) == 0)
|
|
*result = true;
|
|
else
|
|
*result = false;
|
|
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|
|
|
|
/*
|
|
** SUPPORT ROUTINES
|
|
*/
|
|
bool
|
|
g_cube_leaf_consistent(NDBOX *key,
|
|
NDBOX *query,
|
|
StrategyNumber strategy)
|
|
{
|
|
bool retval;
|
|
|
|
switch (strategy)
|
|
{
|
|
case RTOverlapStrategyNumber:
|
|
retval = cube_overlap_v0(key, query);
|
|
break;
|
|
case RTSameStrategyNumber:
|
|
retval = (cube_cmp_v0(key, query) == 0);
|
|
break;
|
|
case RTContainsStrategyNumber:
|
|
case RTOldContainsStrategyNumber:
|
|
retval = cube_contains_v0(key, query);
|
|
break;
|
|
case RTContainedByStrategyNumber:
|
|
case RTOldContainedByStrategyNumber:
|
|
retval = cube_contains_v0(query, key);
|
|
break;
|
|
default:
|
|
retval = false;
|
|
}
|
|
return retval;
|
|
}
|
|
|
|
bool
|
|
g_cube_internal_consistent(NDBOX *key,
|
|
NDBOX *query,
|
|
StrategyNumber strategy)
|
|
{
|
|
bool retval;
|
|
|
|
switch (strategy)
|
|
{
|
|
case RTOverlapStrategyNumber:
|
|
retval = (bool) cube_overlap_v0(key, query);
|
|
break;
|
|
case RTSameStrategyNumber:
|
|
case RTContainsStrategyNumber:
|
|
case RTOldContainsStrategyNumber:
|
|
retval = (bool) cube_contains_v0(key, query);
|
|
break;
|
|
case RTContainedByStrategyNumber:
|
|
case RTOldContainedByStrategyNumber:
|
|
retval = (bool) cube_overlap_v0(key, query);
|
|
break;
|
|
default:
|
|
retval = false;
|
|
}
|
|
return retval;
|
|
}
|
|
|
|
NDBOX *
|
|
g_cube_binary_union(NDBOX *r1, NDBOX *r2, int *sizep)
|
|
{
|
|
NDBOX *retval;
|
|
|
|
retval = cube_union_v0(r1, r2);
|
|
*sizep = VARSIZE(retval);
|
|
|
|
return retval;
|
|
}
|
|
|
|
|
|
/* cube_union_v0 */
|
|
NDBOX *
|
|
cube_union_v0(NDBOX *a, NDBOX *b)
|
|
{
|
|
int i;
|
|
NDBOX *result;
|
|
int dim;
|
|
int size;
|
|
|
|
/* trivial case */
|
|
if (a == b)
|
|
return a;
|
|
|
|
/* swap the arguments if needed, so that 'a' is always larger than 'b' */
|
|
if (DIM(a) < DIM(b))
|
|
{
|
|
NDBOX *tmp = b;
|
|
|
|
b = a;
|
|
a = tmp;
|
|
}
|
|
dim = DIM(a);
|
|
|
|
size = CUBE_SIZE(dim);
|
|
result = palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, dim);
|
|
|
|
/* First compute the union of the dimensions present in both args */
|
|
for (i = 0; i < DIM(b); i++)
|
|
{
|
|
result->x[i] = Min(Min(LL_COORD(a, i), UR_COORD(a, i)),
|
|
Min(LL_COORD(b, i), UR_COORD(b, i)));
|
|
result->x[i + DIM(a)] = Max(Max(LL_COORD(a, i), UR_COORD(a, i)),
|
|
Max(LL_COORD(b, i), UR_COORD(b, i)));
|
|
}
|
|
/* continue on the higher dimensions only present in 'a' */
|
|
for (; i < DIM(a); i++)
|
|
{
|
|
result->x[i] = Min(0,
|
|
Min(LL_COORD(a, i), UR_COORD(a, i))
|
|
);
|
|
result->x[i + dim] = Max(0,
|
|
Max(LL_COORD(a, i), UR_COORD(a, i))
|
|
);
|
|
}
|
|
|
|
/*
|
|
* Check if the result was in fact a point, and set the flag in the datum
|
|
* accordingly. (we don't bother to repalloc it smaller)
|
|
*/
|
|
if (cube_is_point_internal(result))
|
|
{
|
|
size = POINT_SIZE(dim);
|
|
SET_VARSIZE(result, size);
|
|
SET_POINT_BIT(result);
|
|
}
|
|
|
|
return result;
|
|
}
|
|
|
|
Datum
|
|
cube_union(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0);
|
|
NDBOX *b = PG_GETARG_NDBOX_P(1);
|
|
NDBOX *res;
|
|
|
|
res = cube_union_v0(a, b);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_NDBOX_P(res);
|
|
}
|
|
|
|
/* cube_inter */
|
|
Datum
|
|
cube_inter(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0);
|
|
NDBOX *b = PG_GETARG_NDBOX_P(1);
|
|
NDBOX *result;
|
|
bool swapped = false;
|
|
int i;
|
|
int dim;
|
|
int size;
|
|
|
|
/* swap the arguments if needed, so that 'a' is always larger than 'b' */
|
|
if (DIM(a) < DIM(b))
|
|
{
|
|
NDBOX *tmp = b;
|
|
|
|
b = a;
|
|
a = tmp;
|
|
swapped = true;
|
|
}
|
|
dim = DIM(a);
|
|
|
|
size = CUBE_SIZE(dim);
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, dim);
|
|
|
|
/* First compute intersection of the dimensions present in both args */
|
|
for (i = 0; i < DIM(b); i++)
|
|
{
|
|
result->x[i] = Max(Min(LL_COORD(a, i), UR_COORD(a, i)),
|
|
Min(LL_COORD(b, i), UR_COORD(b, i)));
|
|
result->x[i + DIM(a)] = Min(Max(LL_COORD(a, i), UR_COORD(a, i)),
|
|
Max(LL_COORD(b, i), UR_COORD(b, i)));
|
|
}
|
|
/* continue on the higher dimensions only present in 'a' */
|
|
for (; i < DIM(a); i++)
|
|
{
|
|
result->x[i] = Max(0,
|
|
Min(LL_COORD(a, i), UR_COORD(a, i))
|
|
);
|
|
result->x[i + DIM(a)] = Min(0,
|
|
Max(LL_COORD(a, i), UR_COORD(a, i))
|
|
);
|
|
}
|
|
|
|
/*
|
|
* Check if the result was in fact a point, and set the flag in the datum
|
|
* accordingly. (we don't bother to repalloc it smaller)
|
|
*/
|
|
if (cube_is_point_internal(result))
|
|
{
|
|
size = POINT_SIZE(dim);
|
|
result = repalloc(result, size);
|
|
SET_VARSIZE(result, size);
|
|
SET_POINT_BIT(result);
|
|
}
|
|
|
|
if (swapped)
|
|
{
|
|
PG_FREE_IF_COPY(b, 0);
|
|
PG_FREE_IF_COPY(a, 1);
|
|
}
|
|
else
|
|
{
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
}
|
|
|
|
/*
|
|
* Is it OK to return a non-null intersection for non-overlapping boxes?
|
|
*/
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|
|
|
|
/* cube_size */
|
|
Datum
|
|
cube_size(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0);
|
|
double result;
|
|
|
|
rt_cube_size(a, &result);
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_RETURN_FLOAT8(result);
|
|
}
|
|
|
|
void
|
|
rt_cube_size(NDBOX *a, double *size)
|
|
{
|
|
double result;
|
|
int i;
|
|
|
|
if (a == (NDBOX *) NULL)
|
|
{
|
|
/* special case for GiST */
|
|
result = 0.0;
|
|
}
|
|
else if (IS_POINT(a) || DIM(a) == 0)
|
|
{
|
|
/* necessarily has zero size */
|
|
result = 0.0;
|
|
}
|
|
else
|
|
{
|
|
result = 1.0;
|
|
for (i = 0; i < DIM(a); i++)
|
|
result *= Abs(UR_COORD(a, i) - LL_COORD(a, i));
|
|
}
|
|
*size = result;
|
|
}
|
|
|
|
/* make up a metric in which one box will be 'lower' than the other
|
|
-- this can be useful for sorting and to determine uniqueness */
|
|
int32
|
|
cube_cmp_v0(NDBOX *a, NDBOX *b)
|
|
{
|
|
int i;
|
|
int dim;
|
|
|
|
dim = Min(DIM(a), DIM(b));
|
|
|
|
/* compare the common dimensions */
|
|
for (i = 0; i < dim; i++)
|
|
{
|
|
if (Min(LL_COORD(a, i), UR_COORD(a, i)) >
|
|
Min(LL_COORD(b, i), UR_COORD(b, i)))
|
|
return 1;
|
|
if (Min(LL_COORD(a, i), UR_COORD(a, i)) <
|
|
Min(LL_COORD(b, i), UR_COORD(b, i)))
|
|
return -1;
|
|
}
|
|
for (i = 0; i < dim; i++)
|
|
{
|
|
if (Max(LL_COORD(a, i), UR_COORD(a, i)) >
|
|
Max(LL_COORD(b, i), UR_COORD(b, i)))
|
|
return 1;
|
|
if (Max(LL_COORD(a, i), UR_COORD(a, i)) <
|
|
Max(LL_COORD(b, i), UR_COORD(b, i)))
|
|
return -1;
|
|
}
|
|
|
|
/* compare extra dimensions to zero */
|
|
if (DIM(a) > DIM(b))
|
|
{
|
|
for (i = dim; i < DIM(a); i++)
|
|
{
|
|
if (Min(LL_COORD(a, i), UR_COORD(a, i)) > 0)
|
|
return 1;
|
|
if (Min(LL_COORD(a, i), UR_COORD(a, i)) < 0)
|
|
return -1;
|
|
}
|
|
for (i = dim; i < DIM(a); i++)
|
|
{
|
|
if (Max(LL_COORD(a, i), UR_COORD(a, i)) > 0)
|
|
return 1;
|
|
if (Max(LL_COORD(a, i), UR_COORD(a, i)) < 0)
|
|
return -1;
|
|
}
|
|
|
|
/*
|
|
* if all common dimensions are equal, the cube with more dimensions
|
|
* wins
|
|
*/
|
|
return 1;
|
|
}
|
|
if (DIM(a) < DIM(b))
|
|
{
|
|
for (i = dim; i < DIM(b); i++)
|
|
{
|
|
if (Min(LL_COORD(b, i), UR_COORD(b, i)) > 0)
|
|
return -1;
|
|
if (Min(LL_COORD(b, i), UR_COORD(b, i)) < 0)
|
|
return 1;
|
|
}
|
|
for (i = dim; i < DIM(b); i++)
|
|
{
|
|
if (Max(LL_COORD(b, i), UR_COORD(b, i)) > 0)
|
|
return -1;
|
|
if (Max(LL_COORD(b, i), UR_COORD(b, i)) < 0)
|
|
return 1;
|
|
}
|
|
|
|
/*
|
|
* if all common dimensions are equal, the cube with more dimensions
|
|
* wins
|
|
*/
|
|
return -1;
|
|
}
|
|
|
|
/* They're really equal */
|
|
return 0;
|
|
}
|
|
|
|
Datum
|
|
cube_cmp(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
int32 res;
|
|
|
|
res = cube_cmp_v0(a, b);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_INT32(res);
|
|
}
|
|
|
|
|
|
Datum
|
|
cube_eq(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
int32 res;
|
|
|
|
res = cube_cmp_v0(a, b);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_BOOL(res == 0);
|
|
}
|
|
|
|
|
|
Datum
|
|
cube_ne(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
int32 res;
|
|
|
|
res = cube_cmp_v0(a, b);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_BOOL(res != 0);
|
|
}
|
|
|
|
|
|
Datum
|
|
cube_lt(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
int32 res;
|
|
|
|
res = cube_cmp_v0(a, b);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_BOOL(res < 0);
|
|
}
|
|
|
|
|
|
Datum
|
|
cube_gt(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
int32 res;
|
|
|
|
res = cube_cmp_v0(a, b);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_BOOL(res > 0);
|
|
}
|
|
|
|
|
|
Datum
|
|
cube_le(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
int32 res;
|
|
|
|
res = cube_cmp_v0(a, b);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_BOOL(res <= 0);
|
|
}
|
|
|
|
|
|
Datum
|
|
cube_ge(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
int32 res;
|
|
|
|
res = cube_cmp_v0(a, b);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_BOOL(res >= 0);
|
|
}
|
|
|
|
|
|
/* Contains */
|
|
/* Box(A) CONTAINS Box(B) IFF pt(A) < pt(B) */
|
|
bool
|
|
cube_contains_v0(NDBOX *a, NDBOX *b)
|
|
{
|
|
int i;
|
|
|
|
if ((a == NULL) || (b == NULL))
|
|
return false;
|
|
|
|
if (DIM(a) < DIM(b))
|
|
{
|
|
/*
|
|
* the further comparisons will make sense if the excess dimensions of
|
|
* (b) were zeroes Since both UL and UR coordinates must be zero, we
|
|
* can check them all without worrying about which is which.
|
|
*/
|
|
for (i = DIM(a); i < DIM(b); i++)
|
|
{
|
|
if (LL_COORD(b, i) != 0)
|
|
return false;
|
|
if (UR_COORD(b, i) != 0)
|
|
return false;
|
|
}
|
|
}
|
|
|
|
/* Can't care less about the excess dimensions of (a), if any */
|
|
for (i = 0; i < Min(DIM(a), DIM(b)); i++)
|
|
{
|
|
if (Min(LL_COORD(a, i), UR_COORD(a, i)) >
|
|
Min(LL_COORD(b, i), UR_COORD(b, i)))
|
|
return false;
|
|
if (Max(LL_COORD(a, i), UR_COORD(a, i)) <
|
|
Max(LL_COORD(b, i), UR_COORD(b, i)))
|
|
return false;
|
|
}
|
|
|
|
return true;
|
|
}
|
|
|
|
Datum
|
|
cube_contains(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
bool res;
|
|
|
|
res = cube_contains_v0(a, b);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_BOOL(res);
|
|
}
|
|
|
|
/* Contained */
|
|
/* Box(A) Contained by Box(B) IFF Box(B) Contains Box(A) */
|
|
Datum
|
|
cube_contained(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
bool res;
|
|
|
|
res = cube_contains_v0(b, a);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_BOOL(res);
|
|
}
|
|
|
|
/* Overlap */
|
|
/* Box(A) Overlap Box(B) IFF (pt(a)LL < pt(B)UR) && (pt(b)LL < pt(a)UR) */
|
|
bool
|
|
cube_overlap_v0(NDBOX *a, NDBOX *b)
|
|
{
|
|
int i;
|
|
|
|
if ((a == NULL) || (b == NULL))
|
|
return false;
|
|
|
|
/* swap the box pointers if needed */
|
|
if (DIM(a) < DIM(b))
|
|
{
|
|
NDBOX *tmp = b;
|
|
|
|
b = a;
|
|
a = tmp;
|
|
}
|
|
|
|
/* compare within the dimensions of (b) */
|
|
for (i = 0; i < DIM(b); i++)
|
|
{
|
|
if (Min(LL_COORD(a, i), UR_COORD(a, i)) > Max(LL_COORD(b, i), UR_COORD(b, i)))
|
|
return false;
|
|
if (Max(LL_COORD(a, i), UR_COORD(a, i)) < Min(LL_COORD(b, i), UR_COORD(b, i)))
|
|
return false;
|
|
}
|
|
|
|
/* compare to zero those dimensions in (a) absent in (b) */
|
|
for (i = DIM(b); i < DIM(a); i++)
|
|
{
|
|
if (Min(LL_COORD(a, i), UR_COORD(a, i)) > 0)
|
|
return false;
|
|
if (Max(LL_COORD(a, i), UR_COORD(a, i)) < 0)
|
|
return false;
|
|
}
|
|
|
|
return true;
|
|
}
|
|
|
|
|
|
Datum
|
|
cube_overlap(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
bool res;
|
|
|
|
res = cube_overlap_v0(a, b);
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
PG_RETURN_BOOL(res);
|
|
}
|
|
|
|
|
|
/* Distance */
|
|
/* The distance is computed as a per axis sum of the squared distances
|
|
between 1D projections of the boxes onto Cartesian axes. Assuming zero
|
|
distance between overlapping projections, this metric coincides with the
|
|
"common sense" geometric distance */
|
|
Datum
|
|
cube_distance(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
bool swapped = false;
|
|
double d,
|
|
distance;
|
|
int i;
|
|
|
|
/* swap the box pointers if needed */
|
|
if (DIM(a) < DIM(b))
|
|
{
|
|
NDBOX *tmp = b;
|
|
|
|
b = a;
|
|
a = tmp;
|
|
swapped = true;
|
|
}
|
|
|
|
distance = 0.0;
|
|
/* compute within the dimensions of (b) */
|
|
for (i = 0; i < DIM(b); i++)
|
|
{
|
|
d = distance_1D(LL_COORD(a, i), UR_COORD(a, i), LL_COORD(b, i), UR_COORD(b, i));
|
|
distance += d * d;
|
|
}
|
|
|
|
/* compute distance to zero for those dimensions in (a) absent in (b) */
|
|
for (i = DIM(b); i < DIM(a); i++)
|
|
{
|
|
d = distance_1D(LL_COORD(a, i), UR_COORD(a, i), 0.0, 0.0);
|
|
distance += d * d;
|
|
}
|
|
|
|
if (swapped)
|
|
{
|
|
PG_FREE_IF_COPY(b, 0);
|
|
PG_FREE_IF_COPY(a, 1);
|
|
}
|
|
else
|
|
{
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
}
|
|
|
|
PG_RETURN_FLOAT8(sqrt(distance));
|
|
}
|
|
|
|
Datum
|
|
distance_taxicab(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
bool swapped = false;
|
|
double distance;
|
|
int i;
|
|
|
|
/* swap the box pointers if needed */
|
|
if (DIM(a) < DIM(b))
|
|
{
|
|
NDBOX *tmp = b;
|
|
|
|
b = a;
|
|
a = tmp;
|
|
swapped = true;
|
|
}
|
|
|
|
distance = 0.0;
|
|
/* compute within the dimensions of (b) */
|
|
for (i = 0; i < DIM(b); i++)
|
|
distance += fabs(distance_1D(LL_COORD(a, i), UR_COORD(a, i),
|
|
LL_COORD(b, i), UR_COORD(b, i)));
|
|
|
|
/* compute distance to zero for those dimensions in (a) absent in (b) */
|
|
for (i = DIM(b); i < DIM(a); i++)
|
|
distance += fabs(distance_1D(LL_COORD(a, i), UR_COORD(a, i),
|
|
0.0, 0.0));
|
|
|
|
if (swapped)
|
|
{
|
|
PG_FREE_IF_COPY(b, 0);
|
|
PG_FREE_IF_COPY(a, 1);
|
|
}
|
|
else
|
|
{
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
}
|
|
|
|
PG_RETURN_FLOAT8(distance);
|
|
}
|
|
|
|
Datum
|
|
distance_chebyshev(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0),
|
|
*b = PG_GETARG_NDBOX_P(1);
|
|
bool swapped = false;
|
|
double d,
|
|
distance;
|
|
int i;
|
|
|
|
/* swap the box pointers if needed */
|
|
if (DIM(a) < DIM(b))
|
|
{
|
|
NDBOX *tmp = b;
|
|
|
|
b = a;
|
|
a = tmp;
|
|
swapped = true;
|
|
}
|
|
|
|
distance = 0.0;
|
|
/* compute within the dimensions of (b) */
|
|
for (i = 0; i < DIM(b); i++)
|
|
{
|
|
d = fabs(distance_1D(LL_COORD(a, i), UR_COORD(a, i),
|
|
LL_COORD(b, i), UR_COORD(b, i)));
|
|
if (d > distance)
|
|
distance = d;
|
|
}
|
|
|
|
/* compute distance to zero for those dimensions in (a) absent in (b) */
|
|
for (i = DIM(b); i < DIM(a); i++)
|
|
{
|
|
d = fabs(distance_1D(LL_COORD(a, i), UR_COORD(a, i), 0.0, 0.0));
|
|
if (d > distance)
|
|
distance = d;
|
|
}
|
|
|
|
if (swapped)
|
|
{
|
|
PG_FREE_IF_COPY(b, 0);
|
|
PG_FREE_IF_COPY(a, 1);
|
|
}
|
|
else
|
|
{
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_FREE_IF_COPY(b, 1);
|
|
}
|
|
|
|
PG_RETURN_FLOAT8(distance);
|
|
}
|
|
|
|
Datum
|
|
g_cube_distance(PG_FUNCTION_ARGS)
|
|
{
|
|
GISTENTRY *entry = (GISTENTRY *) PG_GETARG_POINTER(0);
|
|
StrategyNumber strategy = (StrategyNumber) PG_GETARG_UINT16(2);
|
|
NDBOX *cube = DatumGetNDBOXP(entry->key);
|
|
double retval;
|
|
|
|
if (strategy == CubeKNNDistanceCoord)
|
|
{
|
|
/*
|
|
* Handle ordering by ~> operator. See comments of cube_coord_llur()
|
|
* for details
|
|
*/
|
|
int coord = PG_GETARG_INT32(1);
|
|
bool isLeaf = GistPageIsLeaf(entry->page);
|
|
bool inverse = false;
|
|
|
|
/* 0 is the only unsupported coordinate value */
|
|
if (coord == 0)
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_ARRAY_ELEMENT_ERROR),
|
|
errmsg("zero cube index is not defined")));
|
|
|
|
/* Return inversed value for negative coordinate */
|
|
if (coord < 0)
|
|
{
|
|
coord = -coord;
|
|
inverse = true;
|
|
}
|
|
|
|
if (coord <= 2 * DIM(cube))
|
|
{
|
|
/* dimension index */
|
|
int index = (coord - 1) / 2;
|
|
|
|
/* whether this is upper bound (lower bound otherwise) */
|
|
bool upper = ((coord - 1) % 2 == 1);
|
|
|
|
if (IS_POINT(cube))
|
|
{
|
|
retval = cube->x[index];
|
|
}
|
|
else
|
|
{
|
|
if (isLeaf)
|
|
{
|
|
/* For leaf just return required upper/lower bound */
|
|
if (upper)
|
|
retval = Max(cube->x[index], cube->x[index + DIM(cube)]);
|
|
else
|
|
retval = Min(cube->x[index], cube->x[index + DIM(cube)]);
|
|
}
|
|
else
|
|
{
|
|
/*
|
|
* For non-leaf we should always return lower bound,
|
|
* because even upper bound of a child in the subtree can
|
|
* be as small as our lower bound. For inversed case we
|
|
* return upper bound because it becomes lower bound for
|
|
* inversed value.
|
|
*/
|
|
if (!inverse)
|
|
retval = Min(cube->x[index], cube->x[index + DIM(cube)]);
|
|
else
|
|
retval = Max(cube->x[index], cube->x[index + DIM(cube)]);
|
|
}
|
|
}
|
|
}
|
|
else
|
|
{
|
|
retval = 0.0;
|
|
}
|
|
|
|
/* Inverse return value if needed */
|
|
if (inverse)
|
|
retval = -retval;
|
|
}
|
|
else
|
|
{
|
|
NDBOX *query = PG_GETARG_NDBOX_P(1);
|
|
|
|
switch (strategy)
|
|
{
|
|
case CubeKNNDistanceTaxicab:
|
|
retval = DatumGetFloat8(DirectFunctionCall2(distance_taxicab,
|
|
PointerGetDatum(cube), PointerGetDatum(query)));
|
|
break;
|
|
case CubeKNNDistanceEuclid:
|
|
retval = DatumGetFloat8(DirectFunctionCall2(cube_distance,
|
|
PointerGetDatum(cube), PointerGetDatum(query)));
|
|
break;
|
|
case CubeKNNDistanceChebyshev:
|
|
retval = DatumGetFloat8(DirectFunctionCall2(distance_chebyshev,
|
|
PointerGetDatum(cube), PointerGetDatum(query)));
|
|
break;
|
|
default:
|
|
elog(ERROR, "unrecognized cube strategy number: %d", strategy);
|
|
retval = 0; /* keep compiler quiet */
|
|
break;
|
|
}
|
|
}
|
|
PG_RETURN_FLOAT8(retval);
|
|
}
|
|
|
|
static double
|
|
distance_1D(double a1, double a2, double b1, double b2)
|
|
{
|
|
/* interval (a) is entirely on the left of (b) */
|
|
if ((a1 <= b1) && (a2 <= b1) && (a1 <= b2) && (a2 <= b2))
|
|
return (Min(b1, b2) - Max(a1, a2));
|
|
|
|
/* interval (a) is entirely on the right of (b) */
|
|
if ((a1 > b1) && (a2 > b1) && (a1 > b2) && (a2 > b2))
|
|
return (Min(a1, a2) - Max(b1, b2));
|
|
|
|
/* the rest are all sorts of intersections */
|
|
return 0.0;
|
|
}
|
|
|
|
/* Test if a box is also a point */
|
|
Datum
|
|
cube_is_point(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *cube = PG_GETARG_NDBOX_P(0);
|
|
bool result;
|
|
|
|
result = cube_is_point_internal(cube);
|
|
PG_FREE_IF_COPY(cube, 0);
|
|
PG_RETURN_BOOL(result);
|
|
}
|
|
|
|
static bool
|
|
cube_is_point_internal(NDBOX *cube)
|
|
{
|
|
int i;
|
|
|
|
if (IS_POINT(cube))
|
|
return true;
|
|
|
|
/*
|
|
* Even if the point-flag is not set, all the lower-left coordinates might
|
|
* match the upper-right coordinates, so that the value is in fact a
|
|
* point. Such values don't arise with current code - the point flag is
|
|
* always set if appropriate - but they might be present on-disk in
|
|
* clusters upgraded from pre-9.4 versions.
|
|
*/
|
|
for (i = 0; i < DIM(cube); i++)
|
|
{
|
|
if (LL_COORD(cube, i) != UR_COORD(cube, i))
|
|
return false;
|
|
}
|
|
return true;
|
|
}
|
|
|
|
/* Return dimensions in use in the data structure */
|
|
Datum
|
|
cube_dim(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *c = PG_GETARG_NDBOX_P(0);
|
|
int dim = DIM(c);
|
|
|
|
PG_FREE_IF_COPY(c, 0);
|
|
PG_RETURN_INT32(dim);
|
|
}
|
|
|
|
/* Return a specific normalized LL coordinate */
|
|
Datum
|
|
cube_ll_coord(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *c = PG_GETARG_NDBOX_P(0);
|
|
int n = PG_GETARG_INT32(1);
|
|
double result;
|
|
|
|
if (DIM(c) >= n && n > 0)
|
|
result = Min(LL_COORD(c, n - 1), UR_COORD(c, n - 1));
|
|
else
|
|
result = 0;
|
|
|
|
PG_FREE_IF_COPY(c, 0);
|
|
PG_RETURN_FLOAT8(result);
|
|
}
|
|
|
|
/* Return a specific normalized UR coordinate */
|
|
Datum
|
|
cube_ur_coord(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *c = PG_GETARG_NDBOX_P(0);
|
|
int n = PG_GETARG_INT32(1);
|
|
double result;
|
|
|
|
if (DIM(c) >= n && n > 0)
|
|
result = Max(LL_COORD(c, n - 1), UR_COORD(c, n - 1));
|
|
else
|
|
result = 0;
|
|
|
|
PG_FREE_IF_COPY(c, 0);
|
|
PG_RETURN_FLOAT8(result);
|
|
}
|
|
|
|
/*
|
|
* Function returns cube coordinate.
|
|
* Numbers from 1 to DIM denotes first corner coordinates.
|
|
* Numbers from DIM+1 to 2*DIM denotes second corner coordinates.
|
|
*/
|
|
Datum
|
|
cube_coord(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *cube = PG_GETARG_NDBOX_P(0);
|
|
int coord = PG_GETARG_INT32(1);
|
|
|
|
if (coord <= 0 || coord > 2 * DIM(cube))
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_ARRAY_ELEMENT_ERROR),
|
|
errmsg("cube index %d is out of bounds", coord)));
|
|
|
|
if (IS_POINT(cube))
|
|
PG_RETURN_FLOAT8(cube->x[(coord - 1) % DIM(cube)]);
|
|
else
|
|
PG_RETURN_FLOAT8(cube->x[coord - 1]);
|
|
}
|
|
|
|
|
|
/*----
|
|
* This function works like cube_coord(), but rearranges coordinates in the
|
|
* way suitable to support coordinate ordering using KNN-GiST. For historical
|
|
* reasons this extension allows us to create cubes in form ((2,1),(1,2)) and
|
|
* instead of normalizing such cube to ((1,1),(2,2)) it stores cube in original
|
|
* way. But in order to get cubes ordered by one of dimensions from the index
|
|
* without explicit sort step we need this representation-independent coordinate
|
|
* getter. Moreover, indexed dataset may contain cubes of different dimensions
|
|
* number. Accordingly, this coordinate getter should be able to return
|
|
* lower/upper bound for particular dimension independently on number of cube
|
|
* dimensions. Also, KNN-GiST supports only ascending sorting. In order to
|
|
* support descending sorting, this function returns inverse of value when
|
|
* negative coordinate is given.
|
|
*
|
|
* Long story short, this function uses following meaning of coordinates:
|
|
* # (2 * N - 1) -- lower bound of Nth dimension,
|
|
* # (2 * N) -- upper bound of Nth dimension,
|
|
* # - (2 * N - 1) -- negative of lower bound of Nth dimension,
|
|
* # - (2 * N) -- negative of upper bound of Nth dimension.
|
|
*
|
|
* When given coordinate exceeds number of cube dimensions, then 0 returned
|
|
* (reproducing logic of GiST indexing of variable-length cubes).
|
|
*/
|
|
Datum
|
|
cube_coord_llur(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *cube = PG_GETARG_NDBOX_P(0);
|
|
int coord = PG_GETARG_INT32(1);
|
|
bool inverse = false;
|
|
float8 result;
|
|
|
|
/* 0 is the only unsupported coordinate value */
|
|
if (coord == 0)
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_ARRAY_ELEMENT_ERROR),
|
|
errmsg("zero cube index is not defined")));
|
|
|
|
/* Return inversed value for negative coordinate */
|
|
if (coord < 0)
|
|
{
|
|
coord = -coord;
|
|
inverse = true;
|
|
}
|
|
|
|
if (coord <= 2 * DIM(cube))
|
|
{
|
|
/* dimension index */
|
|
int index = (coord - 1) / 2;
|
|
|
|
/* whether this is upper bound (lower bound otherwise) */
|
|
bool upper = ((coord - 1) % 2 == 1);
|
|
|
|
if (IS_POINT(cube))
|
|
{
|
|
result = cube->x[index];
|
|
}
|
|
else
|
|
{
|
|
if (upper)
|
|
result = Max(cube->x[index], cube->x[index + DIM(cube)]);
|
|
else
|
|
result = Min(cube->x[index], cube->x[index + DIM(cube)]);
|
|
}
|
|
}
|
|
else
|
|
{
|
|
/*
|
|
* Return zero if coordinate is out of bound. That reproduces logic
|
|
* of how cubes with low dimension number are expanded during GiST
|
|
* indexing.
|
|
*/
|
|
result = 0.0;
|
|
}
|
|
|
|
/* Inverse value if needed */
|
|
if (inverse)
|
|
result = -result;
|
|
|
|
PG_RETURN_FLOAT8(result);
|
|
}
|
|
|
|
/* Increase or decrease box size by a radius in at least n dimensions. */
|
|
Datum
|
|
cube_enlarge(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *a = PG_GETARG_NDBOX_P(0);
|
|
double r = PG_GETARG_FLOAT8(1);
|
|
int32 n = PG_GETARG_INT32(2);
|
|
NDBOX *result;
|
|
int dim = 0;
|
|
int size;
|
|
int i,
|
|
j;
|
|
|
|
if (n > CUBE_MAX_DIM)
|
|
n = CUBE_MAX_DIM;
|
|
if (r > 0 && n > 0)
|
|
dim = n;
|
|
if (DIM(a) > dim)
|
|
dim = DIM(a);
|
|
|
|
size = CUBE_SIZE(dim);
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, dim);
|
|
|
|
for (i = 0, j = dim; i < DIM(a); i++, j++)
|
|
{
|
|
if (LL_COORD(a, i) >= UR_COORD(a, i))
|
|
{
|
|
result->x[i] = UR_COORD(a, i) - r;
|
|
result->x[j] = LL_COORD(a, i) + r;
|
|
}
|
|
else
|
|
{
|
|
result->x[i] = LL_COORD(a, i) - r;
|
|
result->x[j] = UR_COORD(a, i) + r;
|
|
}
|
|
if (result->x[i] > result->x[j])
|
|
{
|
|
result->x[i] = (result->x[i] + result->x[j]) / 2;
|
|
result->x[j] = result->x[i];
|
|
}
|
|
}
|
|
/* dim > a->dim only if r > 0 */
|
|
for (; i < dim; i++, j++)
|
|
{
|
|
result->x[i] = -r;
|
|
result->x[j] = r;
|
|
}
|
|
|
|
/*
|
|
* Check if the result was in fact a point, and set the flag in the datum
|
|
* accordingly. (we don't bother to repalloc it smaller)
|
|
*/
|
|
if (cube_is_point_internal(result))
|
|
{
|
|
size = POINT_SIZE(dim);
|
|
SET_VARSIZE(result, size);
|
|
SET_POINT_BIT(result);
|
|
}
|
|
|
|
PG_FREE_IF_COPY(a, 0);
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|
|
|
|
/* Create a one dimensional box with identical upper and lower coordinates */
|
|
Datum
|
|
cube_f8(PG_FUNCTION_ARGS)
|
|
{
|
|
double x = PG_GETARG_FLOAT8(0);
|
|
NDBOX *result;
|
|
int size;
|
|
|
|
size = POINT_SIZE(1);
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, 1);
|
|
SET_POINT_BIT(result);
|
|
result->x[0] = x;
|
|
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|
|
|
|
/* Create a one dimensional box */
|
|
Datum
|
|
cube_f8_f8(PG_FUNCTION_ARGS)
|
|
{
|
|
double x0 = PG_GETARG_FLOAT8(0);
|
|
double x1 = PG_GETARG_FLOAT8(1);
|
|
NDBOX *result;
|
|
int size;
|
|
|
|
if (x0 == x1)
|
|
{
|
|
size = POINT_SIZE(1);
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, 1);
|
|
SET_POINT_BIT(result);
|
|
result->x[0] = x0;
|
|
}
|
|
else
|
|
{
|
|
size = CUBE_SIZE(1);
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, 1);
|
|
result->x[0] = x0;
|
|
result->x[1] = x1;
|
|
}
|
|
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|
|
|
|
/* Add a dimension to an existing cube with the same values for the new
|
|
coordinate */
|
|
Datum
|
|
cube_c_f8(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *cube = PG_GETARG_NDBOX_P(0);
|
|
double x = PG_GETARG_FLOAT8(1);
|
|
NDBOX *result;
|
|
int size;
|
|
int i;
|
|
|
|
if (DIM(cube) + 1 > CUBE_MAX_DIM)
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
|
|
errmsg("can't extend cube"),
|
|
errdetail("A cube cannot have more than %d dimensions.",
|
|
CUBE_MAX_DIM)));
|
|
|
|
if (IS_POINT(cube))
|
|
{
|
|
size = POINT_SIZE((DIM(cube) + 1));
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, DIM(cube) + 1);
|
|
SET_POINT_BIT(result);
|
|
for (i = 0; i < DIM(cube); i++)
|
|
result->x[i] = cube->x[i];
|
|
result->x[DIM(result) - 1] = x;
|
|
}
|
|
else
|
|
{
|
|
size = CUBE_SIZE((DIM(cube) + 1));
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, DIM(cube) + 1);
|
|
for (i = 0; i < DIM(cube); i++)
|
|
{
|
|
result->x[i] = cube->x[i];
|
|
result->x[DIM(result) + i] = cube->x[DIM(cube) + i];
|
|
}
|
|
result->x[DIM(result) - 1] = x;
|
|
result->x[2 * DIM(result) - 1] = x;
|
|
}
|
|
|
|
PG_FREE_IF_COPY(cube, 0);
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|
|
|
|
/* Add a dimension to an existing cube */
|
|
Datum
|
|
cube_c_f8_f8(PG_FUNCTION_ARGS)
|
|
{
|
|
NDBOX *cube = PG_GETARG_NDBOX_P(0);
|
|
double x1 = PG_GETARG_FLOAT8(1);
|
|
double x2 = PG_GETARG_FLOAT8(2);
|
|
NDBOX *result;
|
|
int size;
|
|
int i;
|
|
|
|
if (DIM(cube) + 1 > CUBE_MAX_DIM)
|
|
ereport(ERROR,
|
|
(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
|
|
errmsg("can't extend cube"),
|
|
errdetail("A cube cannot have more than %d dimensions.",
|
|
CUBE_MAX_DIM)));
|
|
|
|
if (IS_POINT(cube) && (x1 == x2))
|
|
{
|
|
size = POINT_SIZE((DIM(cube) + 1));
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, DIM(cube) + 1);
|
|
SET_POINT_BIT(result);
|
|
for (i = 0; i < DIM(cube); i++)
|
|
result->x[i] = cube->x[i];
|
|
result->x[DIM(result) - 1] = x1;
|
|
}
|
|
else
|
|
{
|
|
size = CUBE_SIZE((DIM(cube) + 1));
|
|
result = (NDBOX *) palloc0(size);
|
|
SET_VARSIZE(result, size);
|
|
SET_DIM(result, DIM(cube) + 1);
|
|
for (i = 0; i < DIM(cube); i++)
|
|
{
|
|
result->x[i] = LL_COORD(cube, i);
|
|
result->x[DIM(result) + i] = UR_COORD(cube, i);
|
|
}
|
|
result->x[DIM(result) - 1] = x1;
|
|
result->x[2 * DIM(result) - 1] = x2;
|
|
}
|
|
|
|
PG_FREE_IF_COPY(cube, 0);
|
|
PG_RETURN_NDBOX_P(result);
|
|
}
|