database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-06-13 07:41:39 +03:00
Code Activity

Fix contrib/btree_gist to handle collations properly.

Browse Source 
Make use of the collation attached to the index column, instead of
hard-wiring DEFAULT_COLLATION_OID.  (Note: in theory this could require
reindexing btree_gist indexes on textual columns, but I rather doubt anyone
has one with a non-default declared collation as yet.)
This commit is contained in:
Tom Lane
2011-04-22 20:19:58 -04:00
parent ae20bf1740
commit bb85030630
7 changed files with 219 additions and 187 deletions
Download Patch File Download Diff File Expand all files Collapse all files

155
contrib/btree_gist/btree_utils_var.c
View File

@ -8,11 +8,24 @@
#include <float.h>
#include "btree_utils_var.h"
#include "catalog/pg_collation.h"
#include "utils/pg_locale.h"
#include "utils/builtins.h"
#include "utils/rel.h"
/* used for key sorting */
typedef struct
{
int i;
GBT_VARKEY *t;
} Vsrt;
typedef struct
{
const gbtree_vinfo *tinfo;
Oid collation;
} gbt_vsrt_arg;
PG_FUNCTION_INFO_V1(gbt_var_decompress);
Datum gbt_var_decompress(PG_FUNCTION_ARGS);
@ -140,13 +153,11 @@ gbt_var_node_cp_len(const GBT_VARKEY *node, const gbtree_vinfo *tinfo)
/*
* returns true, if query matches prefix ( common prefix )
*/
*/
static bool
gbt_bytea_pf_match(const bytea *pf, const bytea *query, const gbtree_vinfo *tinfo)
{
bool out = FALSE;
int32 k = 0;
int32 qlen = VARSIZE(query) - VARHDRSZ;
int32 nlen = VARSIZE(pf) - VARHDRSZ;
@ -155,27 +166,7 @@ gbt_bytea_pf_match(const bytea *pf, const bytea *query, const gbtree_vinfo *tinf
char *q = VARDATA(query);
char *n = VARDATA(pf);
if (tinfo->eml > 1)
{
out = (varstr_cmp(q, nlen, n, nlen, DEFAULT_COLLATION_OID) == 0);
}
else
{
out = TRUE;
for (k = 0; k < nlen; k++)
{
if (*n != *q)
{
out = FALSE;
break;
}
if (k < (nlen - 1))
{
q++;
n++;
}
}
}
out = (memcmp(q, n, nlen) == 0);
}
return out;
@ -184,17 +175,14 @@ gbt_bytea_pf_match(const bytea *pf, const bytea *query, const gbtree_vinfo *tinf
/*
* returns true, if query matches node using common prefix
*/
*/
static bool
gbt_var_node_pf_match(const GBT_VARKEY_R *node, const bytea *query, const gbtree_vinfo *tinfo)
{
return (tinfo->trnc && (
gbt_bytea_pf_match(node->lower, query, tinfo) ||
gbt_bytea_pf_match(node->upper, query, tinfo)
));
}
@ -232,9 +220,9 @@ gbt_var_node_truncate(const GBT_VARKEY *node, int32 cpf_length, const gbtree_vin
void
gbt_var_bin_union(Datum *u, GBT_VARKEY *e, const gbtree_vinfo *tinfo)
gbt_var_bin_union(Datum *u, GBT_VARKEY *e, Oid collation,
const gbtree_vinfo *tinfo)
{
GBT_VARKEY *nk = NULL;
GBT_VARKEY *tmp = NULL;
GBT_VARKEY_R nr;
@ -252,14 +240,14 @@ gbt_var_bin_union(Datum *u, GBT_VARKEY *e, const gbtree_vinfo *tinfo)
GBT_VARKEY_R ro = gbt_var_key_readable((GBT_VARKEY *) DatumGetPointer(*u));
if ((*tinfo->f_cmp) ((bytea *) ro.lower, (bytea *) eo.lower) > 0)
if ((*tinfo->f_cmp) (ro.lower, eo.lower, collation) > 0)
{
nr.lower = eo.lower;
nr.upper = ro.upper;
nk = gbt_var_key_copy(&nr, TRUE);
}
if ((*tinfo->f_cmp) ((bytea *) ro.upper, (bytea *) eo.upper) < 0)
if ((*tinfo->f_cmp) (ro.upper, eo.upper, collation) < 0)
{
nr.upper = eo.upper;
nr.lower = ro.lower;
@ -308,7 +296,8 @@ gbt_var_compress(GISTENTRY *entry, const gbtree_vinfo *tinfo)
GBT_VARKEY *
gbt_var_union(const GistEntryVector *entryvec, int32 *size, const gbtree_vinfo *tinfo)
gbt_var_union(const GistEntryVector *entryvec, int32 *size, Oid collation,
const gbtree_vinfo *tinfo)
{
int i = 0,
@ -326,7 +315,7 @@ gbt_var_union(const GistEntryVector *entryvec, int32 *size, const gbtree_vinfo *
for (i = 1; i < numranges; i++)
{
cur = (GBT_VARKEY *) DatumGetPointer(entryvec->vector[i].key);
gbt_var_bin_union(&out, cur, tinfo);
gbt_var_bin_union(&out, cur, collation, tinfo);
}
@ -347,9 +336,10 @@ gbt_var_union(const GistEntryVector *entryvec, int32 *size, const gbtree_vinfo *
bool
gbt_var_same(bool *result, const Datum d1, const Datum d2, const gbtree_vinfo *tinfo)
gbt_var_same(Datum d1, Datum d2, Oid collation,
const gbtree_vinfo *tinfo)
{
bool result;
GBT_VARKEY *t1 = (GBT_VARKEY *) DatumGetPointer(d1);
GBT_VARKEY *t2 = (GBT_VARKEY *) DatumGetPointer(d2);
GBT_VARKEY_R r1,
@ -359,22 +349,19 @@ gbt_var_same(bool *result, const Datum d1, const Datum d2, const gbtree_vinfo *t
r2 = gbt_var_key_readable(t2);
if (t1 && t2)
{
*result = (((*tinfo->f_cmp) ((bytea *) r1.lower, (bytea *) r2.lower) == 0
&& (*tinfo->f_cmp) ((bytea *) r1.upper, (bytea *) r2.upper) == 0) ? TRUE : FALSE);
}
result = ((*tinfo->f_cmp) (r1.lower, r2.lower, collation) == 0 &&
(*tinfo->f_cmp) (r1.upper, r2.upper, collation) == 0);
else
*result = (t1 == NULL && t2 == NULL) ? TRUE : FALSE;
result = (t1 == NULL && t2 == NULL);
PG_RETURN_POINTER(result);
return result;
}
float *
gbt_var_penalty(float *res, const GISTENTRY *o, const GISTENTRY *n, const gbtree_vinfo *tinfo)
gbt_var_penalty(float *res, const GISTENTRY *o, const GISTENTRY *n,
Oid collation, const gbtree_vinfo *tinfo)
{
GBT_VARKEY *orge = (GBT_VARKEY *) DatumGetPointer(o->key);
GBT_VARKEY *newe = (GBT_VARKEY *) DatumGetPointer(n->key);
GBT_VARKEY_R ok,
@ -394,21 +381,19 @@ gbt_var_penalty(float *res, const GISTENTRY *o, const GISTENTRY *n, const gbtree
if ((VARSIZE(ok.lower) - VARHDRSZ) == 0 && (VARSIZE(ok.upper) - VARHDRSZ) == 0)
*res = 0.0;
else if (!(
(
((*tinfo->f_cmp) (nk.lower, ok.lower) >= 0 || gbt_bytea_pf_match(ok.lower, nk.lower, tinfo)) &&
((*tinfo->f_cmp) (nk.upper, ok.upper) <= 0 || gbt_bytea_pf_match(ok.upper, nk.upper, tinfo))
)
))
else if (!(((*tinfo->f_cmp) (nk.lower, ok.lower, collation) >= 0 ||
gbt_bytea_pf_match(ok.lower, nk.lower, tinfo)) &&
((*tinfo->f_cmp) (nk.upper, ok.upper, collation) <= 0 ||
gbt_bytea_pf_match(ok.upper, nk.upper, tinfo))))
{
Datum d = PointerGetDatum(0);
double dres = 0.0;
int32 ol,
ul;
gbt_var_bin_union(&d, orge, tinfo);
gbt_var_bin_union(&d, orge, collation, tinfo);
ol = gbt_var_node_cp_len((GBT_VARKEY *) DatumGetPointer(d), tinfo);
gbt_var_bin_union(&d, newe, tinfo);
gbt_var_bin_union(&d, newe, collation, tinfo);
ul = gbt_var_node_cp_len((GBT_VARKEY *) DatumGetPointer(d), tinfo);
if (ul < ol)
@ -444,18 +429,19 @@ gbt_vsrt_cmp(const void *a, const void *b, void *arg)
{
GBT_VARKEY_R ar = gbt_var_key_readable(((const Vsrt *) a)->t);
GBT_VARKEY_R br = gbt_var_key_readable(((const Vsrt *) b)->t);
const gbtree_vinfo *tinfo = (const gbtree_vinfo *) arg;
const gbt_vsrt_arg *varg = (const gbt_vsrt_arg *) arg;
int res;
res = (*tinfo->f_cmp) (ar.lower, br.lower);
res = (*varg->tinfo->f_cmp) (ar.lower, br.lower, varg->collation);
if (res == 0)
return (*tinfo->f_cmp) (ar.upper, br.upper);
return (*varg->tinfo->f_cmp) (ar.upper, br.upper, varg->collation);
return res;
}
GIST_SPLITVEC *
gbt_var_picksplit(const GistEntryVector *entryvec, GIST_SPLITVEC *v, const gbtree_vinfo *tinfo)
gbt_var_picksplit(const GistEntryVector *entryvec, GIST_SPLITVEC *v,
Oid collation, const gbtree_vinfo *tinfo)
{
OffsetNumber i,
maxoff = entryvec->n - 1;
@ -464,6 +450,7 @@ gbt_var_picksplit(const GistEntryVector *entryvec, GIST_SPLITVEC *v, const gbtre
nbytes;
char *cur;
GBT_VARKEY **sv = NULL;
gbt_vsrt_arg varg;
arr = (Vsrt *) palloc((maxoff + 1) * sizeof(Vsrt));
nbytes = (maxoff + 2) * sizeof(OffsetNumber);
@ -497,11 +484,13 @@ gbt_var_picksplit(const GistEntryVector *entryvec, GIST_SPLITVEC *v, const gbtre
}
/* sort */
varg.tinfo = tinfo;
varg.collation = collation;
qsort_arg((void *) &arr[FirstOffsetNumber],
maxoff - FirstOffsetNumber + 1,
sizeof(Vsrt),
gbt_vsrt_cmp,
(void *) tinfo);
(void *) &varg);
/* We do simply create two parts */
@ -509,13 +498,13 @@ gbt_var_picksplit(const GistEntryVector *entryvec, GIST_SPLITVEC *v, const gbtre
{
if (i <= (maxoff - FirstOffsetNumber + 1) / 2)
{
gbt_var_bin_union(&v->spl_ldatum, arr[i].t, tinfo);
gbt_var_bin_union(&v->spl_ldatum, arr[i].t, collation, tinfo);
v->spl_left[v->spl_nleft] = arr[i].i;
v->spl_nleft++;
}
else
{
gbt_var_bin_union(&v->spl_rdatum, arr[i].t, tinfo);
gbt_var_bin_union(&v->spl_rdatum, arr[i].t, collation, tinfo);
v->spl_right[v->spl_nright] = arr[i].i;
v->spl_nright++;
}
@ -546,63 +535,61 @@ gbt_var_picksplit(const GistEntryVector *entryvec, GIST_SPLITVEC *v, const gbtre
* The GiST consistent method
*/
bool
gbt_var_consistent(
GBT_VARKEY_R *key,
gbt_var_consistent(GBT_VARKEY_R *key,
const void *query,
const StrategyNumber *strategy,
StrategyNumber strategy,
Oid collation,
bool is_leaf,
const gbtree_vinfo *tinfo
)
const gbtree_vinfo *tinfo)
{
bool retval = FALSE;
switch (*strategy)
switch (strategy)
{
case BTLessEqualStrategyNumber:
if (is_leaf)
retval = (*tinfo->f_ge) (query, (void *) key->lower);
retval = (*tinfo->f_ge) (query, key->lower, collation);
else
retval = (*tinfo->f_cmp) ((bytea *) query, key->lower) >= 0
retval = (*tinfo->f_cmp) (query, key->lower, collation) >= 0
|| gbt_var_node_pf_match(key, query, tinfo);
break;
case BTLessStrategyNumber:
if (is_leaf)
retval = (*tinfo->f_gt) (query, (void *) key->lower);
retval = (*tinfo->f_gt) (query, key->lower, collation);
else
retval = (*tinfo->f_cmp) ((bytea *) query, key->lower) >= 0
retval = (*tinfo->f_cmp) (query, key->lower, collation) >= 0
|| gbt_var_node_pf_match(key, query, tinfo);
break;
case BTEqualStrategyNumber:
if (is_leaf)
retval = (*tinfo->f_eq) (query, (void *) key->lower);
retval = (*tinfo->f_eq) (query, key->lower, collation);
else
retval = (
(
(*tinfo->f_cmp) (key->lower, (bytea *) query) <= 0 &&
(*tinfo->f_cmp) ((bytea *) query, (void *) key->upper) <= 0
) || gbt_var_node_pf_match(key, query, tinfo)
);
retval =
((*tinfo->f_cmp) (key->lower, query, collation) <= 0 &&
(*tinfo->f_cmp) (query, key->upper, collation) <= 0) ||
gbt_var_node_pf_match(key, query, tinfo);
break;
case BTGreaterStrategyNumber:
if (is_leaf)
retval = (*tinfo->f_lt) (query, (void *) key->upper);
retval = (*tinfo->f_lt) (query, key->upper, collation);
else
retval = (*tinfo->f_cmp) ((bytea *) query, key->upper) <= 0
retval = (*tinfo->f_cmp) (query, key->upper, collation) <= 0
|| gbt_var_node_pf_match(key, query, tinfo);
break;
case BTGreaterEqualStrategyNumber:
if (is_leaf)
retval = (*tinfo->f_le) (query, (void *) key->upper);
retval = (*tinfo->f_le) (query, key->upper, collation);
else
retval = (*tinfo->f_cmp) ((bytea *) query, key->upper) <= 0
retval = (*tinfo->f_cmp) (query, key->upper, collation) <= 0
|| gbt_var_node_pf_match(key, query, tinfo);
break;
case BtreeGistNotEqualStrategyNumber:
retval = !((*tinfo->f_eq) (query, key->lower) && (*tinfo->f_eq) (query, key->upper));
retval = !((*tinfo->f_eq) (query, key->lower, collation) &&
(*tinfo->f_eq) (query, key->upper, collation));
break;
default:
retval = FALSE;
}
return (retval);
return retval;
}