database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-07-12 21:01:52 +03:00
Code Activity

Per-column collation support

Browse Source 
This adds collation support for columns and domains, a COLLATE clause
to override it per expression, and B-tree index support.

Peter Eisentraut
reviewed by Pavel Stehule, Itagaki Takahiro, Robert Haas, Noah Misch
This commit is contained in:
Peter Eisentraut
2011-02-08 23:04:18 +02:00
parent 1703f0e8da
commit 414c5a2ea6
156 changed files with 4519 additions and 582 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
src/backend/optimizer/path/costsize.c
View File

@ -1795,6 +1795,7 @@ cost_mergejoin(MergePath *path, PlannerInfo *root, SpecialJoinInfo *sjinfo)
ipathkey = (PathKey *) linitial(ipathkeys);
/* debugging check */
if (opathkey->pk_opfamily != ipathkey->pk_opfamily ||
opathkey->pk_collation != ipathkey->pk_collation ||
opathkey->pk_strategy != ipathkey->pk_strategy ||
opathkey->pk_nulls_first != ipathkey->pk_nulls_first)
elog(ERROR, "left and right pathkeys do not match in mergejoin");
@ -2045,6 +2046,7 @@ cached_scansel(PlannerInfo *root, RestrictInfo *rinfo, PathKey *pathkey)
{
cache = (MergeScanSelCache *) lfirst(lc);
if (cache->opfamily == pathkey->pk_opfamily &&
cache->collation == pathkey->pk_collation &&
cache->strategy == pathkey->pk_strategy &&
cache->nulls_first == pathkey->pk_nulls_first)
return cache;
@ -2054,6 +2056,7 @@ cached_scansel(PlannerInfo *root, RestrictInfo *rinfo, PathKey *pathkey)
mergejoinscansel(root,
(Node *) rinfo->clause,
pathkey->pk_opfamily,
pathkey->pk_collation,
pathkey->pk_strategy,
pathkey->pk_nulls_first,
&leftstartsel,
@ -2066,6 +2069,7 @@ cached_scansel(PlannerInfo *root, RestrictInfo *rinfo, PathKey *pathkey)
cache = (MergeScanSelCache *) palloc(sizeof(MergeScanSelCache));
cache->opfamily = pathkey->pk_opfamily;
cache->collation = pathkey->pk_collation;
cache->strategy = pathkey->pk_strategy;
cache->nulls_first = pathkey->pk_nulls_first;
cache->leftstartsel = leftstartsel;

75
src/backend/optimizer/path/indxpath.c
View File

@ -23,6 +23,7 @@
#include "catalog/pg_opfamily.h"
#include "catalog/pg_type.h"
#include "nodes/makefuncs.h"
#include "nodes/nodeFuncs.h"
#include "optimizer/clauses.h"
#include "optimizer/cost.h"
#include "optimizer/pathnode.h"
@ -99,15 +100,15 @@ static List *find_clauses_for_join(PlannerInfo *root, RelOptInfo *rel,
Relids outer_relids, bool isouterjoin);
static bool match_boolean_index_clause(Node *clause, int indexcol,
IndexOptInfo *index);
static bool match_special_index_operator(Expr *clause, Oid opfamily,
static bool match_special_index_operator(Expr *clause, Oid idxcolcollation, Oid opfamily,
bool indexkey_on_left);
static Expr *expand_boolean_index_clause(Node *clause, int indexcol,
IndexOptInfo *index);
static List *expand_indexqual_opclause(RestrictInfo *rinfo, Oid opfamily);
static List *expand_indexqual_opclause(RestrictInfo *rinfo, Oid opfamily, Oid collation);
static RestrictInfo *expand_indexqual_rowcompare(RestrictInfo *rinfo,
IndexOptInfo *index,
int indexcol);
static List *prefix_quals(Node *leftop, Oid opfamily,
static List *prefix_quals(Node *leftop, Oid opfamily, Oid collation,
Const *prefix, Pattern_Prefix_Status pstatus);
static List *network_prefix_quals(Node *leftop, Oid expr_op, Oid opfamily,
Datum rightop);
@ -1142,7 +1143,9 @@ group_clauses_by_indexkey(IndexOptInfo *index,
* and
* (2) must contain an operator which is in the same family as the index
* operator for this column, or is a "special" operator as recognized
* by match_special_index_operator().
* by match_special_index_operator();
* and
* (3) must match the collation of the index.
*
* Our definition of "const" is pretty liberal: we allow Vars belonging
* to the caller-specified outer_relids relations (which had better not
@ -1198,6 +1201,7 @@ match_clause_to_indexcol(IndexOptInfo *index,
SaOpControl saop_control)
{
Expr *clause = rinfo->clause;
Oid collation = index->indexcollations[indexcol];
Oid opfamily = index->opfamily[indexcol];
Node *leftop,
*rightop;
@ -1280,7 +1284,8 @@ match_clause_to_indexcol(IndexOptInfo *index,
bms_is_subset(right_relids, outer_relids) &&
!contain_volatile_functions(rightop))
{
if (is_indexable_operator(expr_op, opfamily, true))
if (is_indexable_operator(expr_op, opfamily, true) &&
(!collation || collation == exprCollation((Node *) clause)))
return true;
/*
@ -1288,7 +1293,7 @@ match_clause_to_indexcol(IndexOptInfo *index,
* is a "special" indexable operator.
*/
if (plain_op &&
match_special_index_operator(clause, opfamily, true))
match_special_index_operator(clause, collation, opfamily, true))
return true;
return false;
}
@ -1298,14 +1303,15 @@ match_clause_to_indexcol(IndexOptInfo *index,
bms_is_subset(left_relids, outer_relids) &&
!contain_volatile_functions(leftop))
{
if (is_indexable_operator(expr_op, opfamily, false))
if (is_indexable_operator(expr_op, opfamily, false) &&
(!collation || collation == exprCollation((Node *) clause)))
return true;
/*
* If we didn't find a member of the index's opfamily, see whether it
* is a "special" indexable operator.
*/
if (match_special_index_operator(clause, opfamily, false))
if (match_special_index_operator(clause, collation, opfamily, false))
return true;
return false;
}
@ -1391,6 +1397,9 @@ match_rowcompare_to_indexcol(IndexOptInfo *index,
else
return false;
if (index->indexcollations[indexcol] != linitial_oid(clause->collids))
return false;
/* We're good if the operator is the right type of opfamily member */
switch (get_op_opfamily_strategy(expr_op, opfamily))
{
@ -2380,7 +2389,7 @@ match_boolean_index_clause(Node *clause,
* Return 'true' if we can do something with it anyway.
*/
static bool
match_special_index_operator(Expr *clause, Oid opfamily,
match_special_index_operator(Expr *clause, Oid idxcolcollation, Oid opfamily,
bool indexkey_on_left)
{
bool isIndexable = false;
@ -2495,7 +2504,7 @@ match_special_index_operator(Expr *clause, Oid opfamily,
isIndexable =
(opfamily == TEXT_PATTERN_BTREE_FAM_OID) ||
(opfamily == TEXT_BTREE_FAM_OID &&
(pstatus == Pattern_Prefix_Exact || lc_collate_is_c()));
(pstatus == Pattern_Prefix_Exact || lc_collate_is_c(idxcolcollation)));
break;
case OID_BPCHAR_LIKE_OP:
@ -2505,7 +2514,7 @@ match_special_index_operator(Expr *clause, Oid opfamily,
isIndexable =
(opfamily == BPCHAR_PATTERN_BTREE_FAM_OID) ||
(opfamily == BPCHAR_BTREE_FAM_OID &&
(pstatus == Pattern_Prefix_Exact || lc_collate_is_c()));
(pstatus == Pattern_Prefix_Exact || lc_collate_is_c(idxcolcollation)));
break;
case OID_NAME_LIKE_OP:
@ -2526,6 +2535,25 @@ match_special_index_operator(Expr *clause, Oid opfamily,
break;
}
if (!isIndexable)
return false;
/*
* For case-insensitive matching, we also need to check that the
* collations match.
*/
switch (expr_op)
{
case OID_TEXT_ICLIKE_OP:
case OID_TEXT_ICREGEXEQ_OP:
case OID_BPCHAR_ICLIKE_OP:
case OID_BPCHAR_ICREGEXEQ_OP:
case OID_NAME_ICLIKE_OP:
case OID_NAME_ICREGEXEQ_OP:
isIndexable = (idxcolcollation == exprCollation((Node *) clause));
break;
}
return isIndexable;
}
@ -2561,6 +2589,7 @@ expand_indexqual_conditions(IndexOptInfo *index, List *clausegroups)
{
List *clausegroup = (List *) lfirst(lc);
Oid curFamily = index->opfamily[indexcol];
Oid curCollation = index->indexcollations[indexcol];
ListCell *lc2;
foreach(lc2, clausegroup)
@ -2592,7 +2621,8 @@ expand_indexqual_conditions(IndexOptInfo *index, List *clausegroups)
{
resultquals = list_concat(resultquals,
expand_indexqual_opclause(rinfo,
curFamily));
curFamily,
curCollation));
}
else if (IsA(clause, ScalarArrayOpExpr))
{
@ -2693,7 +2723,7 @@ expand_boolean_index_clause(Node *clause,
* expand special cases that were accepted by match_special_index_operator().
*/
static List *
expand_indexqual_opclause(RestrictInfo *rinfo, Oid opfamily)
expand_indexqual_opclause(RestrictInfo *rinfo, Oid opfamily, Oid collation)
{
Expr *clause = rinfo->clause;
@ -2724,7 +2754,7 @@ expand_indexqual_opclause(RestrictInfo *rinfo, Oid opfamily)
{
pstatus = pattern_fixed_prefix(patt, Pattern_Type_Like,
&prefix, &rest);
return prefix_quals(leftop, opfamily, prefix, pstatus);
return prefix_quals(leftop, opfamily, collation, prefix, pstatus);
}
break;
@ -2736,7 +2766,7 @@ expand_indexqual_opclause(RestrictInfo *rinfo, Oid opfamily)
/* the right-hand const is type text for all of these */
pstatus = pattern_fixed_prefix(patt, Pattern_Type_Like_IC,
&prefix, &rest);
return prefix_quals(leftop, opfamily, prefix, pstatus);
return prefix_quals(leftop, opfamily, collation, prefix, pstatus);
}
break;
@ -2748,7 +2778,7 @@ expand_indexqual_opclause(RestrictInfo *rinfo, Oid opfamily)
/* the right-hand const is type text for all of these */
pstatus = pattern_fixed_prefix(patt, Pattern_Type_Regex,
&prefix, &rest);
return prefix_quals(leftop, opfamily, prefix, pstatus);
return prefix_quals(leftop, opfamily, collation, prefix, pstatus);
}
break;
@ -2760,7 +2790,7 @@ expand_indexqual_opclause(RestrictInfo *rinfo, Oid opfamily)
/* the right-hand const is type text for all of these */
pstatus = pattern_fixed_prefix(patt, Pattern_Type_Regex_IC,
&prefix, &rest);
return prefix_quals(leftop, opfamily, prefix, pstatus);
return prefix_quals(leftop, opfamily, collation, prefix, pstatus);
}
break;
@ -2814,6 +2844,7 @@ expand_indexqual_rowcompare(RestrictInfo *rinfo,
ListCell *largs_cell;
ListCell *rargs_cell;
ListCell *opnos_cell;
ListCell *collids_cell;
/* We have to figure out (again) how the first col matches */
var_on_left = match_index_to_operand((Node *) linitial(clause->largs),
@ -2845,6 +2876,7 @@ expand_indexqual_rowcompare(RestrictInfo *rinfo,
largs_cell = lnext(list_head(clause->largs));
rargs_cell = lnext(list_head(clause->rargs));
opnos_cell = lnext(list_head(clause->opnos));
collids_cell = lnext(list_head(clause->collids));
while (largs_cell != NULL)
{
@ -2891,6 +2923,10 @@ expand_indexqual_rowcompare(RestrictInfo *rinfo,
!= op_strategy)
break;
/* Does collation match? */
if (lfirst_oid(collids_cell) != index->indexcollations[i])
break;
/* Add opfamily and datatypes to lists */
get_op_opfamily_properties(expr_op, index->opfamily[i], false,
&op_strategy,
@ -2974,6 +3010,8 @@ expand_indexqual_rowcompare(RestrictInfo *rinfo,
rc->opnos = new_ops;
rc->opfamilies = list_truncate(list_copy(clause->opfamilies),
matching_cols);
rc->collids = list_truncate(list_copy(clause->collids),
matching_cols);
rc->largs = list_truncate((List *) copyObject(clause->largs),
matching_cols);
rc->rargs = list_truncate((List *) copyObject(clause->rargs),
@ -2998,7 +3036,7 @@ expand_indexqual_rowcompare(RestrictInfo *rinfo,
* operators and operand datatypes.
*/
static List *
prefix_quals(Node *leftop, Oid opfamily,
prefix_quals(Node *leftop, Oid opfamily, Oid collation,
Const *prefix_const, Pattern_Prefix_Status pstatus)
{
List *result;
@ -3100,6 +3138,7 @@ prefix_quals(Node *leftop, Oid opfamily,
if (oproid == InvalidOid)
elog(ERROR, "no < operator for opfamily %u", opfamily);
fmgr_info(get_opcode(oproid), &ltproc);
fmgr_info_collation(collation, &ltproc);
greaterstr = make_greater_string(prefix_const, &ltproc);
if (greaterstr)
{

32
src/backend/optimizer/path/pathkeys.c
View File

@ -18,6 +18,7 @@
#include "postgres.h"
#include "access/skey.h"
#include "catalog/pg_collation.h"
#include "catalog/pg_type.h"
#include "nodes/makefuncs.h"
#include "nodes/nodeFuncs.h"
@ -30,10 +31,10 @@
#include "utils/lsyscache.h"
static PathKey *makePathKey(EquivalenceClass *eclass, Oid opfamily,
static PathKey *makePathKey(EquivalenceClass *eclass, Oid opfamily, Oid collation,
int strategy, bool nulls_first);
static PathKey *make_canonical_pathkey(PlannerInfo *root,
EquivalenceClass *eclass, Oid opfamily,
EquivalenceClass *eclass, Oid opfamily, Oid collation,
int strategy, bool nulls_first);
static bool pathkey_is_redundant(PathKey *new_pathkey, List *pathkeys);
static Var *find_indexkey_var(PlannerInfo *root, RelOptInfo *rel,
@ -53,13 +54,14 @@ static bool right_merge_direction(PlannerInfo *root, PathKey *pathkey);
* convenience routine to build the specified node.
*/
static PathKey *
makePathKey(EquivalenceClass *eclass, Oid opfamily,
makePathKey(EquivalenceClass *eclass, Oid opfamily, Oid collation,
int strategy, bool nulls_first)
{
PathKey *pk = makeNode(PathKey);
pk->pk_eclass = eclass;
pk->pk_opfamily = opfamily;
pk->pk_collation = collation;
pk->pk_strategy = strategy;
pk->pk_nulls_first = nulls_first;
@ -77,7 +79,7 @@ makePathKey(EquivalenceClass *eclass, Oid opfamily,
*/
static PathKey *
make_canonical_pathkey(PlannerInfo *root,
EquivalenceClass *eclass, Oid opfamily,
EquivalenceClass *eclass, Oid opfamily, Oid collation,
int strategy, bool nulls_first)
{
PathKey *pk;
@ -93,6 +95,7 @@ make_canonical_pathkey(PlannerInfo *root,
pk = (PathKey *) lfirst(lc);
if (eclass == pk->pk_eclass &&
opfamily == pk->pk_opfamily &&
collation == pk->pk_collation &&
strategy == pk->pk_strategy &&
nulls_first == pk->pk_nulls_first)
return pk;
@ -104,7 +107,7 @@ make_canonical_pathkey(PlannerInfo *root,
*/
oldcontext = MemoryContextSwitchTo(root->planner_cxt);
pk = makePathKey(eclass, opfamily, strategy, nulls_first);
pk = makePathKey(eclass, opfamily, collation, strategy, nulls_first);
root->canon_pathkeys = lappend(root->canon_pathkeys, pk);
MemoryContextSwitchTo(oldcontext);
@ -206,6 +209,7 @@ canonicalize_pathkeys(PlannerInfo *root, List *pathkeys)
cpathkey = make_canonical_pathkey(root,
eclass,
pathkey->pk_opfamily,
pathkey->pk_collation,
pathkey->pk_strategy,
pathkey->pk_nulls_first);
@ -247,6 +251,7 @@ make_pathkey_from_sortinfo(PlannerInfo *root,
Oid equality_op;
List *opfamilies;
EquivalenceClass *eclass;
Oid collation;
strategy = reverse_sort ? BTGreaterStrategyNumber : BTLessStrategyNumber;
@ -301,12 +306,14 @@ make_pathkey_from_sortinfo(PlannerInfo *root,
if (!eclass)
return NULL;
collation = exprCollation((Node *) expr);
/* And finally we can find or create a PathKey node */
if (canonicalize)
return make_canonical_pathkey(root, eclass, opfamily,
return make_canonical_pathkey(root, eclass, opfamily, collation,
strategy, nulls_first);
else
return makePathKey(eclass, opfamily, strategy, nulls_first);
return makePathKey(eclass, opfamily, collation, strategy, nulls_first);
}
/*
@ -605,7 +612,8 @@ find_indexkey_var(PlannerInfo *root, RelOptInfo *rel, AttrNumber varattno)
ListCell *temp;
Index relid;
Oid reloid,
vartypeid;
vartypeid,
varcollid;
int32 type_mod;
foreach(temp, rel->reltargetlist)
@ -620,8 +628,9 @@ find_indexkey_var(PlannerInfo *root, RelOptInfo *rel, AttrNumber varattno)
relid = rel->relid;
reloid = getrelid(relid, root->parse->rtable);
get_atttypetypmod(reloid, varattno, &vartypeid, &type_mod);
varcollid = get_attcollation(reloid, varattno);
return makeVar(relid, varattno, vartypeid, type_mod, 0);
return makeVar(relid, varattno, vartypeid, type_mod, varcollid, 0);
}
/*
@ -703,6 +712,7 @@ convert_subquery_pathkeys(PlannerInfo *root, RelOptInfo *rel,
make_canonical_pathkey(root,
outer_ec,
sub_pathkey->pk_opfamily,
sub_pathkey->pk_collation,
sub_pathkey->pk_strategy,
sub_pathkey->pk_nulls_first);
}
@ -805,6 +815,7 @@ convert_subquery_pathkeys(PlannerInfo *root, RelOptInfo *rel,
outer_pk = make_canonical_pathkey(root,
outer_ec,
sub_pathkey->pk_opfamily,
sub_pathkey->pk_collation,
sub_pathkey->pk_strategy,
sub_pathkey->pk_nulls_first);
/* score = # of equivalence peers */
@ -1326,6 +1337,7 @@ select_outer_pathkeys_for_merge(PlannerInfo *root,
pathkey = make_canonical_pathkey(root,
ec,
linitial_oid(ec->ec_opfamilies),
DEFAULT_COLLATION_OID,
BTLessStrategyNumber,
false);
/* can't be redundant because no duplicate ECs */
@ -1419,6 +1431,7 @@ make_inner_pathkeys_for_merge(PlannerInfo *root,
pathkey = make_canonical_pathkey(root,
ieclass,
opathkey->pk_opfamily,
opathkey->pk_collation,
opathkey->pk_strategy,
opathkey->pk_nulls_first);
@ -1539,6 +1552,7 @@ right_merge_direction(PlannerInfo *root, PathKey *pathkey)
PathKey *query_pathkey = (PathKey *) lfirst(l);
if (pathkey->pk_eclass == query_pathkey->pk_eclass &&
pathkey->pk_collation == query_pathkey->pk_collation &&
pathkey->pk_opfamily == query_pathkey->pk_opfamily)
{
/*