mirror of
https://github.com/postgres/postgres.git
synced 2026-01-05 23:38:41 +03:00
Change the planner to allow indexscan qualification clauses to use
nonconsecutive columns of a multicolumn index, as per discussion around mid-May (pghackers thread "Best way to scan on-disk bitmaps"). This turns out to require only minimal changes in btree, and so far as I can see none at all in GiST. btcostestimate did need some work, but its original assumption that index selectivity == heap selectivity was quite bogus even before this.
This commit is contained in:
@@ -7,7 +7,7 @@
|
||||
* Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
|
||||
* Portions Copyright (c) 1994, Regents of the University of California
|
||||
*
|
||||
* $PostgreSQL: pgsql/src/include/nodes/relation.h,v 1.114 2005/06/10 03:32:25 tgl Exp $
|
||||
* $PostgreSQL: pgsql/src/include/nodes/relation.h,v 1.115 2005/06/13 23:14:49 tgl Exp $
|
||||
*
|
||||
*-------------------------------------------------------------------------
|
||||
*/
|
||||
@@ -328,6 +328,7 @@ typedef struct IndexOptInfo
|
||||
|
||||
bool predOK; /* true if predicate matches query */
|
||||
bool unique; /* true if a unique index */
|
||||
bool amoptionalkey; /* can query omit key for the first column? */
|
||||
} IndexOptInfo;
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user