rdelaage
/
postgresql
mirror of https://git.postgresql.org/git/postgresql.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
postgresql/src/backend/commands/indexcmds.c

950 lines
27 KiB
C
Raw Normal View History

Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
/*-------------------------------------------------------------------------
*
Defend against > INDEX_MAX_KEYS keys in an index.
2000-01-12 06:04:42 +01:00
* indexcmds.c
Disallow non-cachable functions in functional indexes and in index predicates. Per suggestion from Hiroshi.
2001-07-17 23:53:01 +02:00
* POSTGRES define and remove index code.
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*
Update copyrights to 2003.
2003-08-04 04:40:20 +02:00
* Portions Copyright (c) 1996-2003, PostgreSQL Global Development Group
Add: * Portions Copyright (c) 1996-2000, PostgreSQL, Inc to all files copyright Regents of Berkeley. Man, that's a lot of files.
2000-01-26 06:58:53 +01:00
* Portions Copyright (c) 1994, Regents of the University of California
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*
*
* IDENTIFICATION
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
* $PostgreSQL: pgsql/src/backend/commands/indexcmds.c,v 1.120 2004/05/26 04:41:11 neilc Exp $
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*
*-------------------------------------------------------------------------
*/
Another first pass at cleaning up the #includes
1996-11-04 00:57:43 +01:00
Change #include's to use <> and "" as appropriate.
1999-07-16 01:04:24 +02:00
#include "postgres.h"
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
Change #include's to use <> and "" as appropriate.
1999-07-16 01:04:24 +02:00
#include "access/heapam.h"
Make toast-table creation and deletion work somewhat reliably. Don't go through pg_exec_query_dest(), but directly to the execution routines. Also, extend parameter lists so that there's no need to change the global setting of allowSystemTableMods, a hack that was certain to cause trouble in the event of any error.
2000-07-04 08:11:54 +02:00
#include "catalog/catalog.h"
Implement reindex command
2000-02-18 10:30:20 +01:00
#include "catalog/catname.h"
Second phase of committing Rod Taylor's pg_depend/pg_constraint patch. pg_relcheck is gone; CHECK, UNIQUE, PRIMARY KEY, and FOREIGN KEY constraints all have real live entries in pg_constraint. pg_depend exists, and RESTRICT/CASCADE options work on most kinds of DROP; however, pg_depend is not yet very well populated with dependencies. (Most of the ones that are present at this point just replace formerly hardwired associations, such as the implicit drop of a relation's pg_type entry when the relation is dropped.) Need to add more logic to create dependency entries, improve pg_dump to dump constraints in place of indexes and triggers, and add some regression tests.
2002-07-12 20:43:19 +02:00
#include "catalog/dependency.h"
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
#include "catalog/heap.h"
Change #include's to use <> and "" as appropriate.
1999-07-16 01:04:24 +02:00
#include "catalog/index.h"
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
#include "catalog/namespace.h"
Final cleanup.
1999-07-16 07:00:38 +02:00
#include "catalog/pg_opclass.h"
Divide functions into three volatility classes (immutable, stable, and volatile), rather than the old cachable/noncachable distinction. This allows indexscan optimizations in many places where we formerly didn't. Also, add a pronamespace column to pg_proc (it doesn't do anything yet, however).
2002-04-05 02:31:36 +02:00
#include "catalog/pg_proc.h"
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
#include "commands/dbcommands.h"
Change #include's to use <> and "" as appropriate.
1999-07-16 01:04:24 +02:00
#include "commands/defrem.h"
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
#include "commands/tablecmds.h"
Phase 3 of read-only-plans project: ExecInitExpr now builds expression execution state trees, and ExecEvalExpr takes an expression state tree not an expression plan tree. The plan tree is now read-only as far as the executor is concerned. Next step is to begin actually exploiting this property.
2002-12-13 20:46:01 +01:00
#include "executor/executor.h"
Make toast-table creation and deletion work somewhat reliably. Don't go through pg_exec_query_dest(), but directly to the execution routines. Also, extend parameter lists so that there's no need to change the global setting of allowSystemTableMods, a hack that was certain to cause trouble in the event of any error.
2000-07-04 08:11:54 +02:00
#include "miscadmin.h"
Change #include's to use <> and "" as appropriate.
1999-07-16 01:04:24 +02:00
#include "optimizer/clauses.h"
Final cleanup.
1999-07-16 07:00:38 +02:00
#include "optimizer/prep.h"
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
#include "parser/analyze.h"
Final cleanup.
1999-07-16 07:00:38 +02:00
#include "parser/parsetree.h"
Check that user-specified opclass in CREATE INDEX corresponds to operators that will actually work on the column datatype.
2000-04-25 04:45:54 +02:00
#include "parser/parse_coerce.h"
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
#include "parser/parse_expr.h"
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
#include "parser/parse_func.h"
Restructure aclcheck error reporting to make permission-failure messages more uniform and internationalizable: the global array aclcheck_error_strings[] is gone in favor of a subroutine aclcheck_error(). Partial implementation of namespace-related permission checks --- not all done yet.
2002-04-27 05:45:03 +02:00
#include "utils/acl.h"
Final cleanup.
1999-07-16 07:00:38 +02:00
#include "utils/builtins.h"
Disallow non-cachable functions in functional indexes and in index predicates. Per suggestion from Hiroshi.
2001-07-17 23:53:01 +02:00
#include "utils/lsyscache.h"
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
#include "utils/relcache.h"
Final cleanup.
1999-07-16 07:00:38 +02:00
#include "utils/syscache.h"
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
Disallow non-cachable functions in functional indexes and in index predicates. Per suggestion from Hiroshi.
2001-07-17 23:53:01 +02:00
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
/* non-export function prototypes */
Clean up the usage of canonicalize_qual(): in particular, be consistent about whether it is applied before or after eval_const_expressions(). I believe there were some corner cases where the system would fail to recognize that a partial index is applicable because of the previous inconsistency. Store normal rather than 'implicit AND' representations of constraints and index predicates in the catalogs. initdb forced due to representation change of constraints/predicates.
2003-12-28 22:57:37 +01:00
static void CheckPredicate(Expr *predicate);
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
static void ComputeIndexAttrs(IndexInfo *indexInfo, Oid *classOidP,
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
List *attList,
Oid relId,
char *accessMethodName, Oid accessMethodId,
bool isconstraint);
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
static Oid GetIndexOpClass(List *opclass, Oid attrType,
pgindent run.
2003-08-04 02:43:34 +02:00
char *accessMethodName, Oid accessMethodId);
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
static Oid GetDefaultOpClass(Oid attrType, Oid accessMethodId);
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
static char *CreateIndexName(const char *table_name, const char *column_name,
const char *label, Oid inamespace);
static bool relationHasPrimaryKey(Relation rel);
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
/*
pgindent run over code.
1999-05-25 18:15:34 +02:00
* DefineIndex
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
* Creates a new index.
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
* 'heapRelation': the relation the index will apply to.
* 'indexRelationName': the name for the new index, or NULL to indicate
* that a nonconflicting default name should be picked.
* 'accessMethodName': name of the AM to use.
* 'attributeList': a list of IndexElem specifying columns and expressions
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
* to index on.
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
* 'predicate': the partial-index condition, or NULL if none.
* 'rangetable': needed to interpret the predicate.
* 'unique': make the index enforce uniqueness.
* 'primary': mark the index as a primary key in the catalogs.
* 'isconstraint': index is for a PRIMARY KEY or UNIQUE constraint,
* so build a pg_constraint entry for it.
* 'is_alter_table': this is due to an ALTER rather than a CREATE operation.
* 'check_rights': check for CREATE rights in the namespace. (This should
* be true except when ALTER is deleting/recreating an index.)
* 'skip_build': make the catalog entries but leave the index file empty;
* it will be filled later.
* 'quiet': suppress the NOTICE chatter ordinarily provided for constraints.
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*/
void
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
DefineIndex(RangeVar *heapRelation,
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
char *indexRelationName,
char *accessMethodName,
Used modified version of indent that understands over 100 typedefs.
1997-09-08 23:56:23 +02:00
List *attributeList,
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
Expr *predicate,
List *rangetable,
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
bool unique,
The following patch finishes primary key support. Previously, when a field was labelled as a primary key, the system automatically created a unique index on the field. This patch extends it so that the index has the indisprimary field set. You can pull a list of primary keys with the followiing select. SELECT pg_class.relname, pg_attribute.attname FROM pg_class, pg_attribute, pg_index WHERE pg_class.oid = pg_attribute.attrelid AND pg_class.oid = pg_index.indrelid AND pg_index.indkey[0] = pg_attribute.attnum AND pg_index.indisunique = 't'; There is nothing in this patch that modifies the template database to set the indisprimary attribute for system tables. Should they be changed or should we only be concerned with user tables? D'Arcy
1999-01-21 23:48:20 +01:00
bool primary,
Second phase of committing Rod Taylor's pg_depend/pg_constraint patch. pg_relcheck is gone; CHECK, UNIQUE, PRIMARY KEY, and FOREIGN KEY constraints all have real live entries in pg_constraint. pg_depend exists, and RESTRICT/CASCADE options work on most kinds of DROP; however, pg_depend is not yet very well populated with dependencies. (Most of the ones that are present at this point just replace formerly hardwired associations, such as the implicit drop of a relation's pg_type entry when the relation is dropped.) Need to add more logic to create dependency entries, improve pg_dump to dump constraints in place of indexes and triggers, and add some regression tests.
2002-07-12 20:43:19 +02:00
bool isconstraint,
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
bool is_alter_table,
bool check_rights,
bool skip_build,
bool quiet)
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
{
Another PGINDENT run that changes variable indenting and case label indenting. Also static variable indenting.
1997-09-08 04:41:22 +02:00
Oid *classObjectId;
Oid accessMethodId;
Oid relationId;
Restructure aclcheck error reporting to make permission-failure messages more uniform and internationalizable: the global array aclcheck_error_strings[] is gone in favor of a subroutine aclcheck_error(). Partial implementation of namespace-related permission checks --- not all done yet.
2002-04-27 05:45:03 +02:00
Oid namespaceId;
Require ownership permission for CREATE INDEX, per bug report. Disallow CREATE INDEX on system catalogs, non-tables (views, sequences, etc). Disallow CREATE/DROP TRIGGER on system catalogs, non-tables. Disallow ALTER TABLE ADD/DROP CONSTRAINT on system catalogs. Disallow FOREIGN KEY reference to non-table. None of these things can actually work in the present system structure, but the code was letting them pass without complaint.
2002-01-04 00:21:32 +01:00
Relation rel;
Restructure index AM interface for index building and index tuple deletion, per previous discussion on pghackers. Most of the duplicate code in different AMs' ambuild routines has been moved out to a common routine in index.c; this means that all index types now do the right things about inserting recently-dead tuples, etc. (I also removed support for EXTEND INDEX in the ambuild routines, since that's about to go away anyway, and it cluttered the code a lot.) The retail indextuple deletion routines have been replaced by a "bulk delete" routine in which the indexscan is inside the access method. I haven't pushed this change as far as it should go yet, but it should allow considerable simplification of the internal bookkeeping for deletions. Also, add flag columns to pg_am to eliminate various hardcoded tests on AM OIDs, and remove unused pg_am columns. Fix rtree and gist index types to not attempt to store NULLs; before this, gist usually crashed, while rtree managed not to crash but computed wacko bounding boxes for NULL entries (which might have had something to do with the performance problems we've heard about occasionally). Add AtEOXact routines to hash, rtree, and gist, all of which have static state that needs to be reset after an error. We discovered this need long ago for btree, but missed the other guys. Oh, one more thing: concurrent VACUUM is now the default.
2001-07-16 00:48:19 +02:00
HeapTuple tuple;
Form_pg_am accessMethodForm;
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
IndexInfo *indexInfo;
Another PGINDENT run that changes variable indenting and case label indenting. Also static variable indenting.
1997-09-08 04:41:22 +02:00
int numberOfAttributes;
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
/*
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
* count attributes in index
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
*/
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
numberOfAttributes = list_length(attributeList);
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
if (numberOfAttributes <= 0)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
Message editing: remove gratuitous variations in message wording, standardize terms, add some clarifications, fix some untranslatable attempts at dynamic message building.
2003-09-25 08:58:07 +02:00
errmsg("must specify at least one column")));
Defend against > INDEX_MAX_KEYS keys in an index.
2000-01-12 06:04:42 +01:00
if (numberOfAttributes > INDEX_MAX_KEYS)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_TOO_MANY_COLUMNS),
Message editing: remove gratuitous variations in message wording, standardize terms, add some clarifications, fix some untranslatable attempts at dynamic message building.
2003-09-25 08:58:07 +02:00
errmsg("cannot use more than %d columns in an index",
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
INDEX_MAX_KEYS)));
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
/*
Require ownership permission for CREATE INDEX, per bug report. Disallow CREATE INDEX on system catalogs, non-tables (views, sequences, etc). Disallow CREATE/DROP TRIGGER on system catalogs, non-tables. Disallow ALTER TABLE ADD/DROP CONSTRAINT on system catalogs. Disallow FOREIGN KEY reference to non-table. None of these things can actually work in the present system structure, but the code was letting them pass without complaint.
2002-01-04 00:21:32 +01:00
* Open heap relation, acquire a suitable lock on it, remember its OID
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
*/
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
rel = heap_openrv(heapRelation, ShareLock);
Require ownership permission for CREATE INDEX, per bug report. Disallow CREATE INDEX on system catalogs, non-tables (views, sequences, etc). Disallow CREATE/DROP TRIGGER on system catalogs, non-tables. Disallow ALTER TABLE ADD/DROP CONSTRAINT on system catalogs. Disallow FOREIGN KEY reference to non-table. None of these things can actually work in the present system structure, but the code was letting them pass without complaint.
2002-01-04 00:21:32 +01:00
/* Note: during bootstrap may see uncataloged relation */
if (rel->rd_rel->relkind != RELKIND_RELATION &&
rel->rd_rel->relkind != RELKIND_UNCATALOGED)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
Eliminate another gratuitous message wording difference.
2003-09-29 18:37:29 +02:00
errmsg("\"%s\" is not a table",
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
heapRelation->relname)));
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
Require ownership permission for CREATE INDEX, per bug report. Disallow CREATE INDEX on system catalogs, non-tables (views, sequences, etc). Disallow CREATE/DROP TRIGGER on system catalogs, non-tables. Disallow ALTER TABLE ADD/DROP CONSTRAINT on system catalogs. Disallow FOREIGN KEY reference to non-table. None of these things can actually work in the present system structure, but the code was letting them pass without complaint.
2002-01-04 00:21:32 +01:00
relationId = RelationGetRelid(rel);
Restructure aclcheck error reporting to make permission-failure messages more uniform and internationalizable: the global array aclcheck_error_strings[] is gone in favor of a subroutine aclcheck_error(). Partial implementation of namespace-related permission checks --- not all done yet.
2002-04-27 05:45:03 +02:00
namespaceId = RelationGetNamespace(rel);
Require ownership permission for CREATE INDEX, per bug report. Disallow CREATE INDEX on system catalogs, non-tables (views, sequences, etc). Disallow CREATE/DROP TRIGGER on system catalogs, non-tables. Disallow ALTER TABLE ADD/DROP CONSTRAINT on system catalogs. Disallow FOREIGN KEY reference to non-table. None of these things can actually work in the present system structure, but the code was letting them pass without complaint.
2002-01-04 00:21:32 +01:00
Restructure aclcheck error reporting to make permission-failure messages more uniform and internationalizable: the global array aclcheck_error_strings[] is gone in favor of a subroutine aclcheck_error(). Partial implementation of namespace-related permission checks --- not all done yet.
2002-04-27 05:45:03 +02:00
/*
* Verify we (still) have CREATE rights in the rel's namespace.
pgindent run.
2002-09-04 22:31:48 +02:00
* (Presumably we did when the rel was created, but maybe not
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
* anymore.) Skip check if caller doesn't want it. Also skip check
* if bootstrapping, since permissions machinery may not be working yet.
Restructure aclcheck error reporting to make permission-failure messages more uniform and internationalizable: the global array aclcheck_error_strings[] is gone in favor of a subroutine aclcheck_error(). Partial implementation of namespace-related permission checks --- not all done yet.
2002-04-27 05:45:03 +02:00
*/
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
if (check_rights && !IsBootstrapProcessingMode())
Restructure aclcheck error reporting to make permission-failure messages more uniform and internationalizable: the global array aclcheck_error_strings[] is gone in favor of a subroutine aclcheck_error(). Partial implementation of namespace-related permission checks --- not all done yet.
2002-04-27 05:45:03 +02:00
{
AclResult aclresult;
aclresult = pg_namespace_aclcheck(namespaceId, GetUserId(),
ACL_CREATE);
if (aclresult != ACLCHECK_OK)
Adjust 'permission denied' messages to be more useful and consistent.
2003-08-01 02:15:26 +02:00
aclcheck_error(aclresult, ACL_KIND_NAMESPACE,
get_namespace_name(namespaceId));
Restructure aclcheck error reporting to make permission-failure messages more uniform and internationalizable: the global array aclcheck_error_strings[] is gone in favor of a subroutine aclcheck_error(). Partial implementation of namespace-related permission checks --- not all done yet.
2002-04-27 05:45:03 +02:00
}
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
/*
* Select name for index if caller didn't specify
*/
if (indexRelationName == NULL)
{
if (primary)
indexRelationName = CreateIndexName(RelationGetRelationName(rel),
NULL,
"pkey",
namespaceId);
else
{
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
IndexElem *iparam = (IndexElem *) linitial(attributeList);
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
indexRelationName = CreateIndexName(RelationGetRelationName(rel),
iparam->name,
"key",
namespaceId);
}
}
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
/*
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
* look up the access method, verify it can handle the requested
* features
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
*/
Restructure index AM interface for index building and index tuple deletion, per previous discussion on pghackers. Most of the duplicate code in different AMs' ambuild routines has been moved out to a common routine in index.c; this means that all index types now do the right things about inserting recently-dead tuples, etc. (I also removed support for EXTEND INDEX in the ambuild routines, since that's about to go away anyway, and it cluttered the code a lot.) The retail indextuple deletion routines have been replaced by a "bulk delete" routine in which the indexscan is inside the access method. I haven't pushed this change as far as it should go yet, but it should allow considerable simplification of the internal bookkeeping for deletions. Also, add flag columns to pg_am to eliminate various hardcoded tests on AM OIDs, and remove unused pg_am columns. Fix rtree and gist index types to not attempt to store NULLs; before this, gist usually crashed, while rtree managed not to crash but computed wacko bounding boxes for NULL entries (which might have had something to do with the performance problems we've heard about occasionally). Add AtEOXact routines to hash, rtree, and gist, all of which have static state that needs to be reset after an error. We discovered this need long ago for btree, but missed the other guys. Oh, one more thing: concurrent VACUUM is now the default.
2001-07-16 00:48:19 +02:00
tuple = SearchSysCache(AMNAME,
PointerGetDatum(accessMethodName),
0, 0, 0);
if (!HeapTupleIsValid(tuple))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_OBJECT),
errmsg("access method \"%s\" does not exist",
accessMethodName)));
oid is needed, it is added at the end of the struct (after the null bitmap, if present). Per Tom Lane's suggestion the information whether a tuple has an oid or not is carried in the tuple descriptor. For debugging reasons tdhasoid is of type char, not bool. There are predefined values for WITHOID, WITHOUTOID and UNDEFOID. This patch has been generated against a cvs snapshot from last week and I don't expect it to apply cleanly to current sources. While I post it here for public review, I'm working on a new version against a current snapshot. (There's been heavy activity recently; hope to catch up some day ...) This is a long patch; if it is too hard to swallow, I can provide it in smaller pieces: Part 1: Accessor macros Part 2: tdhasoid in TupDesc Part 3: Regression test Part 4: Parameter withoid to heap_addheader Part 5: Eliminate t_oid from HeapTupleHeader Part 2 is the most hairy part because of changes in the executor and even in the parser; the other parts are straightforward. Up to part 4 the patched postmaster stays binary compatible to databases created with an unpatched version. Part 5 is small (100 lines) and finally breaks compatibility. Manfred Koizar
2002-07-20 07:16:59 +02:00
accessMethodId = HeapTupleGetOid(tuple);
Restructure index AM interface for index building and index tuple deletion, per previous discussion on pghackers. Most of the duplicate code in different AMs' ambuild routines has been moved out to a common routine in index.c; this means that all index types now do the right things about inserting recently-dead tuples, etc. (I also removed support for EXTEND INDEX in the ambuild routines, since that's about to go away anyway, and it cluttered the code a lot.) The retail indextuple deletion routines have been replaced by a "bulk delete" routine in which the indexscan is inside the access method. I haven't pushed this change as far as it should go yet, but it should allow considerable simplification of the internal bookkeeping for deletions. Also, add flag columns to pg_am to eliminate various hardcoded tests on AM OIDs, and remove unused pg_am columns. Fix rtree and gist index types to not attempt to store NULLs; before this, gist usually crashed, while rtree managed not to crash but computed wacko bounding boxes for NULL entries (which might have had something to do with the performance problems we've heard about occasionally). Add AtEOXact routines to hash, rtree, and gist, all of which have static state that needs to be reset after an error. We discovered this need long ago for btree, but missed the other guys. Oh, one more thing: concurrent VACUUM is now the default.
2001-07-16 00:48:19 +02:00
accessMethodForm = (Form_pg_am) GETSTRUCT(tuple);
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
if (unique && !accessMethodForm->amcanunique)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
Message editing: remove gratuitous variations in message wording, standardize terms, add some clarifications, fix some untranslatable attempts at dynamic message building.
2003-09-25 08:58:07 +02:00
errmsg("access method \"%s\" does not support unique indexes",
pgindent run.
2003-08-04 02:43:34 +02:00
accessMethodName)));
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
if (numberOfAttributes > 1 && !accessMethodForm->amcanmulticol)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
Message editing: remove gratuitous variations in message wording, standardize terms, add some clarifications, fix some untranslatable attempts at dynamic message building.
2003-09-25 08:58:07 +02:00
errmsg("access method \"%s\" does not support multicolumn indexes",
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
accessMethodName)));
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
Restructure index AM interface for index building and index tuple deletion, per previous discussion on pghackers. Most of the duplicate code in different AMs' ambuild routines has been moved out to a common routine in index.c; this means that all index types now do the right things about inserting recently-dead tuples, etc. (I also removed support for EXTEND INDEX in the ambuild routines, since that's about to go away anyway, and it cluttered the code a lot.) The retail indextuple deletion routines have been replaced by a "bulk delete" routine in which the indexscan is inside the access method. I haven't pushed this change as far as it should go yet, but it should allow considerable simplification of the internal bookkeeping for deletions. Also, add flag columns to pg_am to eliminate various hardcoded tests on AM OIDs, and remove unused pg_am columns. Fix rtree and gist index types to not attempt to store NULLs; before this, gist usually crashed, while rtree managed not to crash but computed wacko bounding boxes for NULL entries (which might have had something to do with the performance problems we've heard about occasionally). Add AtEOXact routines to hash, rtree, and gist, all of which have static state that needs to be reset after an error. We discovered this need long ago for btree, but missed the other guys. Oh, one more thing: concurrent VACUUM is now the default.
2001-07-16 00:48:19 +02:00
ReleaseSysCache(tuple);
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
/*
* If a range table was created then check that only the base rel is
* mentioned.
*/
if (rangetable != NIL)
{
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
if (list_length(rangetable) != 1 || getrelid(1, rangetable) != relationId)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_INVALID_COLUMN_REFERENCE),
Various message fixes, among those fixes for the previous round of fixes
2003-09-26 17:27:37 +02:00
errmsg("index expressions and predicates may refer only to the table being indexed")));
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
}
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
/*
Clean up the usage of canonicalize_qual(): in particular, be consistent about whether it is applied before or after eval_const_expressions(). I believe there were some corner cases where the system would fail to recognize that a partial index is applicable because of the previous inconsistency. Store normal rather than 'implicit AND' representations of constraints and index predicates in the catalogs. initdb forced due to representation change of constraints/predicates.
2003-12-28 22:57:37 +01:00
* Validate predicate, if given
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
*/
Phase 2 of read-only-plans project: restructure expression-tree nodes so that all executable expression nodes inherit from a common supertype Expr. This is somewhat of an exercise in code purity rather than any real functional advance, but getting rid of the extra Oper or Func node formerly used in each operator or function call should provide at least a little space and speed improvement. initdb forced by changes in stored-rules representation.
2002-12-12 16:49:42 +01:00
if (predicate)
Clean up the usage of canonicalize_qual(): in particular, be consistent about whether it is applied before or after eval_const_expressions(). I believe there were some corner cases where the system would fail to recognize that a partial index is applicable because of the previous inconsistency. Store normal rather than 'implicit AND' representations of constraints and index predicates in the catalogs. initdb forced due to representation change of constraints/predicates.
2003-12-28 22:57:37 +01:00
CheckPredicate(predicate);
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
/*
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
* Extra checks when creating a PRIMARY KEY index.
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
*/
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
if (primary)
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
{
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
List *cmds;
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
ListCell *keys;
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
/*
* If ALTER TABLE, check that there isn't already a PRIMARY KEY.
* In CREATE TABLE, we have faith that the parser rejected multiple
* pkey clauses; and CREATE INDEX doesn't have a way to say
* PRIMARY KEY, so it's no problem either.
*/
if (is_alter_table &&
relationHasPrimaryKey(rel))
{
ereport(ERROR,
(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
errmsg("multiple primary keys for table \"%s\" are not allowed",
RelationGetRelationName(rel))));
}
/*
* Check that all of the attributes in a primary key are marked as not
* null, otherwise attempt to ALTER TABLE .. SET NOT NULL
*/
cmds = NIL;
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
foreach(keys, attributeList)
{
pgindent run.
2003-08-04 02:43:34 +02:00
IndexElem *key = (IndexElem *) lfirst(keys);
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
HeapTuple atttuple;
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
if (!key->name)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("primary keys cannot be expressions")));
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
/* System attributes are never null, so no problem */
if (SystemAttributeByName(key->name, rel->rd_rel->relhasoids))
continue;
atttuple = SearchSysCacheAttName(relationId, key->name);
if (HeapTupleIsValid(atttuple))
{
pgindent run.
2003-08-04 02:43:34 +02:00
if (!((Form_pg_attribute) GETSTRUCT(atttuple))->attnotnull)
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
{
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
/* Add a subcommand to make this one NOT NULL */
AlterTableCmd *cmd = makeNode(AlterTableCmd);
cmd->subtype = AT_SetNotNull;
cmd->name = key->name;
cmds = lappend(cmds, cmd);
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
}
ReleaseSysCache(atttuple);
}
else
{
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
/*
* This shouldn't happen during CREATE TABLE, but can
* happen during ALTER TABLE. Keep message in sync with
* transformIndexConstraints() in parser/analyze.c.
*/
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_COLUMN),
pgindent run.
2003-08-04 02:43:34 +02:00
errmsg("column \"%s\" named in key does not exist",
key->name)));
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
}
}
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
/*
* XXX: Shouldn't the ALTER TABLE .. SET NOT NULL cascade
* to child tables? Currently, since the PRIMARY KEY
* itself doesn't cascade, we don't cascade the
* notnull constraint(s) either; but this is pretty debatable.
*
* XXX: possible future improvement: when being called from
* ALTER TABLE, it would be more efficient to merge this with
* the outer ALTER TABLE, so as to avoid two scans. But that
* seems to complicate DefineIndex's API unduly.
*/
if (cmds)
AlterTableInternal(relationId, cmds, false);
Enforces NOT NULL constraints to be applied against new PRIMARY KEY columns in DefineIndex. So, ALTER TABLE ... PRIMARY KEY will now automatically add the NOT NULL constraint. It appeared the alter_table regression test wanted this to occur, as after the change the regression test better matched in inline 'fails'/'succeeds' comments. Rod Taylor
2003-01-02 20:29:22 +01:00
}
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
/*
pgindent run. Make it all clean.
2001-03-22 05:01:46 +01:00
* Prepare arguments for index_create, primarily an IndexInfo
Clean up the usage of canonicalize_qual(): in particular, be consistent about whether it is applied before or after eval_const_expressions(). I believe there were some corner cases where the system would fail to recognize that a partial index is applicable because of the previous inconsistency. Store normal rather than 'implicit AND' representations of constraints and index predicates in the catalogs. initdb forced due to representation change of constraints/predicates.
2003-12-28 22:57:37 +01:00
* structure. Note that ii_Predicate must be in implicit-AND format.
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
*/
indexInfo = makeNode(IndexInfo);
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
indexInfo->ii_NumIndexAttrs = numberOfAttributes;
pgindent run.
2003-08-04 02:43:34 +02:00
indexInfo->ii_Expressions = NIL; /* for now */
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
indexInfo->ii_ExpressionsState = NIL;
Clean up the usage of canonicalize_qual(): in particular, be consistent about whether it is applied before or after eval_const_expressions(). I believe there were some corner cases where the system would fail to recognize that a partial index is applicable because of the previous inconsistency. Store normal rather than 'implicit AND' representations of constraints and index predicates in the catalogs. initdb forced due to representation change of constraints/predicates.
2003-12-28 22:57:37 +01:00
indexInfo->ii_Predicate = make_ands_implicit(predicate);
Revise executor APIs so that all per-query state structure is built in a per-query memory context created by CreateExecutorState --- and destroyed by FreeExecutorState. This provides a final solution to the longstanding problem of memory leaked by various ExecEndNode calls.
2002-12-15 17:17:59 +01:00
indexInfo->ii_PredicateState = NIL;
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
indexInfo->ii_Unique = unique;
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
classObjectId = (Oid *) palloc(numberOfAttributes * sizeof(Oid));
ComputeIndexAttrs(indexInfo, classObjectId, attributeList,
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
relationId, accessMethodName, accessMethodId,
isconstraint);
heap_close(rel, NoLock);
/*
* Report index creation if appropriate (delay this till after most
* of the error checks)
*/
if (isconstraint && !quiet)
ereport(NOTICE,
(errmsg("%s %s will create implicit index \"%s\" for table \"%s\"",
is_alter_table ? "ALTER TABLE / ADD" : "CREATE TABLE /",
primary ? "PRIMARY KEY" : "UNIQUE",
indexRelationName, RelationGetRelationName(rel))));
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
index_create(relationId, indexRelationName,
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
indexInfo, accessMethodId, classObjectId,
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
primary, isconstraint,
allowSystemTableMods, skip_build);
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
Fix performance problems with pg_index lookups (see, for example, discussion of 5/19/00). pg_index is now searched for indexes of a relation using an indexscan. Moreover, this is done once and cached in the relcache entry for the relation, in the form of a list of OIDs for the indexes. This list is used by the parser and executor to drive lookups in the pg_index syscache when they want to know the properties of the indexes. Net result: index information will be fully cached for repetitive operations such as inserts.
2000-06-17 23:49:04 +02:00
/*
* We update the relation's pg_class tuple even if it already has
pgindent run. Make it all clean.
2001-03-22 05:01:46 +01:00
* relhasindex = true. This is needed to cause a shared-cache-inval
Fix performance problems with pg_index lookups (see, for example, discussion of 5/19/00). pg_index is now searched for indexes of a relation using an indexscan. Moreover, this is done once and cached in the relcache entry for the relation, in the form of a list of OIDs for the indexes. This list is used by the parser and executor to drive lookups in the pg_index syscache when they want to know the properties of the indexes. Net result: index information will be fully cached for repetitive operations such as inserts.
2000-06-17 23:49:04 +02:00
* message to be sent for the pg_class tuple, which will cause other
* backends to flush their relcache entries and in particular their
* cached lists of the indexes for this relation.
*/
Make OIDs optional, per discussions in pghackers. WITH OIDS is still the default, but OIDS are removed from many system catalogs that don't need them. Some interesting side effects: TOAST pointers are 20 bytes not 32 now; pg_description has a three-column key instead of one. Bugs fixed in passing: BINARY cursors work again; pg_class.relhaspkey has some usefulness; pg_dump dumps comments on indexes, rules, and triggers in a valid order. initdb forced.
2001-08-10 20:57:42 +02:00
setRelhasindex(relationId, true, primary, InvalidOid);
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
}
/*
* CheckPredicate
Clean up the usage of canonicalize_qual(): in particular, be consistent about whether it is applied before or after eval_const_expressions(). I believe there were some corner cases where the system would fail to recognize that a partial index is applicable because of the previous inconsistency. Store normal rather than 'implicit AND' representations of constraints and index predicates in the catalogs. initdb forced due to representation change of constraints/predicates.
2003-12-28 22:57:37 +01:00
* Checks that the given partial-index predicate is valid.
Partial indexes work again, courtesy of Martijn van Oosterhout. Note: I didn't force an initdb, figuring that one today was enough. However, there is a new function in pg_proc.h, and pg_dump won't be able to dump partial indexes until you add that function.
2001-07-16 07:07:00 +02:00
*
* This used to also constrain the form of the predicate to forms that
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
* indxpath.c could do something with. However, that seems overly
Partial indexes work again, courtesy of Martijn van Oosterhout. Note: I didn't force an initdb, figuring that one today was enough. However, there is a new function in pg_proc.h, and pg_dump won't be able to dump partial indexes until you add that function.
2001-07-16 07:07:00 +02:00
* restrictive. One useful application of partial indexes is to apply
* a UNIQUE constraint across a subset of a table, and in that scenario
* any evaluatable predicate will work. So accept any predicate here
* (except ones requiring a plan), and let indxpath.c fend for itself.
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*/
static void
Clean up the usage of canonicalize_qual(): in particular, be consistent about whether it is applied before or after eval_const_expressions(). I believe there were some corner cases where the system would fail to recognize that a partial index is applicable because of the previous inconsistency. Store normal rather than 'implicit AND' representations of constraints and index predicates in the catalogs. initdb forced due to representation change of constraints/predicates.
2003-12-28 22:57:37 +01:00
CheckPredicate(Expr *predicate)
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
{
Disallow non-cachable functions in functional indexes and in index predicates. Per suggestion from Hiroshi.
2001-07-17 23:53:01 +02:00
/*
* We don't currently support generation of an actual query plan for a
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
* predicate, only simple scalar expressions; hence these
* restrictions.
Disallow non-cachable functions in functional indexes and in index predicates. Per suggestion from Hiroshi.
2001-07-17 23:53:01 +02:00
*/
Clean up the usage of canonicalize_qual(): in particular, be consistent about whether it is applied before or after eval_const_expressions(). I believe there were some corner cases where the system would fail to recognize that a partial index is applicable because of the previous inconsistency. Store normal rather than 'implicit AND' representations of constraints and index predicates in the catalogs. initdb forced due to representation change of constraints/predicates.
2003-12-28 22:57:37 +01:00
if (contain_subplans((Node *) predicate))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
Message editing: remove gratuitous variations in message wording, standardize terms, add some clarifications, fix some untranslatable attempts at dynamic message building.
2003-09-25 08:58:07 +02:00
errmsg("cannot use subquery in index predicate")));
Clean up the usage of canonicalize_qual(): in particular, be consistent about whether it is applied before or after eval_const_expressions(). I believe there were some corner cases where the system would fail to recognize that a partial index is applicable because of the previous inconsistency. Store normal rather than 'implicit AND' representations of constraints and index predicates in the catalogs. initdb forced due to representation change of constraints/predicates.
2003-12-28 22:57:37 +01:00
if (contain_agg_clause((Node *) predicate))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_GROUPING_ERROR),
errmsg("cannot use aggregate in index predicate")));
Disallow non-cachable functions in functional indexes and in index predicates. Per suggestion from Hiroshi.
2001-07-17 23:53:01 +02:00
/*
pgindent run.
2002-09-04 22:31:48 +02:00
* A predicate using mutable functions is probably wrong, for the same
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
* reasons that we don't allow an index expression to use one.
Disallow non-cachable functions in functional indexes and in index predicates. Per suggestion from Hiroshi.
2001-07-17 23:53:01 +02:00
*/
Clean up the usage of canonicalize_qual(): in particular, be consistent about whether it is applied before or after eval_const_expressions(). I believe there were some corner cases where the system would fail to recognize that a partial index is applicable because of the previous inconsistency. Store normal rather than 'implicit AND' representations of constraints and index predicates in the catalogs. initdb forced due to representation change of constraints/predicates.
2003-12-28 22:57:37 +01:00
if (contain_mutable_functions((Node *) predicate))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
pgindent run.
2003-08-04 02:43:34 +02:00
errmsg("functions in index predicate must be marked IMMUTABLE")));
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
}
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
static void
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
ComputeIndexAttrs(IndexInfo *indexInfo,
Oid *classOidP,
List *attList, /* list of IndexElem's */
Oid relId,
char *accessMethodName,
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
Oid accessMethodId,
bool isconstraint)
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
{
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
ListCell *rest;
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
int attn = 0;
Fixes a bug in 'create index' Submitted by: Dan McGuirk <mcguirk@indirect.com>
1996-08-15 09:42:52 +02:00
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
/*
* process attributeList
*/
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
foreach(rest, attList)
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
{
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
IndexElem *attribute = (IndexElem *) lfirst(rest);
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
Oid atttype;
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
if (attribute->name != NULL)
{
/* Simple index attribute */
HeapTuple atttuple;
Form_pg_attribute attform;
Assert(attribute->expr == NULL);
atttuple = SearchSysCacheAttName(relId, attribute->name);
if (!HeapTupleIsValid(atttuple))
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
{
/* difference in error message spellings is historical */
if (isconstraint)
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_COLUMN),
errmsg("column \"%s\" named in key does not exist",
attribute->name)));
else
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_COLUMN),
errmsg("column \"%s\" does not exist",
attribute->name)));
}
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
attform = (Form_pg_attribute) GETSTRUCT(atttuple);
indexInfo->ii_KeyAttrNumbers[attn] = attform->attnum;
atttype = attform->atttypid;
ReleaseSysCache(atttuple);
}
else if (attribute->expr && IsA(attribute->expr, Var))
{
/* Tricky tricky, he wrote (column) ... treat as simple attr */
pgindent run.
2003-08-04 02:43:34 +02:00
Var *var = (Var *) attribute->expr;
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
indexInfo->ii_KeyAttrNumbers[attn] = var->varattno;
atttype = get_atttype(relId, var->varattno);
}
else
{
/* Index expression */
Assert(attribute->expr != NULL);
pgindent run.
2003-08-04 02:43:34 +02:00
indexInfo->ii_KeyAttrNumbers[attn] = 0; /* marks expression */
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
indexInfo->ii_Expressions = lappend(indexInfo->ii_Expressions,
attribute->expr);
atttype = exprType(attribute->expr);
/*
pgindent run.
2003-08-04 02:43:34 +02:00
* We don't currently support generation of an actual query
* plan for an index expression, only simple scalar
* expressions; hence these restrictions.
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
*/
if (contain_subplans(attribute->expr))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
Message editing: remove gratuitous variations in message wording, standardize terms, add some clarifications, fix some untranslatable attempts at dynamic message building.
2003-09-25 08:58:07 +02:00
errmsg("cannot use subquery in index expression")));
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
if (contain_agg_clause(attribute->expr))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_GROUPING_ERROR),
Message editing: remove gratuitous variations in message wording, standardize terms, add some clarifications, fix some untranslatable attempts at dynamic message building.
2003-09-25 08:58:07 +02:00
errmsg("cannot use aggregate function in index expression")));
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
/*
* A expression using mutable functions is probably wrong,
pgindent run.
2003-08-04 02:43:34 +02:00
* since if you aren't going to get the same result for the
* same data every time, it's not clear what the index entries
* mean at all.
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
*/
if (contain_mutable_functions(attribute->expr))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_INVALID_OBJECT_DEFINITION),
errmsg("functions in index expression must be marked IMMUTABLE")));
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
}
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
classOidP[attn] = GetIndexOpClass(attribute->opclass,
atttype,
accessMethodName,
accessMethodId);
Cleanup of code for creating index entries. Functional indexes with pass-by-ref data types --- eg, an index on lower(textfield) --- no longer leak memory during index creation or update. Clean up a lot of redundant code ... did you know that copy, vacuum, truncate, reindex, extend index, and bootstrap each basically duplicated the main executor's logic for extracting information about an index and preparing index entries? Functional indexes should be a little faster now too, due to removal of repeated function lookups. CREATE INDEX 'opt_type' clause is deimplemented by these changes, but I haven't removed it from the parser yet (need to merge with Thomas' latest change set first).
2000-07-15 00:18:02 +02:00
attn++;
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
}
}
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
/*
* Resolve possibly-defaulted operator class specification
*/
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
static Oid
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
GetIndexOpClass(List *opclass, Oid attrType,
char *accessMethodName, Oid accessMethodId)
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
{
Centralize code for interpreting schema references, which had gotten copied more places than I first thought it would. This fixes a bug: a couple of these places were neglecting to enforce USAGE access on explicitly-referenced schemas.
2002-07-30 01:46:35 +02:00
char *schemaname;
char *opcname;
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
HeapTuple tuple;
Check that user-specified opclass in CREATE INDEX corresponds to operators that will actually work on the column datatype.
2000-04-25 04:45:54 +02:00
Oid opClassId,
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
opInputType;
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
/*
pgindent run.
2003-08-04 02:43:34 +02:00
* Release 7.0 removed network_ops, timespan_ops, and datetime_ops, so
* we ignore those opclass names so the default *_ops is used. This
* can be removed in some later release. bjm 2000/02/07
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
*
pgindent run.
2003-08-04 02:43:34 +02:00
* Release 7.1 removes lztext_ops, so suppress that too for a while. tgl
* 2000/07/30
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
*
pgindent run.
2003-08-04 02:43:34 +02:00
* Release 7.2 renames timestamp_ops to timestamptz_ops, so suppress that
* too for awhile. I'm starting to think we need a better approach.
* tgl 2000/10/01
Cross-data-type comparisons are now indexable by btrees, pursuant to my pghackers proposal of 8-Nov. All the existing cross-type comparison operators (int2/int4/int8 and float4/float8) have appropriate support. The original proposal of storing the right-hand-side datatype as part of the primary key for pg_amop and pg_amproc got modified a bit in the event; it is easier to store zero as the 'default' case and only store a nonzero when the operator is actually cross-type. Along the way, remove the long-since-defunct bigbox_ops operator class.
2003-11-12 22:15:59 +01:00
*
* Release 7.5 removes bigbox_ops (which was dead code for a long while
* anyway). tgl 2003/11/11
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
*/
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
if (list_length(opclass) == 1)
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
{
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
char *claname = strVal(linitial(opclass));
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
if (strcmp(claname, "network_ops") == 0 ||
strcmp(claname, "timespan_ops") == 0 ||
strcmp(claname, "datetime_ops") == 0 ||
strcmp(claname, "lztext_ops") == 0 ||
Cross-data-type comparisons are now indexable by btrees, pursuant to my pghackers proposal of 8-Nov. All the existing cross-type comparison operators (int2/int4/int8 and float4/float8) have appropriate support. The original proposal of storing the right-hand-side datatype as part of the primary key for pg_amop and pg_amproc got modified a bit in the event; it is easier to store zero as the 'default' case and only store a nonzero when the operator is actually cross-type. Along the way, remove the long-since-defunct bigbox_ops operator class.
2003-11-12 22:15:59 +01:00
strcmp(claname, "timestamp_ops") == 0 ||
strcmp(claname, "bigbox_ops") == 0)
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
opclass = NIL;
}
if (opclass == NIL)
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
{
/* no operator class specified, so find the default */
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
opClassId = GetDefaultOpClass(attrType, accessMethodId);
if (!OidIsValid(opClassId))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_OBJECT),
errmsg("data type %s has no default operator class for access method \"%s\"",
format_type_be(attrType), accessMethodName),
errhint("You must specify an operator class for the index or define a default operator class for the data type.")));
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
return opClassId;
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
}
Produce an appropriate error message when opclass is not supported by specified index access method. Clean up wording of some existing error messages, too.
2000-04-23 03:44:55 +02:00
/*
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
* Specific opclass name given, so look up the opclass.
Produce an appropriate error message when opclass is not supported by specified index access method. Clean up wording of some existing error messages, too.
2000-04-23 03:44:55 +02:00
*/
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
/* deconstruct the name list */
Replace functional-index facility with expressional indexes. Any column of an index can now be a computed expression instead of a simple variable. Restrictions on expressions are the same as for predicates (only immutable functions, no sub-selects). This fixes problems recently introduced with inlining SQL functions, because the inlining transformation is applied to both expression trees so the planner can still match them up. Along the way, improve efficiency of handling index predicates (both predicates and index expressions are now cached by the relcache) and fix 7.3 oversight that didn't record dependencies of predicate expressions.
2003-05-28 18:04:02 +02:00
DeconstructQualifiedName(opclass, &schemaname, &opcname);
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
if (schemaname)
{
/* Look in specific schema only */
pgindent run.
2002-09-04 22:31:48 +02:00
Oid namespaceId;
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
Centralize code for interpreting schema references, which had gotten copied more places than I first thought it would. This fixes a bug: a couple of these places were neglecting to enforce USAGE access on explicitly-referenced schemas.
2002-07-30 01:46:35 +02:00
namespaceId = LookupExplicitNamespace(schemaname);
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
tuple = SearchSysCache(CLAAMNAMENSP,
ObjectIdGetDatum(accessMethodId),
PointerGetDatum(opcname),
ObjectIdGetDatum(namespaceId),
0);
}
else
{
/* Unqualified opclass name, so search the search path */
opClassId = OpclassnameGetOpcid(accessMethodId, opcname);
if (!OidIsValid(opClassId))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_OBJECT),
errmsg("operator class \"%s\" does not exist for access method \"%s\"",
opcname, accessMethodName)));
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
tuple = SearchSysCache(CLAOID,
ObjectIdGetDatum(opClassId),
0, 0, 0);
}
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
if (!HeapTupleIsValid(tuple))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_OBJECT),
errmsg("operator class \"%s\" does not exist for access method \"%s\"",
NameListToString(opclass), accessMethodName)));
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
/*
pgindent run.
2002-09-04 22:31:48 +02:00
* Verify that the index operator class accepts this datatype. Note
* we will accept binary compatibility.
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
*/
oid is needed, it is added at the end of the struct (after the null bitmap, if present). Per Tom Lane's suggestion the information whether a tuple has an oid or not is carried in the tuple descriptor. For debugging reasons tdhasoid is of type char, not bool. There are predefined values for WITHOID, WITHOUTOID and UNDEFOID. This patch has been generated against a cvs snapshot from last week and I don't expect it to apply cleanly to current sources. While I post it here for public review, I'm working on a new version against a current snapshot. (There's been heavy activity recently; hope to catch up some day ...) This is a long patch; if it is too hard to swallow, I can provide it in smaller pieces: Part 1: Accessor macros Part 2: tdhasoid in TupDesc Part 3: Regression test Part 4: Parameter withoid to heap_addheader Part 5: Eliminate t_oid from HeapTupleHeader Part 2 is the most hairy part because of changes in the executor and even in the parser; the other parts are straightforward. Up to part 4 the patched postmaster stays binary compatible to databases created with an unpatched version. Part 5 is small (100 lines) and finally breaks compatibility. Manfred Koizar
2002-07-20 07:16:59 +02:00
opClassId = HeapTupleGetOid(tuple);
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
opInputType = ((Form_pg_opclass) GETSTRUCT(tuple))->opcintype;
Produce an appropriate error message when opclass is not supported by specified index access method. Clean up wording of some existing error messages, too.
2000-04-23 03:44:55 +02:00
Extend pg_cast castimplicit column to a three-way value; this allows us to be flexible about assignment casts without introducing ambiguity in operator/function resolution. Introduce a well-defined promotion hierarchy for numeric datatypes (int2->int4->int8->numeric->float4->float8). Change make_const to initially label numeric literals as int4, int8, or numeric (never float8 anymore). Explicitly mark Func and RelabelType nodes to indicate whether they came from a function call, explicit cast, or implicit cast; use this to do reverse-listing more accurately and without so many heuristics. Explicit casts to char, varchar, bit, varbit will truncate or pad without raising an error (the pre-7.2 behavior), while assigning to a column without any explicit cast will still raise an error for wrong-length data like 7.3. This more nearly follows the SQL spec than 7.2 behavior (we should be reporting a 'completion condition' in the explicit-cast cases, but we have no mechanism for that, so just do silent truncation). Fix some problems with enforcement of typmod for array elements; it didn't work at all in 'UPDATE ... SET array[n] = foo', for example. Provide a generalized array_length_coerce() function to replace the specialized per-array-type functions that used to be needed (and were missing for NUMERIC as well as all the datetime types). Add missing conversions int8<->float4, text<->numeric, oid<->int8. initdb forced.
2002-09-18 23:35:25 +02:00
if (!IsBinaryCoercible(attrType, opInputType))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_DATATYPE_MISMATCH),
pgindent run.
2003-08-04 02:43:34 +02:00
errmsg("operator class \"%s\" does not accept data type %s",
NameListToString(opclass), format_type_be(attrType))));
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
ReleaseSysCache(tuple);
Check that user-specified opclass in CREATE INDEX corresponds to operators that will actually work on the column datatype.
2000-04-25 04:45:54 +02:00
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
return opClassId;
}
static Oid
GetDefaultOpClass(Oid attrType, Oid accessMethodId)
{
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
OpclassCandidateList opclass;
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
int nexact = 0;
int ncompatible = 0;
Oid exactOid = InvalidOid;
Oid compatibleOid = InvalidOid;
Fix longstanding bug that kept functional indexes from working when you defaulted the opclass. This addresses TODO item * Allow creation of functional indexes to use default types (Does that make it a feature? Oh dear...)
2000-02-25 03:58:48 +01:00
Make domain types indexable. Rod Taylor & Tom Lane.
2002-08-16 22:55:09 +02:00
/* If it's a domain, look at the base type instead */
attrType = getBaseType(attrType);
Check that user-specified opclass in CREATE INDEX corresponds to operators that will actually work on the column datatype.
2000-04-25 04:45:54 +02:00
/*
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
* We scan through all the opclasses available for the access method,
* looking for one that is marked default and matches the target type
* (either exactly or binary-compatibly, but prefer an exact match).
*
* We could find more than one binary-compatible match, in which case we
pgindent run on all C files. Java run to follow. initdb/regression tests pass.
2001-10-25 07:50:21 +02:00
* require the user to specify which one he wants. If we find more
* than one exact match, then someone put bogus entries in pg_opclass.
Check that user-specified opclass in CREATE INDEX corresponds to operators that will actually work on the column datatype.
2000-04-25 04:45:54 +02:00
*
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
* The initial search is done by namespace.c so that we only consider
Create a 'type cache' that keeps track of the data needed for any particular datatype by array_eq and array_cmp; use this to solve problems with memory leaks in array indexing support. The parser's equality_oper and ordering_oper routines also use the cache. Change the operator search algorithms to look for appropriate btree or hash index opclasses, instead of assuming operators named '<' or '=' have the right semantics. (ORDER BY ASC/DESC now also look at opclasses, instead of assuming '<' and '>' are the right things.) Add several more index opclasses so that there is no regression in functionality for base datatypes. initdb forced due to catalog additions.
2003-08-17 21:58:06 +02:00
* opclasses visible in the current namespace search path. (See also
* typcache.c, which applies the same logic, but over all opclasses.)
Check that user-specified opclass in CREATE INDEX corresponds to operators that will actually work on the column datatype.
2000-04-25 04:45:54 +02:00
*/
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
for (opclass = OpclassGetCandidates(accessMethodId);
opclass != NULL;
opclass = opclass->next)
Check that user-specified opclass in CREATE INDEX corresponds to operators that will actually work on the column datatype.
2000-04-25 04:45:54 +02:00
{
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
if (opclass->opcdefault)
Check that user-specified opclass in CREATE INDEX corresponds to operators that will actually work on the column datatype.
2000-04-25 04:45:54 +02:00
{
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
if (opclass->opcintype == attrType)
{
nexact++;
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
exactOid = opclass->oid;
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
}
Extend pg_cast castimplicit column to a three-way value; this allows us to be flexible about assignment casts without introducing ambiguity in operator/function resolution. Introduce a well-defined promotion hierarchy for numeric datatypes (int2->int4->int8->numeric->float4->float8). Change make_const to initially label numeric literals as int4, int8, or numeric (never float8 anymore). Explicitly mark Func and RelabelType nodes to indicate whether they came from a function call, explicit cast, or implicit cast; use this to do reverse-listing more accurately and without so many heuristics. Explicit casts to char, varchar, bit, varbit will truncate or pad without raising an error (the pre-7.2 behavior), while assigning to a column without any explicit cast will still raise an error for wrong-length data like 7.3. This more nearly follows the SQL spec than 7.2 behavior (we should be reporting a 'completion condition' in the explicit-cast cases, but we have no mechanism for that, so just do silent truncation). Fix some problems with enforcement of typmod for array elements; it didn't work at all in 'UPDATE ... SET array[n] = foo', for example. Provide a generalized array_length_coerce() function to replace the specialized per-array-type functions that used to be needed (and were missing for NUMERIC as well as all the datetime types). Add missing conversions int8<->float4, text<->numeric, oid<->int8. initdb forced.
2002-09-18 23:35:25 +02:00
else if (IsBinaryCoercible(attrType, opclass->opcintype))
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
{
ncompatible++;
Opclasses live in namespaces. I also took the opportunity to create an 'opclass owner' column in pg_opclass. Nothing is done with it at present, but since there are plans to invent a CREATE OPERATOR CLASS command soon, we'll probably want DROP OPERATOR CLASS too, which suggests that a notion of ownership would be a good idea.
2002-04-17 22:57:57 +02:00
compatibleOid = opclass->oid;
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
}
Check that user-specified opclass in CREATE INDEX corresponds to operators that will actually work on the column datatype.
2000-04-25 04:45:54 +02:00
}
}
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
if (nexact == 1)
return exactOid;
if (nexact != 0)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_DUPLICATE_OBJECT),
errmsg("there are multiple default operator classes for data type %s",
format_type_be(attrType))));
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
if (ncompatible == 1)
return compatibleOid;
Change SearchSysCache coding conventions so that a reference count is maintained for each cache entry. A cache entry will not be freed until the matching ReleaseSysCache call has been executed. This eliminates worries about cache entries getting dropped while still in use. See my posting to pg-hackers of even date for more info.
2000-11-16 23:30:52 +01:00
Restructure pg_opclass, pg_amop, and pg_amproc per previous discussions in pgsql-hackers. pg_opclass now has a row for each opclass supported by each index AM, not a row for each opclass name. This allows pg_opclass to show directly whether an AM supports an opclass, and furthermore makes it possible to store additional information about an opclass that might be AM-dependent. pg_opclass and pg_amop now store "lossy" and "haskeytype" information that we previously expected the user to remember to provide in CREATE INDEX commands. Lossiness is no longer an index-level property, but is associated with the use of a particular operator in a particular index opclass. Along the way, IndexSupportInitialize now uses the syscaches to retrieve pg_amop and pg_amproc entries. I find this reduces backend launch time by about ten percent, at the cost of a couple more special cases in catcache.c's IndexScanOK. Initial work by Oleg Bartunov and Teodor Sigaev, further hacking by Tom Lane. initdb forced.
2001-08-21 18:36:06 +02:00
return InvalidOid;
Fixes a bug in 'create index' Submitted by: Dan McGuirk <mcguirk@indirect.com>
1996-08-15 09:42:52 +02:00
}
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
/*
* Select a nonconflicting name for an index.
*/
static char *
CreateIndexName(const char *table_name, const char *column_name,
const char *label, Oid inamespace)
{
int pass = 0;
char *iname = NULL;
char typename[NAMEDATALEN];
/*
* The type name for makeObjectName is label, or labelN if that's
* necessary to prevent collision with existing indexes.
*/
strncpy(typename, label, sizeof(typename));
for (;;)
{
iname = makeObjectName(table_name, column_name, typename);
if (!OidIsValid(get_relname_relid(iname, inamespace)))
break;
/* found a conflict, so try a new name component */
pfree(iname);
snprintf(typename, sizeof(typename), "%s%d", label, ++pass);
}
return iname;
}
/*
* relationHasPrimaryKey -
*
* See whether an existing relation has a primary key.
*/
static bool
relationHasPrimaryKey(Relation rel)
{
bool result = false;
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
List *indexoidlist;
ListCell *indexoidscan;
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
/*
* Get the list of index OIDs for the table from the relcache, and
* look up each one in the pg_index syscache until we find one marked
* primary key (hopefully there isn't more than one such).
*/
indexoidlist = RelationGetIndexList(rel);
foreach(indexoidscan, indexoidlist)
{
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
Oid indexoid = lfirst_oid(indexoidscan);
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
HeapTuple indexTuple;
indexTuple = SearchSysCache(INDEXRELID,
ObjectIdGetDatum(indexoid),
0, 0, 0);
if (!HeapTupleIsValid(indexTuple)) /* should not happen */
elog(ERROR, "cache lookup failed for index %u", indexoid);
result = ((Form_pg_index) GETSTRUCT(indexTuple))->indisprimary;
ReleaseSysCache(indexTuple);
if (result)
break;
}
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
list_free(indexoidlist);
ALTER TABLE rewrite. New cool stuff: * ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL spec. A default is implemented by rewriting the table with the new value stored in each row. * ALTER COLUMN TYPE. You can change a column's datatype to anything you want, so long as you can specify how to convert the old value. Rewrites the table. (Possible future improvement: optimize no-op conversions such as varchar(N) to varchar(N+1).) * Multiple ALTER actions in a single ALTER TABLE command. You can perform any number of column additions, type changes, and constraint additions with only one pass over the table contents. Basic documentation provided in ALTER TABLE ref page, but some more docs work is needed. Original patch from Rod Taylor, additional work from Tom Lane.
2004-05-05 06:48:48 +02:00
return result;
}
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
/*
pgindent run over code.
1999-05-25 18:15:34 +02:00
* RemoveIndex
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
* Deletes an index.
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*/
void
First phase of applying Rod Taylor's pg_depend patch. This just adds RESTRICT/CASCADE syntax to the DROP commands that need it, and propagates the behavioral option through the parser to the routines that execute drops. Doesn't do anything useful yet, but I figured I'd commit these changes so I could get out of the parser area while working on the rest.
2002-07-01 17:27:56 +02:00
RemoveIndex(RangeVar *relation, DropBehavior behavior)
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
{
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
Oid indOid;
Tweak heap.c to refuse attempts to create table columns of standalone composite types. Add a couple more lsyscache.c routines to support this, and make use of them in some other places that were doing lookups the hard way.
2002-09-20 01:40:56 +02:00
char relkind;
Second phase of committing Rod Taylor's pg_depend/pg_constraint patch. pg_relcheck is gone; CHECK, UNIQUE, PRIMARY KEY, and FOREIGN KEY constraints all have real live entries in pg_constraint. pg_depend exists, and RESTRICT/CASCADE options work on most kinds of DROP; however, pg_depend is not yet very well populated with dependencies. (Most of the ones that are present at this point just replace formerly hardwired associations, such as the implicit drop of a relation's pg_type entry when the relation is dropped.) Need to add more logic to create dependency entries, improve pg_dump to dump constraints in place of indexes and triggers, and add some regression tests.
2002-07-12 20:43:19 +02:00
ObjectAddress object;
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
indOid = RangeVarGetRelid(relation, false);
Tweak heap.c to refuse attempts to create table columns of standalone composite types. Add a couple more lsyscache.c routines to support this, and make use of them in some other places that were doing lookups the hard way.
2002-09-20 01:40:56 +02:00
relkind = get_rel_relkind(indOid);
if (relkind != RELKIND_INDEX)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
Message editing: remove gratuitous variations in message wording, standardize terms, add some clarifications, fix some untranslatable attempts at dynamic message building.
2003-09-25 08:58:07 +02:00
errmsg("\"%s\" is not an index",
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
relation->relname)));
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
Second phase of committing Rod Taylor's pg_depend/pg_constraint patch. pg_relcheck is gone; CHECK, UNIQUE, PRIMARY KEY, and FOREIGN KEY constraints all have real live entries in pg_constraint. pg_depend exists, and RESTRICT/CASCADE options work on most kinds of DROP; however, pg_depend is not yet very well populated with dependencies. (Most of the ones that are present at this point just replace formerly hardwired associations, such as the implicit drop of a relation's pg_type entry when the relation is dropped.) Need to add more logic to create dependency entries, improve pg_dump to dump constraints in place of indexes and triggers, and add some regression tests.
2002-07-12 20:43:19 +02:00
object.classId = RelOid_pg_class;
object.objectId = indOid;
object.objectSubId = 0;
performDeletion(&object, behavior);
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
}
Implement reindex command
2000-02-18 10:30:20 +01:00
/*
Cause REINDEX to regard TOAST tables as regular relations, not system tables that need special defenses. I believe this is okay even for TOAST tables that belong to system tables.
2002-08-29 17:56:20 +02:00
* ReindexIndex
Implement reindex command
2000-02-18 10:30:20 +01:00
* Recreate an index.
*/
void
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
ReindexIndex(RangeVar *indexRelation, bool force /* currently unused */ )
Implement reindex command
2000-02-18 10:30:20 +01:00
{
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
Oid indOid;
Implement reindex command
2000-02-18 10:30:20 +01:00
HeapTuple tuple;
Make DROP TABLE rollback-able: postpone physical file delete until commit. (WAL logging for this is not done yet, however.) Clean up a number of really crufty things that are no longer needed now that DROP behaves nicely. Make temp table mapper do the right things when drop or rename affecting a temp table is rolled back. Also, remove "relation modified while in use" error check, in favor of locking tables at first reference and holding that lock throughout the statement.
2000-11-08 23:10:03 +01:00
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
indOid = RangeVarGetRelid(indexRelation, false);
tuple = SearchSysCache(RELOID,
ObjectIdGetDatum(indOid),
Change SearchSysCache coding conventions so that a reference count is maintained for each cache entry. A cache entry will not be freed until the matching ReleaseSysCache call has been executed. This eliminates worries about cache entries getting dropped while still in use. See my posting to pg-hackers of even date for more info.
2000-11-16 23:30:52 +01:00
0, 0, 0);
pgindent run.
2003-08-04 02:43:34 +02:00
if (!HeapTupleIsValid(tuple)) /* shouldn't happen */
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
elog(ERROR, "cache lookup failed for relation %u", indOid);
Implement reindex command
2000-02-18 10:30:20 +01:00
if (((Form_pg_class) GETSTRUCT(tuple))->relkind != RELKIND_INDEX)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
Message editing: remove gratuitous variations in message wording, standardize terms, add some clarifications, fix some untranslatable attempts at dynamic message building.
2003-09-25 08:58:07 +02:00
errmsg("\"%s\" is not an index",
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
indexRelation->relname)));
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
/* Check permissions */
if (!pg_class_ownercheck(indOid, GetUserId()))
aclcheck_error(ACLCHECK_NOT_OWNER, ACL_KIND_CLASS,
indexRelation->relname);
Checking to decide whether relations are system relations now depends on the namespace not the name; pg_ is not a reserved prefix for table names anymore. From Fernando Nasser.
2002-04-12 22:38:31 +02:00
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
ReleaseSysCache(tuple);
Implement reindex command
2000-02-18 10:30:20 +01:00
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
reindex_index(indOid);
Implement reindex command
2000-02-18 10:30:20 +01:00
}
/*
* ReindexTable
* Recreate indexes of a table.
*/
void
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
ReindexTable(RangeVar *relation, bool force /* currently unused */ )
Implement reindex command
2000-02-18 10:30:20 +01:00
{
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
Oid heapOid;
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
HeapTuple tuple;
Implement reindex command
2000-02-18 10:30:20 +01:00
pg_class has a relnamespace column. You can create and access tables in schemas other than the system namespace; however, there's no search path yet, and not all operations work yet on tables outside the system namespace.
2002-03-26 20:17:02 +01:00
heapOid = RangeVarGetRelid(relation, false);
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
tuple = SearchSysCache(RELOID,
ObjectIdGetDatum(heapOid),
0, 0, 0);
if (!HeapTupleIsValid(tuple)) /* shouldn't happen */
elog(ERROR, "cache lookup failed for relation %u", heapOid);
Implement reindex command
2000-02-18 10:30:20 +01:00
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
if (((Form_pg_class) GETSTRUCT(tuple))->relkind != RELKIND_RELATION &&
((Form_pg_class) GETSTRUCT(tuple))->relkind != RELKIND_TOASTVALUE)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
Eliminate another gratuitous message wording difference.
2003-09-29 18:37:29 +02:00
errmsg("\"%s\" is not a table",
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
relation->relname)));
Change SearchSysCache coding conventions so that a reference count is maintained for each cache entry. A cache entry will not be freed until the matching ReleaseSysCache call has been executed. This eliminates worries about cache entries getting dropped while still in use. See my posting to pg-hackers of even date for more info.
2000-11-16 23:30:52 +01:00
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
/* Check permissions */
if (!pg_class_ownercheck(heapOid, GetUserId()))
aclcheck_error(ACLCHECK_NOT_OWNER, ACL_KIND_CLASS,
relation->relname);
/* Can't reindex shared tables except in standalone mode */
if (((Form_pg_class) GETSTRUCT(tuple))->relisshared && IsUnderPostmaster)
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
More message editing, some suggested by Alvaro Herrera
2003-09-29 02:05:25 +02:00
errmsg("shared table \"%s\" can only be reindexed in stand-alone mode",
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
relation->relname)));
ReleaseSysCache(tuple);
Fix places that were using IsTransactionBlock() as an (inadequate) check that they'd get to commit immediately on finishing. There's now a centralized routine PreventTransactionChain() that implements the necessary tests.
2002-10-22 00:06:20 +02:00
Get rid of cluster.c's apparatus for rebuilding a relation's indexes in favor of using the REINDEX TABLE apparatus, which does the same thing simpler and faster. Also, make TRUNCATE not use cluster.c at all, but just assign a new relfilenode and REINDEX. This partially addresses Hartmut Raschick's complaint from last December that 7.4's TRUNCATE is an order of magnitude slower than prior releases. By getting rid of a lot of unnecessary catalog updates, these changes buy back about a factor of two (on my system). The remaining overhead seems associated with creating and deleting storage files, which we may not be able to do much about without abandoning transaction safety for TRUNCATE.
2004-05-08 02:34:49 +02:00
if (!reindex_relation(heapOid, true))
Change some notices to warnings and vice versa according to criteria developed on -hackers.
2003-10-02 08:34:04 +02:00
ereport(NOTICE,
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
(errmsg("table \"%s\" has no indexes",
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
relation->relname)));
Implement reindex command
2000-02-18 10:30:20 +01:00
}
/*
* ReindexDatabase
* Recreate indexes of a database.
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
*
* To reduce the probability of deadlocks, each table is reindexed in a
* separate transaction, so we can release the lock on it right away.
Implement reindex command
2000-02-18 10:30:20 +01:00
*/
void
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
ReindexDatabase(const char *dbname, bool force /* currently unused */,
bool all)
Implement reindex command
2000-02-18 10:30:20 +01:00
{
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
Relation relationRelation;
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
HeapScanDesc scan;
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
HeapTuple tuple;
First phase of memory management rewrite (see backend/utils/mmgr/README for details). It doesn't really do that much yet, since there are no short-term memory contexts in the executor, but the infrastructure is in place and long-term contexts are handled reasonably. A few long- standing bugs have been fixed, such as 'VACUUM; anything' in a single query string crashing. Also, out-of-memory is now considered a recoverable ERROR, not FATAL. Eliminate a large amount of crufty, now-dead code in and around memory management. Fix problem with holding off SIGTRAP, SIGSEGV, etc in postmaster and backend startup.
2000-06-28 05:33:33 +02:00
MemoryContext private_context;
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
MemoryContext old;
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
List *relids = NIL;
ListCell *l;
Implement reindex command
2000-02-18 10:30:20 +01:00
AssertArg(dbname);
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
if (strcmp(dbname, get_database_name(MyDatabaseId)) != 0)
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("can only reindex the currently open database")));
First phase of memory management rewrite (see backend/utils/mmgr/README for details). It doesn't really do that much yet, since there are no short-term memory contexts in the executor, but the infrastructure is in place and long-term contexts are handled reasonably. A few long- standing bugs have been fixed, such as 'VACUUM; anything' in a single query string crashing. Also, out-of-memory is now considered a recoverable ERROR, not FATAL. Eliminate a large amount of crufty, now-dead code in and around memory management. Fix problem with holding off SIGTRAP, SIGSEGV, etc in postmaster and backend startup.
2000-06-28 05:33:33 +02:00
First batch of object rename commands.
2003-06-27 16:45:32 +02:00
if (!pg_database_ownercheck(MyDatabaseId, GetUserId()))
Adjust 'permission denied' messages to be more useful and consistent.
2003-08-01 02:15:26 +02:00
aclcheck_error(ACLCHECK_NOT_OWNER, ACL_KIND_DATABASE,
dbname);
Implement reindex command
2000-02-18 10:30:20 +01:00
First phase of memory management rewrite (see backend/utils/mmgr/README for details). It doesn't really do that much yet, since there are no short-term memory contexts in the executor, but the infrastructure is in place and long-term contexts are handled reasonably. A few long- standing bugs have been fixed, such as 'VACUUM; anything' in a single query string crashing. Also, out-of-memory is now considered a recoverable ERROR, not FATAL. Eliminate a large amount of crufty, now-dead code in and around memory management. Fix problem with holding off SIGTRAP, SIGSEGV, etc in postmaster and backend startup.
2000-06-28 05:33:33 +02:00
/*
pgindent run. Make it all clean.
2001-03-22 05:01:46 +01:00
* We cannot run inside a user transaction block; if we were inside a
* transaction, then our commit- and start-transaction-command calls
* would not have the intended effect!
First phase of memory management rewrite (see backend/utils/mmgr/README for details). It doesn't really do that much yet, since there are no short-term memory contexts in the executor, but the infrastructure is in place and long-term contexts are handled reasonably. A few long- standing bugs have been fixed, such as 'VACUUM; anything' in a single query string crashing. Also, out-of-memory is now considered a recoverable ERROR, not FATAL. Eliminate a large amount of crufty, now-dead code in and around memory management. Fix problem with holding off SIGTRAP, SIGSEGV, etc in postmaster and backend startup.
2000-06-28 05:33:33 +02:00
*/
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
PreventTransactionChain((void *) dbname, "REINDEX DATABASE");
Invert logic in pg_exec_query_string() so that we set a snapshot for all utility statement types *except* a short list, per discussion a few days ago. Add missing SetQuerySnapshot calls in VACUUM and REINDEX, and guard against calling REINDEX DATABASE from a function (has same problem as VACUUM).
2002-10-19 22:15:09 +02:00
First phase of memory management rewrite (see backend/utils/mmgr/README for details). It doesn't really do that much yet, since there are no short-term memory contexts in the executor, but the infrastructure is in place and long-term contexts are handled reasonably. A few long- standing bugs have been fixed, such as 'VACUUM; anything' in a single query string crashing. Also, out-of-memory is now considered a recoverable ERROR, not FATAL. Eliminate a large amount of crufty, now-dead code in and around memory management. Fix problem with holding off SIGTRAP, SIGSEGV, etc in postmaster and backend startup.
2000-06-28 05:33:33 +02:00
/*
pgindent run. Make it all clean.
2001-03-22 05:01:46 +01:00
* Create a memory context that will survive forced transaction
Portal and memory management infrastructure for extended query protocol. Both plannable queries and utility commands are now always executed within Portals, which have been revamped so that they can handle the load (they used to be good only for single SELECT queries). Restructure code to push command-completion-tag selection logic out of postgres.c, so that it won't have to be duplicated between simple and extended queries. initdb forced due to addition of a field to Query nodes.
2003-05-02 22:54:36 +02:00
* commits we do below. Since it is a child of PortalContext, it will
pgindent run. Make it all clean.
2001-03-22 05:01:46 +01:00
* go away eventually even if we suffer an error; there's no need for
* special abort cleanup logic.
First phase of memory management rewrite (see backend/utils/mmgr/README for details). It doesn't really do that much yet, since there are no short-term memory contexts in the executor, but the infrastructure is in place and long-term contexts are handled reasonably. A few long- standing bugs have been fixed, such as 'VACUUM; anything' in a single query string crashing. Also, out-of-memory is now considered a recoverable ERROR, not FATAL. Eliminate a large amount of crufty, now-dead code in and around memory management. Fix problem with holding off SIGTRAP, SIGSEGV, etc in postmaster and backend startup.
2000-06-28 05:33:33 +02:00
*/
Portal and memory management infrastructure for extended query protocol. Both plannable queries and utility commands are now always executed within Portals, which have been revamped so that they can handle the load (they used to be good only for single SELECT queries). Restructure code to push command-completion-tag selection logic out of postgres.c, so that it won't have to be duplicated between simple and extended queries. initdb forced due to addition of a field to Query nodes.
2003-05-02 22:54:36 +02:00
private_context = AllocSetContextCreate(PortalContext,
First phase of memory management rewrite (see backend/utils/mmgr/README for details). It doesn't really do that much yet, since there are no short-term memory contexts in the executor, but the infrastructure is in place and long-term contexts are handled reasonably. A few long- standing bugs have been fixed, such as 'VACUUM; anything' in a single query string crashing. Also, out-of-memory is now considered a recoverable ERROR, not FATAL. Eliminate a large amount of crufty, now-dead code in and around memory management. Fix problem with holding off SIGTRAP, SIGSEGV, etc in postmaster and backend startup.
2000-06-28 05:33:33 +02:00
"ReindexDatabase",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
Implement reindex command
2000-02-18 10:30:20 +01:00
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
/*
* We always want to reindex pg_class first. This ensures that if
* there is any corruption in pg_class' indexes, they will be fixed
* before we process any other tables. This is critical because
* reindexing itself will try to update pg_class.
*/
old = MemoryContextSwitchTo(private_context);
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
relids = lappend_oid(relids, RelOid_pg_class);
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
MemoryContextSwitchTo(old);
Some minor tweaks of REINDEX processing: grab exclusive lock a little earlier, make error checks more uniform.
2001-11-20 03:46:13 +01:00
/*
* Scan pg_class to build a list of the relations we need to reindex.
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
*
* We only consider plain relations here (toast rels will be processed
* indirectly by reindex_relation).
Some minor tweaks of REINDEX processing: grab exclusive lock a little earlier, make error checks more uniform.
2001-11-20 03:46:13 +01:00
*/
Implement reindex command
2000-02-18 10:30:20 +01:00
relationRelation = heap_openr(RelationRelationName, AccessShareLock);
Restructure indexscan API (index_beginscan, index_getnext) per yesterday's proposal to pghackers. Also remove unnecessary parameters to heap_beginscan, heap_rescan. I modified pg_proc.h to reflect the new numbers of parameters for the AM interface routines, but did not force an initdb because nothing actually looks at those fields.
2002-05-21 01:51:44 +02:00
scan = heap_beginscan(relationRelation, SnapshotNow, 0, NULL);
while ((tuple = heap_getnext(scan, ForwardScanDirection)) != NULL)
Implement reindex command
2000-02-18 10:30:20 +01:00
{
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
Form_pg_class classtuple = (Form_pg_class) GETSTRUCT(tuple);
if (classtuple->relkind != RELKIND_RELATION)
continue;
Cause REINDEX to regard TOAST tables as regular relations, not system tables that need special defenses. I believe this is okay even for TOAST tables that belong to system tables.
2002-08-29 17:56:20 +02:00
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
if (!all) /* only system tables? */
Implement reindex command
2000-02-18 10:30:20 +01:00
{
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
if (!IsSystemClass(classtuple))
Implement reindex command
2000-02-18 10:30:20 +01:00
continue;
}
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
if (IsUnderPostmaster) /* silently ignore shared tables */
Implement reindex command
2000-02-18 10:30:20 +01:00
{
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
if (classtuple->relisshared)
continue;
Implement reindex command
2000-02-18 10:30:20 +01:00
}
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
if (HeapTupleGetOid(tuple) == RelOid_pg_class)
continue; /* got it already */
old = MemoryContextSwitchTo(private_context);
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
relids = lappend_oid(relids, HeapTupleGetOid(tuple));
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
MemoryContextSwitchTo(old);
Implement reindex command
2000-02-18 10:30:20 +01:00
}
heap_endscan(scan);
heap_close(relationRelation, AccessShareLock);
Some minor tweaks of REINDEX processing: grab exclusive lock a little earlier, make error checks more uniform.
2001-11-20 03:46:13 +01:00
/* Now reindex each rel in a separate transaction */
Backend support for autocommit removed, per recent discussions. The only remnant of this failed experiment is that the server will take SET AUTOCOMMIT TO ON. Still TODO: provide some client-side autocommit logic in libpq.
2003-05-14 05:26:03 +02:00
CommitTransactionCommand();
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
foreach(l, relids)
Implement reindex command
2000-02-18 10:30:20 +01:00
{
Reimplement the linked list data structure used throughout the backend. In the past, we used a 'Lispy' linked list implementation: a "list" was merely a pointer to the head node of the list. The problem with that design is that it makes lappend() and length() linear time. This patch fixes that problem (and others) by maintaining a count of the list length and a pointer to the tail node along with each head node pointer. A "list" is now a pointer to a structure containing some meta-data about the list; the head and tail pointers in that structure refer to ListCell structures that maintain the actual linked list of nodes. The function names of the list API have also been changed to, I hope, be more logically consistent. By default, the old function names are still available; they will be disabled-by-default once the rest of the tree has been updated to use the new API names.
2004-05-26 06:41:50 +02:00
Oid relid = lfirst_oid(l);
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
Backend support for autocommit removed, per recent discussions. The only remnant of this failed experiment is that the server will take SET AUTOCOMMIT TO ON. Still TODO: provide some client-side autocommit logic in libpq.
2003-05-14 05:26:03 +02:00
StartTransactionCommand();
pgindent run.
2003-08-04 02:43:34 +02:00
SetQuerySnapshot(); /* might be needed for functions in
* indexes */
Get rid of cluster.c's apparatus for rebuilding a relation's indexes in favor of using the REINDEX TABLE apparatus, which does the same thing simpler and faster. Also, make TRUNCATE not use cluster.c at all, but just assign a new relfilenode and REINDEX. This partially addresses Hartmut Raschick's complaint from last December that 7.4's TRUNCATE is an order of magnitude slower than prior releases. By getting rid of a lot of unnecessary catalog updates, these changes buy back about a factor of two (on my system). The remaining overhead seems associated with creating and deleting storage files, which we may not be able to do much about without abandoning transaction safety for TRUNCATE.
2004-05-08 02:34:49 +02:00
if (reindex_relation(relid, true))
Another round of error message editing, covering backend/commands/.
2003-07-20 23:56:35 +02:00
ereport(NOTICE,
Repair some REINDEX problems per recent discussions. The relcache is now able to cope with assigning new relfilenode values to nailed-in-cache indexes, so they can be reindexed using the fully crash-safe method. This leaves only shared system indexes as special cases. Remove the 'index deactivation' code, since it provides no useful protection in the shared- index case. Require reindexing of shared indexes to be done in standalone mode, but remove other restrictions on REINDEX. -P (IgnoreSystemIndexes) now prevents using indexes for lookups, but does not disable index updates. It is therefore safe to allow from PGOPTIONS. Upshot: reindexing system catalogs can be done without a standalone backend for all cases except shared catalogs.
2003-09-24 20:54:02 +02:00
(errmsg("table \"%s\" was reindexed",
get_rel_name(relid))));
Backend support for autocommit removed, per recent discussions. The only remnant of this failed experiment is that the server will take SET AUTOCOMMIT TO ON. Still TODO: provide some client-side autocommit logic in libpq.
2003-05-14 05:26:03 +02:00
CommitTransactionCommand();
Implement reindex command
2000-02-18 10:30:20 +01:00
}
Backend support for autocommit removed, per recent discussions. The only remnant of this failed experiment is that the server will take SET AUTOCOMMIT TO ON. Still TODO: provide some client-side autocommit logic in libpq.
2003-05-14 05:26:03 +02:00
StartTransactionCommand();
First phase of memory management rewrite (see backend/utils/mmgr/README for details). It doesn't really do that much yet, since there are no short-term memory contexts in the executor, but the infrastructure is in place and long-term contexts are handled reasonably. A few long- standing bugs have been fixed, such as 'VACUUM; anything' in a single query string crashing. Also, out-of-memory is now considered a recoverable ERROR, not FATAL. Eliminate a large amount of crufty, now-dead code in and around memory management. Fix problem with holding off SIGTRAP, SIGSEGV, etc in postmaster and backend startup.
2000-06-28 05:33:33 +02:00
MemoryContextDelete(private_context);
Implement reindex command
2000-02-18 10:30:20 +01:00
}