2016-03-24 03:01:35 +01:00
|
|
|
/*-------------------------------------------------------------------------
|
|
|
|
*
|
|
|
|
* amcmds.c
|
|
|
|
* Routines for SQL commands that manipulate access methods.
|
|
|
|
*
|
2020-01-01 18:21:45 +01:00
|
|
|
* Portions Copyright (c) 1996-2020, PostgreSQL Global Development Group
|
2016-03-24 03:01:35 +01:00
|
|
|
* Portions Copyright (c) 1994, Regents of the University of California
|
|
|
|
*
|
|
|
|
*
|
|
|
|
* IDENTIFICATION
|
|
|
|
* src/backend/commands/amcmds.c
|
|
|
|
*-------------------------------------------------------------------------
|
|
|
|
*/
|
|
|
|
#include "postgres.h"
|
|
|
|
|
|
|
|
#include "access/htup_details.h"
|
2019-01-21 19:18:20 +01:00
|
|
|
#include "access/table.h"
|
Remove WITH OIDS support, change oid catalog column visibility.
Previously tables declared WITH OIDS, including a significant fraction
of the catalog tables, stored the oid column not as a normal column,
but as part of the tuple header.
This special column was not shown by default, which was somewhat odd,
as it's often (consider e.g. pg_class.oid) one of the more important
parts of a row. Neither pg_dump nor COPY included the contents of the
oid column by default.
The fact that the oid column was not an ordinary column necessitated a
significant amount of special case code to support oid columns. That
already was painful for the existing, but upcoming work aiming to make
table storage pluggable, would have required expanding and duplicating
that "specialness" significantly.
WITH OIDS has been deprecated since 2005 (commit ff02d0a05280e0).
Remove it.
Removing includes:
- CREATE TABLE and ALTER TABLE syntax for declaring the table to be
WITH OIDS has been removed (WITH (oids[ = true]) will error out)
- pg_dump does not support dumping tables declared WITH OIDS and will
issue a warning when dumping one (and ignore the oid column).
- restoring an pg_dump archive with pg_restore will warn when
restoring a table with oid contents (and ignore the oid column)
- COPY will refuse to load binary dump that includes oids.
- pg_upgrade will error out when encountering tables declared WITH
OIDS, they have to be altered to remove the oid column first.
- Functionality to access the oid of the last inserted row (like
plpgsql's RESULT_OID, spi's SPI_lastoid, ...) has been removed.
The syntax for declaring a table WITHOUT OIDS (or WITH (oids = false)
for CREATE TABLE) is still supported. While that requires a bit of
support code, it seems unnecessary to break applications / dumps that
do not use oids, and are explicit about not using them.
The biggest user of WITH OID columns was postgres' catalog. This
commit changes all 'magic' oid columns to be columns that are normally
declared and stored. To reduce unnecessary query breakage all the
newly added columns are still named 'oid', even if a table's column
naming scheme would indicate 'reloid' or such. This obviously
requires adapting a lot code, mostly replacing oid access via
HeapTupleGetOid() with access to the underlying Form_pg_*->oid column.
The bootstrap process now assigns oids for all oid columns in
genbki.pl that do not have an explicit value (starting at the largest
oid previously used), only oids assigned later by oids will be above
FirstBootstrapObjectId. As the oid column now is a normal column the
special bootstrap syntax for oids has been removed.
Oids are not automatically assigned during insertion anymore, all
backend code explicitly assigns oids with GetNewOidWithIndex(). For
the rare case that insertions into the catalog via SQL are called for
the new pg_nextoid() function can be used (which only works on catalog
tables).
The fact that oid columns on system tables are now normal columns
means that they will be included in the set of columns expanded
by * (i.e. SELECT * FROM pg_class will now include the table's oid,
previously it did not). It'd not technically be hard to hide oid
column by default, but that'd mean confusing behavior would either
have to be carried forward forever, or it'd cause breakage down the
line.
While it's not unlikely that further adjustments are needed, the
scope/invasiveness of the patch makes it worthwhile to get merge this
now. It's painful to maintain externally, too complicated to commit
after the code code freeze, and a dependency of a number of other
patches.
Catversion bump, for obvious reasons.
Author: Andres Freund, with contributions by John Naylor
Discussion: https://postgr.es/m/20180930034810.ywp2c7awz7opzcfr@alap3.anarazel.de
2018-11-21 00:36:57 +01:00
|
|
|
#include "catalog/catalog.h"
|
2016-03-24 03:01:35 +01:00
|
|
|
#include "catalog/dependency.h"
|
|
|
|
#include "catalog/indexing.h"
|
|
|
|
#include "catalog/pg_am.h"
|
|
|
|
#include "catalog/pg_proc.h"
|
|
|
|
#include "catalog/pg_type.h"
|
|
|
|
#include "commands/defrem.h"
|
|
|
|
#include "miscadmin.h"
|
|
|
|
#include "parser/parse_func.h"
|
|
|
|
#include "utils/builtins.h"
|
|
|
|
#include "utils/lsyscache.h"
|
|
|
|
#include "utils/rel.h"
|
|
|
|
#include "utils/syscache.h"
|
|
|
|
|
|
|
|
|
tableam: introduce table AM infrastructure.
This introduces the concept of table access methods, i.e. CREATE
ACCESS METHOD ... TYPE TABLE and
CREATE TABLE ... USING (storage-engine).
No table access functionality is delegated to table AMs as of this
commit, that'll be done in following commits.
Subsequent commits will incrementally abstract table access
functionality to be routed through table access methods. That change
is too large to be reviewed & committed at once, so it'll be done
incrementally.
Docs will be updated at the end, as adding them incrementally would
likely make them less coherent, and definitely is a lot more work,
without a lot of benefit.
Table access methods are specified similar to index access methods,
i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with
callbacks. In contrast to index AMs that struct needs to live as long
as a backend, typically that's achieved by just returning a pointer to
a constant struct.
Psql's \d+ now displays a table's access method. That can be disabled
with HIDE_TABLEAM=true, which is mainly useful so regression tests can
be run against different AMs. It's quite possible that this behaviour
still needs to be fine tuned.
For now it's not allowed to set a table AM for a partitioned table, as
we've not resolved how partitions would inherit that. Disallowing
allows us to introduce, if we decide that's the way forward, such a
behaviour without a compatibility break.
Catversion bumped, to add the heap table AM and references to it.
Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others
Discussion:
https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de
https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
|
|
|
static Oid lookup_am_handler_func(List *handler_name, char amtype);
|
2016-03-24 22:22:24 +01:00
|
|
|
static const char *get_am_type_string(char amtype);
|
2016-03-24 03:01:35 +01:00
|
|
|
|
|
|
|
|
|
|
|
/*
|
2017-02-06 10:33:58 +01:00
|
|
|
* CreateAccessMethod
|
2016-03-24 03:01:35 +01:00
|
|
|
* Registers a new access method.
|
|
|
|
*/
|
|
|
|
ObjectAddress
|
|
|
|
CreateAccessMethod(CreateAmStmt *stmt)
|
|
|
|
{
|
|
|
|
Relation rel;
|
|
|
|
ObjectAddress myself;
|
|
|
|
ObjectAddress referenced;
|
|
|
|
Oid amoid;
|
|
|
|
Oid amhandler;
|
|
|
|
bool nulls[Natts_pg_am];
|
|
|
|
Datum values[Natts_pg_am];
|
|
|
|
HeapTuple tup;
|
|
|
|
|
2019-01-21 19:32:19 +01:00
|
|
|
rel = table_open(AccessMethodRelationId, RowExclusiveLock);
|
2016-03-24 03:01:35 +01:00
|
|
|
|
|
|
|
/* Must be super user */
|
|
|
|
if (!superuser())
|
|
|
|
ereport(ERROR,
|
|
|
|
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
|
|
|
|
errmsg("permission denied to create access method \"%s\"",
|
|
|
|
stmt->amname),
|
|
|
|
errhint("Must be superuser to create an access method.")));
|
|
|
|
|
|
|
|
/* Check if name is used */
|
2019-05-22 18:55:34 +02:00
|
|
|
amoid = GetSysCacheOid1(AMNAME, Anum_pg_am_oid,
|
Remove WITH OIDS support, change oid catalog column visibility.
Previously tables declared WITH OIDS, including a significant fraction
of the catalog tables, stored the oid column not as a normal column,
but as part of the tuple header.
This special column was not shown by default, which was somewhat odd,
as it's often (consider e.g. pg_class.oid) one of the more important
parts of a row. Neither pg_dump nor COPY included the contents of the
oid column by default.
The fact that the oid column was not an ordinary column necessitated a
significant amount of special case code to support oid columns. That
already was painful for the existing, but upcoming work aiming to make
table storage pluggable, would have required expanding and duplicating
that "specialness" significantly.
WITH OIDS has been deprecated since 2005 (commit ff02d0a05280e0).
Remove it.
Removing includes:
- CREATE TABLE and ALTER TABLE syntax for declaring the table to be
WITH OIDS has been removed (WITH (oids[ = true]) will error out)
- pg_dump does not support dumping tables declared WITH OIDS and will
issue a warning when dumping one (and ignore the oid column).
- restoring an pg_dump archive with pg_restore will warn when
restoring a table with oid contents (and ignore the oid column)
- COPY will refuse to load binary dump that includes oids.
- pg_upgrade will error out when encountering tables declared WITH
OIDS, they have to be altered to remove the oid column first.
- Functionality to access the oid of the last inserted row (like
plpgsql's RESULT_OID, spi's SPI_lastoid, ...) has been removed.
The syntax for declaring a table WITHOUT OIDS (or WITH (oids = false)
for CREATE TABLE) is still supported. While that requires a bit of
support code, it seems unnecessary to break applications / dumps that
do not use oids, and are explicit about not using them.
The biggest user of WITH OID columns was postgres' catalog. This
commit changes all 'magic' oid columns to be columns that are normally
declared and stored. To reduce unnecessary query breakage all the
newly added columns are still named 'oid', even if a table's column
naming scheme would indicate 'reloid' or such. This obviously
requires adapting a lot code, mostly replacing oid access via
HeapTupleGetOid() with access to the underlying Form_pg_*->oid column.
The bootstrap process now assigns oids for all oid columns in
genbki.pl that do not have an explicit value (starting at the largest
oid previously used), only oids assigned later by oids will be above
FirstBootstrapObjectId. As the oid column now is a normal column the
special bootstrap syntax for oids has been removed.
Oids are not automatically assigned during insertion anymore, all
backend code explicitly assigns oids with GetNewOidWithIndex(). For
the rare case that insertions into the catalog via SQL are called for
the new pg_nextoid() function can be used (which only works on catalog
tables).
The fact that oid columns on system tables are now normal columns
means that they will be included in the set of columns expanded
by * (i.e. SELECT * FROM pg_class will now include the table's oid,
previously it did not). It'd not technically be hard to hide oid
column by default, but that'd mean confusing behavior would either
have to be carried forward forever, or it'd cause breakage down the
line.
While it's not unlikely that further adjustments are needed, the
scope/invasiveness of the patch makes it worthwhile to get merge this
now. It's painful to maintain externally, too complicated to commit
after the code code freeze, and a dependency of a number of other
patches.
Catversion bump, for obvious reasons.
Author: Andres Freund, with contributions by John Naylor
Discussion: https://postgr.es/m/20180930034810.ywp2c7awz7opzcfr@alap3.anarazel.de
2018-11-21 00:36:57 +01:00
|
|
|
CStringGetDatum(stmt->amname));
|
2016-03-24 03:01:35 +01:00
|
|
|
if (OidIsValid(amoid))
|
|
|
|
{
|
|
|
|
ereport(ERROR,
|
|
|
|
(errcode(ERRCODE_DUPLICATE_OBJECT),
|
|
|
|
errmsg("access method \"%s\" already exists",
|
|
|
|
stmt->amname)));
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Get the handler function oid, verifying the AM type while at it.
|
|
|
|
*/
|
tableam: introduce table AM infrastructure.
This introduces the concept of table access methods, i.e. CREATE
ACCESS METHOD ... TYPE TABLE and
CREATE TABLE ... USING (storage-engine).
No table access functionality is delegated to table AMs as of this
commit, that'll be done in following commits.
Subsequent commits will incrementally abstract table access
functionality to be routed through table access methods. That change
is too large to be reviewed & committed at once, so it'll be done
incrementally.
Docs will be updated at the end, as adding them incrementally would
likely make them less coherent, and definitely is a lot more work,
without a lot of benefit.
Table access methods are specified similar to index access methods,
i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with
callbacks. In contrast to index AMs that struct needs to live as long
as a backend, typically that's achieved by just returning a pointer to
a constant struct.
Psql's \d+ now displays a table's access method. That can be disabled
with HIDE_TABLEAM=true, which is mainly useful so regression tests can
be run against different AMs. It's quite possible that this behaviour
still needs to be fine tuned.
For now it's not allowed to set a table AM for a partitioned table, as
we've not resolved how partitions would inherit that. Disallowing
allows us to introduce, if we decide that's the way forward, such a
behaviour without a compatibility break.
Catversion bumped, to add the heap table AM and references to it.
Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others
Discussion:
https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de
https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
|
|
|
amhandler = lookup_am_handler_func(stmt->handler_name, stmt->amtype);
|
2016-03-24 03:01:35 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Insert tuple into pg_am.
|
|
|
|
*/
|
|
|
|
memset(values, 0, sizeof(values));
|
|
|
|
memset(nulls, false, sizeof(nulls));
|
|
|
|
|
Remove WITH OIDS support, change oid catalog column visibility.
Previously tables declared WITH OIDS, including a significant fraction
of the catalog tables, stored the oid column not as a normal column,
but as part of the tuple header.
This special column was not shown by default, which was somewhat odd,
as it's often (consider e.g. pg_class.oid) one of the more important
parts of a row. Neither pg_dump nor COPY included the contents of the
oid column by default.
The fact that the oid column was not an ordinary column necessitated a
significant amount of special case code to support oid columns. That
already was painful for the existing, but upcoming work aiming to make
table storage pluggable, would have required expanding and duplicating
that "specialness" significantly.
WITH OIDS has been deprecated since 2005 (commit ff02d0a05280e0).
Remove it.
Removing includes:
- CREATE TABLE and ALTER TABLE syntax for declaring the table to be
WITH OIDS has been removed (WITH (oids[ = true]) will error out)
- pg_dump does not support dumping tables declared WITH OIDS and will
issue a warning when dumping one (and ignore the oid column).
- restoring an pg_dump archive with pg_restore will warn when
restoring a table with oid contents (and ignore the oid column)
- COPY will refuse to load binary dump that includes oids.
- pg_upgrade will error out when encountering tables declared WITH
OIDS, they have to be altered to remove the oid column first.
- Functionality to access the oid of the last inserted row (like
plpgsql's RESULT_OID, spi's SPI_lastoid, ...) has been removed.
The syntax for declaring a table WITHOUT OIDS (or WITH (oids = false)
for CREATE TABLE) is still supported. While that requires a bit of
support code, it seems unnecessary to break applications / dumps that
do not use oids, and are explicit about not using them.
The biggest user of WITH OID columns was postgres' catalog. This
commit changes all 'magic' oid columns to be columns that are normally
declared and stored. To reduce unnecessary query breakage all the
newly added columns are still named 'oid', even if a table's column
naming scheme would indicate 'reloid' or such. This obviously
requires adapting a lot code, mostly replacing oid access via
HeapTupleGetOid() with access to the underlying Form_pg_*->oid column.
The bootstrap process now assigns oids for all oid columns in
genbki.pl that do not have an explicit value (starting at the largest
oid previously used), only oids assigned later by oids will be above
FirstBootstrapObjectId. As the oid column now is a normal column the
special bootstrap syntax for oids has been removed.
Oids are not automatically assigned during insertion anymore, all
backend code explicitly assigns oids with GetNewOidWithIndex(). For
the rare case that insertions into the catalog via SQL are called for
the new pg_nextoid() function can be used (which only works on catalog
tables).
The fact that oid columns on system tables are now normal columns
means that they will be included in the set of columns expanded
by * (i.e. SELECT * FROM pg_class will now include the table's oid,
previously it did not). It'd not technically be hard to hide oid
column by default, but that'd mean confusing behavior would either
have to be carried forward forever, or it'd cause breakage down the
line.
While it's not unlikely that further adjustments are needed, the
scope/invasiveness of the patch makes it worthwhile to get merge this
now. It's painful to maintain externally, too complicated to commit
after the code code freeze, and a dependency of a number of other
patches.
Catversion bump, for obvious reasons.
Author: Andres Freund, with contributions by John Naylor
Discussion: https://postgr.es/m/20180930034810.ywp2c7awz7opzcfr@alap3.anarazel.de
2018-11-21 00:36:57 +01:00
|
|
|
amoid = GetNewOidWithIndex(rel, AmOidIndexId, Anum_pg_am_oid);
|
|
|
|
values[Anum_pg_am_oid - 1] = ObjectIdGetDatum(amoid);
|
2016-03-24 03:01:35 +01:00
|
|
|
values[Anum_pg_am_amname - 1] =
|
|
|
|
DirectFunctionCall1(namein, CStringGetDatum(stmt->amname));
|
|
|
|
values[Anum_pg_am_amhandler - 1] = ObjectIdGetDatum(amhandler);
|
|
|
|
values[Anum_pg_am_amtype - 1] = CharGetDatum(stmt->amtype);
|
|
|
|
|
|
|
|
tup = heap_form_tuple(RelationGetDescr(rel), values, nulls);
|
|
|
|
|
Remove WITH OIDS support, change oid catalog column visibility.
Previously tables declared WITH OIDS, including a significant fraction
of the catalog tables, stored the oid column not as a normal column,
but as part of the tuple header.
This special column was not shown by default, which was somewhat odd,
as it's often (consider e.g. pg_class.oid) one of the more important
parts of a row. Neither pg_dump nor COPY included the contents of the
oid column by default.
The fact that the oid column was not an ordinary column necessitated a
significant amount of special case code to support oid columns. That
already was painful for the existing, but upcoming work aiming to make
table storage pluggable, would have required expanding and duplicating
that "specialness" significantly.
WITH OIDS has been deprecated since 2005 (commit ff02d0a05280e0).
Remove it.
Removing includes:
- CREATE TABLE and ALTER TABLE syntax for declaring the table to be
WITH OIDS has been removed (WITH (oids[ = true]) will error out)
- pg_dump does not support dumping tables declared WITH OIDS and will
issue a warning when dumping one (and ignore the oid column).
- restoring an pg_dump archive with pg_restore will warn when
restoring a table with oid contents (and ignore the oid column)
- COPY will refuse to load binary dump that includes oids.
- pg_upgrade will error out when encountering tables declared WITH
OIDS, they have to be altered to remove the oid column first.
- Functionality to access the oid of the last inserted row (like
plpgsql's RESULT_OID, spi's SPI_lastoid, ...) has been removed.
The syntax for declaring a table WITHOUT OIDS (or WITH (oids = false)
for CREATE TABLE) is still supported. While that requires a bit of
support code, it seems unnecessary to break applications / dumps that
do not use oids, and are explicit about not using them.
The biggest user of WITH OID columns was postgres' catalog. This
commit changes all 'magic' oid columns to be columns that are normally
declared and stored. To reduce unnecessary query breakage all the
newly added columns are still named 'oid', even if a table's column
naming scheme would indicate 'reloid' or such. This obviously
requires adapting a lot code, mostly replacing oid access via
HeapTupleGetOid() with access to the underlying Form_pg_*->oid column.
The bootstrap process now assigns oids for all oid columns in
genbki.pl that do not have an explicit value (starting at the largest
oid previously used), only oids assigned later by oids will be above
FirstBootstrapObjectId. As the oid column now is a normal column the
special bootstrap syntax for oids has been removed.
Oids are not automatically assigned during insertion anymore, all
backend code explicitly assigns oids with GetNewOidWithIndex(). For
the rare case that insertions into the catalog via SQL are called for
the new pg_nextoid() function can be used (which only works on catalog
tables).
The fact that oid columns on system tables are now normal columns
means that they will be included in the set of columns expanded
by * (i.e. SELECT * FROM pg_class will now include the table's oid,
previously it did not). It'd not technically be hard to hide oid
column by default, but that'd mean confusing behavior would either
have to be carried forward forever, or it'd cause breakage down the
line.
While it's not unlikely that further adjustments are needed, the
scope/invasiveness of the patch makes it worthwhile to get merge this
now. It's painful to maintain externally, too complicated to commit
after the code code freeze, and a dependency of a number of other
patches.
Catversion bump, for obvious reasons.
Author: Andres Freund, with contributions by John Naylor
Discussion: https://postgr.es/m/20180930034810.ywp2c7awz7opzcfr@alap3.anarazel.de
2018-11-21 00:36:57 +01:00
|
|
|
CatalogTupleInsert(rel, tup);
|
2016-03-24 03:01:35 +01:00
|
|
|
heap_freetuple(tup);
|
|
|
|
|
|
|
|
myself.classId = AccessMethodRelationId;
|
|
|
|
myself.objectId = amoid;
|
|
|
|
myself.objectSubId = 0;
|
|
|
|
|
|
|
|
/* Record dependency on handler function */
|
|
|
|
referenced.classId = ProcedureRelationId;
|
|
|
|
referenced.objectId = amhandler;
|
|
|
|
referenced.objectSubId = 0;
|
|
|
|
|
|
|
|
recordDependencyOn(&myself, &referenced, DEPENDENCY_NORMAL);
|
|
|
|
|
|
|
|
recordDependencyOnCurrentExtension(&myself, false);
|
|
|
|
|
2019-01-21 19:32:19 +01:00
|
|
|
table_close(rel, RowExclusiveLock);
|
2016-03-24 03:01:35 +01:00
|
|
|
|
|
|
|
return myself;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Guts of access method deletion.
|
|
|
|
*/
|
|
|
|
void
|
|
|
|
RemoveAccessMethodById(Oid amOid)
|
|
|
|
{
|
|
|
|
Relation relation;
|
|
|
|
HeapTuple tup;
|
|
|
|
|
|
|
|
if (!superuser())
|
|
|
|
ereport(ERROR,
|
|
|
|
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
|
|
|
|
errmsg("must be superuser to drop access methods")));
|
|
|
|
|
2019-01-21 19:32:19 +01:00
|
|
|
relation = table_open(AccessMethodRelationId, RowExclusiveLock);
|
2016-03-24 03:01:35 +01:00
|
|
|
|
|
|
|
tup = SearchSysCache1(AMOID, ObjectIdGetDatum(amOid));
|
|
|
|
if (!HeapTupleIsValid(tup))
|
|
|
|
elog(ERROR, "cache lookup failed for access method %u", amOid);
|
|
|
|
|
2017-02-01 22:13:30 +01:00
|
|
|
CatalogTupleDelete(relation, &tup->t_self);
|
2016-03-24 03:01:35 +01:00
|
|
|
|
|
|
|
ReleaseSysCache(tup);
|
|
|
|
|
2019-01-21 19:32:19 +01:00
|
|
|
table_close(relation, RowExclusiveLock);
|
2016-03-24 03:01:35 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* get_am_type_oid
|
2016-06-10 00:02:36 +02:00
|
|
|
* Worker for various get_am_*_oid variants
|
2016-03-24 03:01:35 +01:00
|
|
|
*
|
|
|
|
* If missing_ok is false, throw an error if access method not found. If
|
|
|
|
* true, just return InvalidOid.
|
|
|
|
*
|
|
|
|
* If amtype is not '\0', an error is raised if the AM found is not of the
|
|
|
|
* given type.
|
|
|
|
*/
|
|
|
|
static Oid
|
|
|
|
get_am_type_oid(const char *amname, char amtype, bool missing_ok)
|
|
|
|
{
|
|
|
|
HeapTuple tup;
|
|
|
|
Oid oid = InvalidOid;
|
|
|
|
|
|
|
|
tup = SearchSysCache1(AMNAME, CStringGetDatum(amname));
|
|
|
|
if (HeapTupleIsValid(tup))
|
|
|
|
{
|
|
|
|
Form_pg_am amform = (Form_pg_am) GETSTRUCT(tup);
|
|
|
|
|
|
|
|
if (amtype != '\0' &&
|
|
|
|
amform->amtype != amtype)
|
|
|
|
ereport(ERROR,
|
|
|
|
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
|
|
|
|
errmsg("access method \"%s\" is not of type %s",
|
|
|
|
NameStr(amform->amname),
|
|
|
|
get_am_type_string(amtype))));
|
|
|
|
|
Remove WITH OIDS support, change oid catalog column visibility.
Previously tables declared WITH OIDS, including a significant fraction
of the catalog tables, stored the oid column not as a normal column,
but as part of the tuple header.
This special column was not shown by default, which was somewhat odd,
as it's often (consider e.g. pg_class.oid) one of the more important
parts of a row. Neither pg_dump nor COPY included the contents of the
oid column by default.
The fact that the oid column was not an ordinary column necessitated a
significant amount of special case code to support oid columns. That
already was painful for the existing, but upcoming work aiming to make
table storage pluggable, would have required expanding and duplicating
that "specialness" significantly.
WITH OIDS has been deprecated since 2005 (commit ff02d0a05280e0).
Remove it.
Removing includes:
- CREATE TABLE and ALTER TABLE syntax for declaring the table to be
WITH OIDS has been removed (WITH (oids[ = true]) will error out)
- pg_dump does not support dumping tables declared WITH OIDS and will
issue a warning when dumping one (and ignore the oid column).
- restoring an pg_dump archive with pg_restore will warn when
restoring a table with oid contents (and ignore the oid column)
- COPY will refuse to load binary dump that includes oids.
- pg_upgrade will error out when encountering tables declared WITH
OIDS, they have to be altered to remove the oid column first.
- Functionality to access the oid of the last inserted row (like
plpgsql's RESULT_OID, spi's SPI_lastoid, ...) has been removed.
The syntax for declaring a table WITHOUT OIDS (or WITH (oids = false)
for CREATE TABLE) is still supported. While that requires a bit of
support code, it seems unnecessary to break applications / dumps that
do not use oids, and are explicit about not using them.
The biggest user of WITH OID columns was postgres' catalog. This
commit changes all 'magic' oid columns to be columns that are normally
declared and stored. To reduce unnecessary query breakage all the
newly added columns are still named 'oid', even if a table's column
naming scheme would indicate 'reloid' or such. This obviously
requires adapting a lot code, mostly replacing oid access via
HeapTupleGetOid() with access to the underlying Form_pg_*->oid column.
The bootstrap process now assigns oids for all oid columns in
genbki.pl that do not have an explicit value (starting at the largest
oid previously used), only oids assigned later by oids will be above
FirstBootstrapObjectId. As the oid column now is a normal column the
special bootstrap syntax for oids has been removed.
Oids are not automatically assigned during insertion anymore, all
backend code explicitly assigns oids with GetNewOidWithIndex(). For
the rare case that insertions into the catalog via SQL are called for
the new pg_nextoid() function can be used (which only works on catalog
tables).
The fact that oid columns on system tables are now normal columns
means that they will be included in the set of columns expanded
by * (i.e. SELECT * FROM pg_class will now include the table's oid,
previously it did not). It'd not technically be hard to hide oid
column by default, but that'd mean confusing behavior would either
have to be carried forward forever, or it'd cause breakage down the
line.
While it's not unlikely that further adjustments are needed, the
scope/invasiveness of the patch makes it worthwhile to get merge this
now. It's painful to maintain externally, too complicated to commit
after the code code freeze, and a dependency of a number of other
patches.
Catversion bump, for obvious reasons.
Author: Andres Freund, with contributions by John Naylor
Discussion: https://postgr.es/m/20180930034810.ywp2c7awz7opzcfr@alap3.anarazel.de
2018-11-21 00:36:57 +01:00
|
|
|
oid = amform->oid;
|
2016-03-24 03:01:35 +01:00
|
|
|
ReleaseSysCache(tup);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!OidIsValid(oid) && !missing_ok)
|
|
|
|
ereport(ERROR,
|
|
|
|
(errcode(ERRCODE_UNDEFINED_OBJECT),
|
|
|
|
errmsg("access method \"%s\" does not exist", amname)));
|
|
|
|
return oid;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* get_index_am_oid - given an access method name, look up its OID
|
|
|
|
* and verify it corresponds to an index AM.
|
|
|
|
*/
|
|
|
|
Oid
|
|
|
|
get_index_am_oid(const char *amname, bool missing_ok)
|
|
|
|
{
|
|
|
|
return get_am_type_oid(amname, AMTYPE_INDEX, missing_ok);
|
|
|
|
}
|
|
|
|
|
2019-04-05 02:17:50 +02:00
|
|
|
/*
|
|
|
|
* get_table_am_oid - given an access method name, look up its OID
|
|
|
|
* and verify it corresponds to an table AM.
|
|
|
|
*/
|
|
|
|
Oid
|
|
|
|
get_table_am_oid(const char *amname, bool missing_ok)
|
|
|
|
{
|
|
|
|
return get_am_type_oid(amname, AMTYPE_TABLE, missing_ok);
|
|
|
|
}
|
|
|
|
|
2016-03-24 03:01:35 +01:00
|
|
|
/*
|
|
|
|
* get_am_oid - given an access method name, look up its OID.
|
2016-06-10 00:02:36 +02:00
|
|
|
* The type is not checked.
|
2016-03-24 03:01:35 +01:00
|
|
|
*/
|
|
|
|
Oid
|
|
|
|
get_am_oid(const char *amname, bool missing_ok)
|
|
|
|
{
|
|
|
|
return get_am_type_oid(amname, '\0', missing_ok);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* get_am_name - given an access method OID name and type, look up its name.
|
|
|
|
*/
|
|
|
|
char *
|
|
|
|
get_am_name(Oid amOid)
|
|
|
|
{
|
|
|
|
HeapTuple tup;
|
|
|
|
char *result = NULL;
|
|
|
|
|
|
|
|
tup = SearchSysCache1(AMOID, ObjectIdGetDatum(amOid));
|
|
|
|
if (HeapTupleIsValid(tup))
|
|
|
|
{
|
|
|
|
Form_pg_am amform = (Form_pg_am) GETSTRUCT(tup);
|
|
|
|
|
|
|
|
result = pstrdup(NameStr(amform->amname));
|
|
|
|
ReleaseSysCache(tup);
|
|
|
|
}
|
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2016-03-24 22:22:24 +01:00
|
|
|
* Convert single-character access method type into string for error reporting.
|
2016-03-24 03:01:35 +01:00
|
|
|
*/
|
2016-03-24 22:22:24 +01:00
|
|
|
static const char *
|
2016-03-24 03:01:35 +01:00
|
|
|
get_am_type_string(char amtype)
|
|
|
|
{
|
|
|
|
switch (amtype)
|
|
|
|
{
|
|
|
|
case AMTYPE_INDEX:
|
|
|
|
return "INDEX";
|
tableam: introduce table AM infrastructure.
This introduces the concept of table access methods, i.e. CREATE
ACCESS METHOD ... TYPE TABLE and
CREATE TABLE ... USING (storage-engine).
No table access functionality is delegated to table AMs as of this
commit, that'll be done in following commits.
Subsequent commits will incrementally abstract table access
functionality to be routed through table access methods. That change
is too large to be reviewed & committed at once, so it'll be done
incrementally.
Docs will be updated at the end, as adding them incrementally would
likely make them less coherent, and definitely is a lot more work,
without a lot of benefit.
Table access methods are specified similar to index access methods,
i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with
callbacks. In contrast to index AMs that struct needs to live as long
as a backend, typically that's achieved by just returning a pointer to
a constant struct.
Psql's \d+ now displays a table's access method. That can be disabled
with HIDE_TABLEAM=true, which is mainly useful so regression tests can
be run against different AMs. It's quite possible that this behaviour
still needs to be fine tuned.
For now it's not allowed to set a table AM for a partitioned table, as
we've not resolved how partitions would inherit that. Disallowing
allows us to introduce, if we decide that's the way forward, such a
behaviour without a compatibility break.
Catversion bumped, to add the heap table AM and references to it.
Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others
Discussion:
https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de
https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
|
|
|
case AMTYPE_TABLE:
|
|
|
|
return "TABLE";
|
2016-03-24 03:01:35 +01:00
|
|
|
default:
|
|
|
|
/* shouldn't happen */
|
|
|
|
elog(ERROR, "invalid access method type '%c'", amtype);
|
2016-03-24 22:22:24 +01:00
|
|
|
return NULL; /* keep compiler quiet */
|
2016-03-24 03:01:35 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Convert a handler function name to an Oid. If the return type of the
|
|
|
|
* function doesn't match the given AM type, an error is raised.
|
|
|
|
*
|
|
|
|
* This function either return valid function Oid or throw an error.
|
|
|
|
*/
|
|
|
|
static Oid
|
tableam: introduce table AM infrastructure.
This introduces the concept of table access methods, i.e. CREATE
ACCESS METHOD ... TYPE TABLE and
CREATE TABLE ... USING (storage-engine).
No table access functionality is delegated to table AMs as of this
commit, that'll be done in following commits.
Subsequent commits will incrementally abstract table access
functionality to be routed through table access methods. That change
is too large to be reviewed & committed at once, so it'll be done
incrementally.
Docs will be updated at the end, as adding them incrementally would
likely make them less coherent, and definitely is a lot more work,
without a lot of benefit.
Table access methods are specified similar to index access methods,
i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with
callbacks. In contrast to index AMs that struct needs to live as long
as a backend, typically that's achieved by just returning a pointer to
a constant struct.
Psql's \d+ now displays a table's access method. That can be disabled
with HIDE_TABLEAM=true, which is mainly useful so regression tests can
be run against different AMs. It's quite possible that this behaviour
still needs to be fine tuned.
For now it's not allowed to set a table AM for a partitioned table, as
we've not resolved how partitions would inherit that. Disallowing
allows us to introduce, if we decide that's the way forward, such a
behaviour without a compatibility break.
Catversion bumped, to add the heap table AM and references to it.
Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others
Discussion:
https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de
https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
|
|
|
lookup_am_handler_func(List *handler_name, char amtype)
|
2016-03-24 03:01:35 +01:00
|
|
|
{
|
|
|
|
Oid handlerOid;
|
tableam: introduce table AM infrastructure.
This introduces the concept of table access methods, i.e. CREATE
ACCESS METHOD ... TYPE TABLE and
CREATE TABLE ... USING (storage-engine).
No table access functionality is delegated to table AMs as of this
commit, that'll be done in following commits.
Subsequent commits will incrementally abstract table access
functionality to be routed through table access methods. That change
is too large to be reviewed & committed at once, so it'll be done
incrementally.
Docs will be updated at the end, as adding them incrementally would
likely make them less coherent, and definitely is a lot more work,
without a lot of benefit.
Table access methods are specified similar to index access methods,
i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with
callbacks. In contrast to index AMs that struct needs to live as long
as a backend, typically that's achieved by just returning a pointer to
a constant struct.
Psql's \d+ now displays a table's access method. That can be disabled
with HIDE_TABLEAM=true, which is mainly useful so regression tests can
be run against different AMs. It's quite possible that this behaviour
still needs to be fine tuned.
For now it's not allowed to set a table AM for a partitioned table, as
we've not resolved how partitions would inherit that. Disallowing
allows us to introduce, if we decide that's the way forward, such a
behaviour without a compatibility break.
Catversion bumped, to add the heap table AM and references to it.
Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others
Discussion:
https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de
https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
|
|
|
Oid funcargtypes[1] = {INTERNALOID};
|
|
|
|
Oid expectedType = InvalidOid;
|
2016-03-24 03:01:35 +01:00
|
|
|
|
|
|
|
if (handler_name == NIL)
|
|
|
|
ereport(ERROR,
|
|
|
|
(errcode(ERRCODE_UNDEFINED_FUNCTION),
|
|
|
|
errmsg("handler function is not specified")));
|
|
|
|
|
|
|
|
/* handlers have one argument of type internal */
|
|
|
|
handlerOid = LookupFuncName(handler_name, 1, funcargtypes, false);
|
|
|
|
|
|
|
|
/* check that handler has the correct return type */
|
|
|
|
switch (amtype)
|
|
|
|
{
|
|
|
|
case AMTYPE_INDEX:
|
tableam: introduce table AM infrastructure.
This introduces the concept of table access methods, i.e. CREATE
ACCESS METHOD ... TYPE TABLE and
CREATE TABLE ... USING (storage-engine).
No table access functionality is delegated to table AMs as of this
commit, that'll be done in following commits.
Subsequent commits will incrementally abstract table access
functionality to be routed through table access methods. That change
is too large to be reviewed & committed at once, so it'll be done
incrementally.
Docs will be updated at the end, as adding them incrementally would
likely make them less coherent, and definitely is a lot more work,
without a lot of benefit.
Table access methods are specified similar to index access methods,
i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with
callbacks. In contrast to index AMs that struct needs to live as long
as a backend, typically that's achieved by just returning a pointer to
a constant struct.
Psql's \d+ now displays a table's access method. That can be disabled
with HIDE_TABLEAM=true, which is mainly useful so regression tests can
be run against different AMs. It's quite possible that this behaviour
still needs to be fine tuned.
For now it's not allowed to set a table AM for a partitioned table, as
we've not resolved how partitions would inherit that. Disallowing
allows us to introduce, if we decide that's the way forward, such a
behaviour without a compatibility break.
Catversion bumped, to add the heap table AM and references to it.
Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others
Discussion:
https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de
https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
|
|
|
expectedType = INDEX_AM_HANDLEROID;
|
|
|
|
break;
|
|
|
|
case AMTYPE_TABLE:
|
|
|
|
expectedType = TABLE_AM_HANDLEROID;
|
2016-03-24 03:01:35 +01:00
|
|
|
break;
|
|
|
|
default:
|
|
|
|
elog(ERROR, "unrecognized access method type \"%c\"", amtype);
|
|
|
|
}
|
|
|
|
|
tableam: introduce table AM infrastructure.
This introduces the concept of table access methods, i.e. CREATE
ACCESS METHOD ... TYPE TABLE and
CREATE TABLE ... USING (storage-engine).
No table access functionality is delegated to table AMs as of this
commit, that'll be done in following commits.
Subsequent commits will incrementally abstract table access
functionality to be routed through table access methods. That change
is too large to be reviewed & committed at once, so it'll be done
incrementally.
Docs will be updated at the end, as adding them incrementally would
likely make them less coherent, and definitely is a lot more work,
without a lot of benefit.
Table access methods are specified similar to index access methods,
i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with
callbacks. In contrast to index AMs that struct needs to live as long
as a backend, typically that's achieved by just returning a pointer to
a constant struct.
Psql's \d+ now displays a table's access method. That can be disabled
with HIDE_TABLEAM=true, which is mainly useful so regression tests can
be run against different AMs. It's quite possible that this behaviour
still needs to be fine tuned.
For now it's not allowed to set a table AM for a partitioned table, as
we've not resolved how partitions would inherit that. Disallowing
allows us to introduce, if we decide that's the way forward, such a
behaviour without a compatibility break.
Catversion bumped, to add the heap table AM and references to it.
Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others
Discussion:
https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.de
https://postgr.es/m/20160812231527.GA690404@alvherre.pgsql
https://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.de
https://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
2019-03-06 18:54:38 +01:00
|
|
|
if (get_func_rettype(handlerOid) != expectedType)
|
|
|
|
ereport(ERROR,
|
|
|
|
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
|
|
|
|
errmsg("function %s must return type %s",
|
|
|
|
get_func_name(handlerOid),
|
|
|
|
format_type_extended(expectedType, -1, 0))));
|
|
|
|
|
2016-03-24 03:01:35 +01:00
|
|
|
return handlerOid;
|
|
|
|
}
|