pgsql: Remove WITH OIDS support, change oid catalog column visibility.

Started by Andres Freundover 7 years ago11 messageshackers

andres@anarazel.de

over 7 years ago

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: /messages/by-id/20180930034810.ywp2c7awz7opzcfr@alap3.anarazel.de

Branch
------
master

Details
-------
https://git.postgresql.org/pg/commitdiff/578b229718e8f15fa779e20f086c4b6bb3776106

Christoph Berg

myon@debian.org

over 7 years ago

In reply to: Andres Freund (#1)

Re: pgsql: Remove WITH OIDS support, change oid catalog column visibility.

Re: Andres Freund 2018-11-21 <E1gPG3j-0003cy-03@gemulon.postgresql.org>

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.

postgres=# \d+ pg_class
[...]
Indexe:
"pg_class_oid_index" UNIQUE, btree (oid)

Now that oid is a proper column, shouldn't that be a PRIMARY KEY?
(Just for the looks.)

Christoph

Gavin Flower

GavinFlower@archidevsys.co.nz

over 7 years ago

In reply to: Christoph Berg (#2)

Re: pgsql: Remove WITH OIDS support, change oid catalog column visibility.

On 21/11/2018 21:20, Christoph Berg wrote:

Re: Andres Freund 2018-11-21 <E1gPG3j-0003cy-03@gemulon.postgresql.org>

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.

postgres=# \d+ pg_class
[...]
Indexe:
"pg_class_oid_index" UNIQUE, btree (oid)

Now that oid is a proper column, shouldn't that be a PRIMARY KEY?
(Just for the looks.)

Christoph

Curious, is there a reason 'Index' is spelt with a trailing 'e'?

Cheers,
Gavin

Christoph Berg

myon@debian.org

over 7 years ago

In reply to: Gavin Flower (#3)

Re: pgsql: Remove WITH OIDS support, change oid catalog column visibility.

Re: Gavin Flower 2018-11-21 <f6f8ce89-e152-0c80-85a5-41aeee08603e@archidevsys.co.nz>

Curious, is there a reason 'Index' is spelt with a trailing 'e'?

LANG=de_DE.UTF-8

Christoph

Peter Eisentraut

peter_e@gmx.net

over 7 years ago

In reply to: Christoph Berg (#2)

Re: pgsql: Remove WITH OIDS support, change oid catalog column visibility.

On 21/11/2018 09:20, Christoph Berg wrote:

Re: Andres Freund 2018-11-21 <E1gPG3j-0003cy-03@gemulon.postgresql.org>

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.

postgres=# \d+ pg_class
[...]
Indexe:
"pg_class_oid_index" UNIQUE, btree (oid)

Now that oid is a proper column, shouldn't that be a PRIMARY KEY?
(Just for the looks.)

This is an independent consideration. There was nothing before that
prevented an index on an oid column from being a primary key that isn't
still there.

--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Tom Lane

tgl@sss.pgh.pa.us

over 7 years ago

In reply to: Peter Eisentraut (#5)

Re: pgsql: Remove WITH OIDS support, change oid catalog column visibility.

I noticed that this patch has broken restores of existing dump files:

psql:testbed.public.backup:82: ERROR: unrecognized configuration parameter "default_with_oids"

Quite aside from the fact that this won't work if the user tries to
restore in single-transaction or no-error mode, this is really really
unhelpful because it's impossible to tell whether the error is
ignorable or not, except by digging through the dump file.

What you should do is restore that GUC, but add a check-hook that throws
an error for an attempt to set it to "true".

regards, tom lane

Andres Freund

andres@anarazel.de

over 7 years ago

In reply to: Tom Lane (#6)

Re: pgsql: Remove WITH OIDS support, change oid catalog column visibility.

Hi,

On 2019-01-03 13:40:42 -0500, Tom Lane wrote:

I noticed that this patch has broken restores of existing dump files:

psql:testbed.public.backup:82: ERROR: unrecognized configuration parameter "default_with_oids"

Quite aside from the fact that this won't work if the user tries to
restore in single-transaction or no-error mode, this is really really
unhelpful because it's impossible to tell whether the error is
ignorable or not, except by digging through the dump file.

What you should do is restore that GUC, but add a check-hook that throws
an error for an attempt to set it to "true".

Makes sense, I (or a colleague) will look into that next week. Wanna get my
head out of pluggable storage first...

Greetings,

Andres Freund

Amit Khandekar

amitdkhan.pg@gmail.com

over 7 years ago

In reply to: Andres Freund (#7)

Re: pgsql: Remove WITH OIDS support, change oid catalog column visibility.

On Sat, 5 Jan 2019 at 02:09, Andres Freund <andres@anarazel.de> wrote:

Hi,

On 2019-01-03 13:40:42 -0500, Tom Lane wrote:

I noticed that this patch has broken restores of existing dump files:

psql:testbed.public.backup:82: ERROR: unrecognized configuration parameter "default_with_oids"

Quite aside from the fact that this won't work if the user tries to
restore in single-transaction or no-error mode, this is really really
unhelpful because it's impossible to tell whether the error is
ignorable or not, except by digging through the dump file.

What you should do is restore that GUC, but add a check-hook that throws
an error for an attempt to set it to "true".

Attached is a patch that does this.

Makes sense, I (or a colleague) will look into that next week. Wanna get my
head out of pluggable storage first...

Greetings,

Andres Freund

--
Thanks,
-Amit Khandekar
EnterpriseDB Corporation
The Postgres Database Company

Andres Freund

andres@anarazel.de

about 7 years ago

In reply to: Amit Khandekar (#8)

Re: pgsql: Remove WITH OIDS support, change oid catalog column visibility.

Hi,

On 2019-01-07 17:03:20 +0530, Amit Khandekar wrote:

On Sat, 5 Jan 2019 at 02:09, Andres Freund <andres@anarazel.de> wrote:

On 2019-01-03 13:40:42 -0500, Tom Lane wrote:

I noticed that this patch has broken restores of existing dump files:

psql:testbed.public.backup:82: ERROR: unrecognized configuration parameter "default_with_oids"

Quite aside from the fact that this won't work if the user tries to
restore in single-transaction or no-error mode, this is really really
unhelpful because it's impossible to tell whether the error is
ignorable or not, except by digging through the dump file.

What you should do is restore that GUC, but add a check-hook that throws
an error for an attempt to set it to "true".

Attached is a patch that does this.

Thanks! I've pushed this, after some minor editorialization (hiding the
GUC, shortened description, shortened error message).

Regards,

Andres

#10

Peter Eisentraut

peter_e@gmx.net

about 7 years ago

In reply to: Andres Freund (#1)

Re: pgsql: Remove WITH OIDS support, change oid catalog column visibility.

On 2018-11-21 01:07, Andres Freund wrote:

Remove WITH OIDS support, change oid catalog column visibility.

I think you may have accidentally duplicated a line in this patch:

@@ -1602,20 +1602,9 @@ ExecFetchSlotHeapTupleDatum(TupleTableSlot *slot)
 void
 ExecInitResultTypeTL(PlanState *planstate)
 {
-   bool        hasoid;
-   TupleDesc   tupDesc;
-
-   if (ExecContextForcesOids(planstate, &hasoid))
-   {
-       /* context forces OID choice; hasoid is now set correctly */
-   }
-   else
-   {
-       /* given free choice, don't leave space for OIDs in result tuples */
-       hasoid = false;
-   }
+   TupleDesc   tupDesc = ExecTypeFromTL(planstate->plan->targetlist);

-   tupDesc = ExecTypeFromTL(planstate->plan->targetlist, hasoid);
+   tupDesc = ExecTypeFromTL(planstate->plan->targetlist);
    planstate->ps_ResultTupleDesc = tupDesc;
 }

--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

#11

Andres Freund

andres@anarazel.de

about 7 years ago

In reply to: Peter Eisentraut (#10)

Re: pgsql: Remove WITH OIDS support, change oid catalog column visibility.

On 2019-02-16 11:14:21 +0100, Peter Eisentraut wrote:

On 2018-11-21 01:07, Andres Freund wrote:

Remove WITH OIDS support, change oid catalog column visibility.

I think you may have accidentally duplicated a line in this patch:

@@ -1602,20 +1602,9 @@ ExecFetchSlotHeapTupleDatum(TupleTableSlot *slot)
void
ExecInitResultTypeTL(PlanState *planstate)
{
-   bool        hasoid;
-   TupleDesc   tupDesc;
-
-   if (ExecContextForcesOids(planstate, &hasoid))
-   {
-       /* context forces OID choice; hasoid is now set correctly */
-   }
-   else
-   {
-       /* given free choice, don't leave space for OIDs in result tuples */
-       hasoid = false;
-   }
+   TupleDesc   tupDesc = ExecTypeFromTL(planstate->plan->targetlist);

-   tupDesc = ExecTypeFromTL(planstate->plan->targetlist, hasoid);
+   tupDesc = ExecTypeFromTL(planstate->plan->targetlist);
planstate->ps_ResultTupleDesc = tupDesc;
}

Indeed! Thanks for noticing. Fixed.

pgsql: Remove WITH OIDS support, change oid catalog column visibility.

Attachments: