Re: optionally schema-qualified for table_name
On Wed, Mar 25, 2020 at 01:46:54PM +0100, Peter Eisentraut wrote: > On 2020-03-24 21:58, Bruce Momjian wrote: > > That is a good point. I used CREATE VIEW as an example because that is > > what the user reported, but it seems only create_view and reindexed use > > a schema name qualification: > > > > $ grep -l 'schema' *.sgml > > create_view.sgml > > reindexdb.sgml > > > > The reindexdb use is because of -S (reindex schema), which makes sense. > > The create view case is used in an example of CREATE RECURSIVE VIEW and > > should probably be removed. > > The CREATE RECURSIVE VIEW example is making a specific point about schema > qualification, which is explained below the example. OK, so I guess everything is fine and we can just go back to other business. :-) Sorry for the distraction. -- Bruce Momjian https://momjian.us EnterpriseDB https://enterprisedb.com + As you are, so once was I. As I am, so you will be. + + Ancient Roman grave inscription +
Re: optionally schema-qualified for table_name
On 2020-03-24 21:58, Bruce Momjian wrote: That is a good point. I used CREATE VIEW as an example because that is what the user reported, but it seems only create_view and reindexed use a schema name qualification: $ grep -l 'schema' *.sgml create_view.sgml reindexdb.sgml The reindexdb use is because of -S (reindex schema), which makes sense. The create view case is used in an example of CREATE RECURSIVE VIEW and should probably be removed. The CREATE RECURSIVE VIEW example is making a specific point about schema qualification, which is explained below the example. -- Peter Eisentraut http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
Re: optionally schema-qualified for table_name
On Tue, Mar 24, 2020 at 09:35:25PM +0100, Peter Eisentraut wrote: > On 2020-03-23 02:27, Bruce Momjian wrote: > > On Sun, Mar 22, 2020 at 06:20:04PM -0400, Tom Lane wrote: > > > Bruce Momjian writes: > > > > On Sun, Mar 22, 2020 at 03:05:01PM -0400, Tom Lane wrote: > > > > > I don't really think this is an improvement, mainly because that > > > > > error message is inventing a notation that we do not use in any > > > > > other error message. > > > > > > > What do you suggest? The current message is: > > > > > > > Specify OWNED BY table.column or OWNED BY NONE. > > > > > > Yeah, and I think that's okay as-is, or at least we can't make it better > > > without fairly whole-sale changes of our documentation practices. > > > The fact that a table name can be schema-qualified is usually implicit, > > > and I don't see why this place cries out for making it explicit > > > more than other places. You could as well complain that there's > > > nothing explicit here about double-quoting practices. > > > > OK, I will do just the documentation patch for this then. > > The same criticism applies to the documentation patch, I think. We don't > usually make the schema part explicit. That is a good point. I used CREATE VIEW as an example because that is what the user reported, but it seems only create_view and reindexed use a schema name qualification: $ grep -l 'schema' *.sgml create_view.sgml reindexdb.sgml The reindexdb use is because of -S (reindex schema), which makes sense. The create view case is used in an example of CREATE RECURSIVE VIEW and should probably be removed. -- Bruce Momjian https://momjian.us EnterpriseDB https://enterprisedb.com + As you are, so once was I. As I am, so you will be. + + Ancient Roman grave inscription +
Re: optionally schema-qualified for table_name
On 2020-03-23 02:27, Bruce Momjian wrote: On Sun, Mar 22, 2020 at 06:20:04PM -0400, Tom Lane wrote: Bruce Momjian writes: On Sun, Mar 22, 2020 at 03:05:01PM -0400, Tom Lane wrote: I don't really think this is an improvement, mainly because that error message is inventing a notation that we do not use in any other error message. What do you suggest? The current message is: Specify OWNED BY table.column or OWNED BY NONE. Yeah, and I think that's okay as-is, or at least we can't make it better without fairly whole-sale changes of our documentation practices. The fact that a table name can be schema-qualified is usually implicit, and I don't see why this place cries out for making it explicit more than other places. You could as well complain that there's nothing explicit here about double-quoting practices. OK, I will do just the documentation patch for this then. The same criticism applies to the documentation patch, I think. We don't usually make the schema part explicit. -- Peter Eisentraut http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
Re: optionally schema-qualified for table_name
On Sun, Mar 22, 2020 at 06:20:04PM -0400, Tom Lane wrote: > Bruce Momjian writes: > > On Sun, Mar 22, 2020 at 03:05:01PM -0400, Tom Lane wrote: > >> I don't really think this is an improvement, mainly because that > >> error message is inventing a notation that we do not use in any > >> other error message. > > > What do you suggest? The current message is: > > > Specify OWNED BY table.column or OWNED BY NONE. > > Yeah, and I think that's okay as-is, or at least we can't make it better > without fairly whole-sale changes of our documentation practices. > The fact that a table name can be schema-qualified is usually implicit, > and I don't see why this place cries out for making it explicit > more than other places. You could as well complain that there's > nothing explicit here about double-quoting practices. OK, I will do just the documentation patch for this then. -- Bruce Momjian https://momjian.us EnterpriseDB https://enterprisedb.com + As you are, so once was I. As I am, so you will be. + + Ancient Roman grave inscription +
Re: optionally schema-qualified for table_name
Bruce Momjian writes: > On Sun, Mar 22, 2020 at 03:05:01PM -0400, Tom Lane wrote: >> I don't really think this is an improvement, mainly because that >> error message is inventing a notation that we do not use in any >> other error message. > What do you suggest? The current message is: > Specify OWNED BY table.column or OWNED BY NONE. Yeah, and I think that's okay as-is, or at least we can't make it better without fairly whole-sale changes of our documentation practices. The fact that a table name can be schema-qualified is usually implicit, and I don't see why this place cries out for making it explicit more than other places. You could as well complain that there's nothing explicit here about double-quoting practices. regards, tom lane
Re: optionally schema-qualified for table_name
On Sun, Mar 22, 2020 at 03:05:01PM -0400, Tom Lane wrote: > Bruce Momjian writes: > > I see what you mean. The attached patch fixes this, as well as > > adjusting the error message. I didn't see any other cases. > > I don't really think this is an improvement, mainly because that > error message is inventing a notation that we do not use in any > other error message. What do you suggest? The current message is: Specify OWNED BY table.column or OWNED BY NONE. I don't see any other messages with "table.column". Do you want? Specify OWNED BY column or OWNED BY NONE. -- Bruce Momjian https://momjian.us EnterpriseDB https://enterprisedb.com + As you are, so once was I. As I am, so you will be. + + Ancient Roman grave inscription +
Re: optionally schema-qualified for table_name
Bruce Momjian writes: > I see what you mean. The attached patch fixes this, as well as > adjusting the error message. I didn't see any other cases. I don't really think this is an improvement, mainly because that error message is inventing a notation that we do not use in any other error message. regards, tom lane
Re: optionally schema-qualified for table_name
On Thu, Mar 12, 2020 at 05:58:02AM +, PG Doc comments form wrote:
> The following documentation comment has been logged on the website:
>
> Page: https://www.postgresql.org/docs/12/sql-altersequence.html
> Description:
>
> Although I can see that table_name in OWNED BY clause can be optionally
> schema-qualified by á…źcarefully reading "The specified table must have the
> same owner and be in the same schema as the sequence.", it would be good if
> "optionally schema-qualified" is explicitly noted somehow like other pages
> such as CREATE TABLE and CREATE VIEW. The same applies to CREATE SEQUENCE
> page.
I see what you mean. The attached patch fixes this, as well as
adjusting the error message. I didn't see any other cases.
I thought maybe the schema wasn't mentioned because the table.column
defaults to the sequence's schema, but it does not --- you have to
specify the column's schema if would not be normally be found via
search_path:
CREATE SCHEMA zz;
SET search_path = zz, public;
CREATE TABLE zz.test (x INT);
CREATE SEQUENCE zz.ss;
ALTER SEQUENCE zz.ss OWNED BY test.x;
SET search_path = public, zz;
ALTER SEQUENCE zz.ss OWNED BY test.x;
SET search_path = public;
ALTER SEQUENCE zz.ss OWNED BY test.x;
--> ERROR: relation "test" does not exist
ALTER SEQUENCE zz.ss OWNED BY zz.test.x;
--
Bruce Momjian https://momjian.us
EnterpriseDB https://enterprisedb.com
+ As you are, so once was I. As I am, so you will be. +
+ Ancient Roman grave inscription +
diff --git a/doc/src/sgml/ref/alter_sequence.sgml b/doc/src/sgml/ref/alter_sequence.sgml
index bfd20af6d3..b732a68d4e 100644
--- a/doc/src/sgml/ref/alter_sequence.sgml
+++ b/doc/src/sgml/ref/alter_sequence.sgml
@@ -30,7 +30,7 @@ ALTER SEQUENCE [ IF EXISTS ] name
[ START [ WITH ] start ]
[ RESTART [ [ WITH ] restart ] ]
[ CACHE cache ] [ [ NO ] CYCLE ]
-[ OWNED BY { table_name.column_name | NONE } ]
+[ OWNED BY { [ schema . ] table_name.column_name | NONE } ]
ALTER SEQUENCE [ IF EXISTS ] name OWNER TO { new_owner | CURRENT_USER | SESSION_USER }
ALTER SEQUENCE [ IF EXISTS ] name RENAME TO new_name
ALTER SEQUENCE [ IF EXISTS ] name SET SCHEMA new_schema
@@ -238,7 +238,7 @@ ALTER SEQUENCE [ IF EXISTS ] name S
-OWNED BY table_name.column_name
+OWNED BY [ schema . ] table_name.column_name
OWNED BY NONE
diff --git a/doc/src/sgml/ref/create_sequence.sgml b/doc/src/sgml/ref/create_sequence.sgml
index e4085804a4..cd72293df9 100644
--- a/doc/src/sgml/ref/create_sequence.sgml
+++ b/doc/src/sgml/ref/create_sequence.sgml
@@ -26,7 +26,7 @@ CREATE [ TEMPORARY | TEMP ] SEQUENCE [ IF NOT EXISTS ] increment ]
[ MINVALUE minvalue | NO MINVALUE ] [ MAXVALUE maxvalue | NO MAXVALUE ]
[ START [ WITH ] start ] [ CACHE cache ] [ [ NO ] CYCLE ]
-[ OWNED BY { table_name.column_name | NONE } ]
+[ OWNED BY { [ schema . ] table_name.column_name | NONE } ]
@@ -225,7 +225,7 @@ SELECT * FROM name;
-OWNED BY table_name.column_name
+OWNED BY [ schema . ] table_name.column_name
OWNED BY NONE
diff --git a/src/backend/commands/sequence.c b/src/backend/commands/sequence.c
index 6aab73bfd4..f9750d2cdf 100644
--- a/src/backend/commands/sequence.c
+++ b/src/backend/commands/sequence.c
@@ -1656,7 +1656,7 @@ process_owned_by(Relation seqrel, List *owned_by, bool for_identity)
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("invalid OWNED BY option"),
- errhint("Specify OWNED BY table.column or OWNED BY NONE.")));
+ errhint("Specify OWNED BY [schema.]table.column or OWNED BY NONE.")));
tablerel = NULL;
attnum = 0;
}
diff --git a/src/test/regress/expected/sequence.out b/src/test/regress/expected/sequence.out
index 8b928b2888..e66059fd75 100644
--- a/src/test/regress/expected/sequence.out
+++ b/src/test/regress/expected/sequence.out
@@ -19,7 +19,7 @@ ERROR: CACHE (0) must be greater than zero
-- OWNED BY errors
CREATE SEQUENCE sequence_testx OWNED BY nobody; -- nonsense word
ERROR: invalid OWNED BY option
-HINT: Specify OWNED BY table.column or OWNED BY NONE.
+HINT: Specify OWNED BY [schema.]table.column or OWNED BY NONE.
CREATE SEQUENCE sequence_testx OWNED BY pg_class_oid_index.oid; -- not a table
ERROR: referenced relation "pg_class_oid_index" is not a table or foreign table
CREATE SEQUENCE sequence_testx OWNED BY pg_class.relname; -- not same schema
