sql: pg_sequences implemented on pg_catalog

[mnovelodou]

fixes #65522

Previously, pg_sequences was added for compatiblity but not implemented
This was inadequate because these tables return values in postgres
To address this, this patch implements pg_sequences

Release note (sql change): pg_sequences table was implemented on pg_catalog

[blathers-crl]

Thank you for contributing to CockroachDB. Please ensure you have followed the guidelines for creating a PR.

My owl senses detect your PR is good for review. Please keep an eye out for any test failures in CI.

I was unable to automatically find a reviewer. You can try CCing one of the following members:

• A person you worked with closely on this PR.
• The person who created the ticket, or a CRDB organization member involved with the ticket (author, commenter, etc.).
• Join our community slack channel and ask on #contributors.
• Try find someone else from here.

_{🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is otan.}

[cockroach-teamcity]

This change is 

[blathers-crl]

Thank you for updating your pull request.

My owl senses detect your PR is good for review. Please keep an eye out for any test failures in CI.

_{🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is otan.}

[arulajmani]

Reviewable status: complete! 0 of 0 LGTMs obtained (waiting on @mnovelodou and @rafiss)

---

pkg/sql/pg_catalog.go, line 2904 at r1 (raw file):

				if sequenceValue != opts.Start-opts.Increment {
					// Before using for the first time, sequenceValue will be:
					// opts.Start - opts.Increment.

nit: comment should be above the if block

---

pkg/sql/pg_catalog.go, line 2911 at r1 (raw file):

					tree.NewDString(scName),                 // schemaname
					tree.NewDString(table.GetName()),        // sequencename
					getOwnerName(table),                     // sequenceowner

Should this be the sequence owner which is on the sequence opts? Could you check against postgres?

Also, if I'm correct, it would be worth adding a test which sets a sequence owner.

---

pkg/sql/logictest/testdata/logic_test/pg_catalog, line 3309 at r1 (raw file):

SELECT * FROM pg_seclabel

statement ok

Could you add a test that sets the CACHE option as well?

[mnovelodou]

Reviewable status: complete! 0 of 0 LGTMs obtained (waiting on @arulajmani and @rafiss)

---

pkg/sql/pg_catalog.go, line 2904 at r1 (raw file):

Previously, arulajmani (Arul Ajmani) wrote…

nit: comment should be above the if block

Done.

---

pkg/sql/pg_catalog.go, line 2911 at r1 (raw file):

Previously, arulajmani (Arul Ajmani) wrote…

Should this be the sequence owner which is on the sequence opts? Could you check against postgres?

Also, if I'm correct, it would be worth adding a test which sets a sequence owner.

Nice catch!

I took a look at postgres and seems like there are 2 different things:
"OWNED BY" and "OWNER", You can ALTER SEQUENCE and change both values.
The first one (OWNED BY) refers to a table.column which in case the column or the entire table gets deleted, the sequence will get deleted as well. sequence opts owner refers to this table.column:

type TableDescriptor_SequenceOpts_SequenceOwner struct {
// Sequence Owner's Column ID
OwnerColumnID ColumnID protobuf:"varint,1,opt,name=owner_column_id,json=ownerColumnId,casttype=ColumnID" json:"owner_column_id"
// Sequence Owner's Table ID
OwnerTableID ID protobuf:"varint,2,opt,name=owner_table_id,json=ownerTableId,casttype=ID" json:"owner_table_id"
}

Owner refers to the username that owns the sequence. Which in this case is correct, but I will add a test case that will verify this.

Thank you for raising this up!

---

pkg/sql/logictest/testdata/logic_test/pg_catalog, line 3309 at r1 (raw file):

Previously, arulajmani (Arul Ajmani) wrote…

Could you add a test that sets the CACHE option as well?

Done.

[arulajmani]

Reviewable status: complete! 1 of 0 LGTMs obtained (waiting on @arulajmani, @mnovelodou, and @rafiss)

---

pkg/sql/pg_catalog.go, line 2911 at r1 (raw file):

Previously, mnovelodou (Miguel Novelo) wrote…

Nice catch!

I took a look at postgres and seems like there are 2 different things:
"OWNED BY" and "OWNER", You can ALTER SEQUENCE and change both values.
The first one (OWNED BY) refers to a table.column which in case the column or the entire table gets deleted, the sequence will get deleted as well. sequence opts owner refers to this table.column:

type TableDescriptor_SequenceOpts_SequenceOwner struct {
// Sequence Owner's Column ID
OwnerColumnID ColumnID protobuf:"varint,1,opt,name=owner_column_id,json=ownerColumnId,casttype=ColumnID" json:"owner_column_id"
// Sequence Owner's Table ID
OwnerTableID ID protobuf:"varint,2,opt,name=owner_table_id,json=ownerTableId,casttype=ID" json:"owner_table_id"
}

Owner refers to the username that owns the sequence. Which in this case is correct, but I will add a test case that will verify this.

Thank you for raising this up!

Thanks for confirming!

[arulajmani]

bors r+

[craig]

Build succeeded:

• GitHub CI (Cockroach)