sql/parser: Move DatumToOid map to Oid() method on parser.Type

[nvanbenschoten]

This initial change removes a map from parser.Type to oid.Oid and
replaces it with a method on parser.Type. This allows us to avoid a
previous issue with reflection, and allows us to simplify some code that
was performing the same mapping in pgwire.

This is the first commit of #12641, but is a nice cleanup in isolation as well.

---

This change is 

[knz]

modulo the two comments

---

Reviewed 6 of 6 files at r1.
Review status: all files reviewed at latest revision, 2 unresolved discussions, some commit checks failed.

---

pkg/sql/parser/type.go, line 410 at r1 (raw file):

	case TypeString:
		return oid.T__text
	case TypeAny:

See my comment about this in #12641 (comment)

---

pkg/sql/parser/type.go, line 449 at r1 (raw file):

// Oid implements the Type interface.
func (TTable) Oid() oid.Oid {
	panic("TODO(knz) unknown Oid for TTable")

T_anyelement See my comment about this in #12641 (comment)

---

Comments from Reviewable

[jordanlewis]

Nice!

---

Review status: all files reviewed at latest revision, 3 unresolved discussions, some commit checks failed.

---

pkg/sql/parser/type.go, line 126 at r1 (raw file):

	oid.T_text:        TypeString,
	oid.T__text:       TypeStringArray,
	oid.T__int2:       TypeIntArray,

Not sure if it matters but you're missing an entry for oid.T_unknown here. I see that it's not a unique mapping - perhaps that's an issue?

Also missing an entry for T_record.

---

Comments from Reviewable

[nvanbenschoten]

TFTRs!

---

Review status: all files reviewed at latest revision, 3 unresolved discussions, some commit checks failed.

---

pkg/sql/parser/type.go, line 126 at r1 (raw file):

Previously, jordanlewis (Jordan Lewis) wrote…

Not sure if it matters but you're missing an entry for oid.T_unknown here. I see that it's not a unique mapping - perhaps that's an issue?

Also missing an entry for T_record.

Yeah I'm not quite sure that it makes sense to add an entry for oid.T_unknown, based on how we use this mapping. For now, I'll leave it out and wait to see if it's ever needed.

Good catch on the T_record. Done.

---

pkg/sql/parser/type.go, line 410 at r1 (raw file):

Previously, knz (kena) wrote…

See my comment about this in #12641 (comment)

Done.

---

pkg/sql/parser/type.go, line 449 at r1 (raw file):

Previously, knz (kena) wrote…

T_anyelement See my comment about this in #12641 (comment)

Done.

---

Comments from Reviewable

[knz]

Reviewed 4 of 4 files at r2.
Review status: all files reviewed at latest revision, 1 unresolved discussion, all commit checks successful.

---

Comments from Reviewable