SQL/JSON: add standard JSON constructor functions

This commit introduces the SQL/JSON standard-conforming constructors for
JSON types:

JSON_ARRAY()
JSON_ARRAYAGG()
JSON_OBJECT()
JSON_OBJECTAGG()

Most of the functionality was already present in PostgreSQL-specific
functions, but these include some new functionality such as the ability
to skip or include NULL values, and to allow duplicate keys or throw
error when they are found, as well as the standard specified syntax to
specify output type and format.

Author: Nikita Glukhov <n.gluhov@postgrespro.ru>
Author: Teodor Sigaev <teodor@sigaev.ru>
Author: Oleg Bartunov <obartunov@gmail.com>
Author: Alexander Korotkov <aekorotkov@gmail.com>
Author: Amit Langote <amitlangote09@gmail.com>

Reviewers have included (in no particular order) Andres Freund, Alexander
Korotkov, Pavel Stehule, Andrew Alsup, Erik Rijkers, Zihong Yu,
Himanshu Upadhyaya, Daniel Gustafsson, Justin Pryzby.

Discussion: https://postgr.es/m/CAF4Au4w2x-5LTnN_bxky-mq4=WOqsGsxSpENCzHRAzSnEd8+WQ@mail.gmail.com
Discussion: https://postgr.es/m/cd0bb935-0158-78a7-08b5-904886deac4b@postgrespro.ru
Discussion: https://postgr.es/m/20220616233130.rparivafipt6doj3@alap3.anarazel.de
Discussion: https://postgr.es/m/abd9b83b-aa66-f230-3d6d-734817f0995d%40postgresql.org

1 files changed, 221 insertions, 0 deletions

diff --git a/src/backend/nodes/nodeFuncs.c b/src/backend/nodes/nodeFuncs.c
index dc8415a693..69907fbcde 100644
--- a/src/backend/nodes/nodeFuncs.c
+++ b/src/backend/nodes/nodeFuncs.c
@@ -249,6 +249,18 @@ exprType(const Node *expr)
 		case T_PlaceHolderVar:
 			type = exprType((Node *) ((const PlaceHolderVar *) expr)->phexpr);
 			break;
+		case T_JsonValueExpr:
+			{
+				const JsonValueExpr *jve = (const JsonValueExpr *) expr;
+
+				type = exprType((Node *)
+								(jve->formatted_expr ? jve->formatted_expr :
+								 jve->raw_expr));
+			}
+			break;
+		case T_JsonConstructorExpr:
+			type = ((const JsonConstructorExpr *) expr)->returning->typid;
+			break;
 		default:
 			elog(ERROR, "unrecognized node type: %d", (int) nodeTag(expr));
 			type = InvalidOid;	/* keep compiler quiet */
@@ -479,6 +491,10 @@ exprTypmod(const Node *expr)
 			return ((const SetToDefault *) expr)->typeMod;
 		case T_PlaceHolderVar:
 			return exprTypmod((Node *) ((const PlaceHolderVar *) expr)->phexpr);
+		case T_JsonValueExpr:
+			return exprTypmod((Node *) ((const JsonValueExpr *) expr)->formatted_expr);
+		case T_JsonConstructorExpr:
+			return -1;			/* XXX maybe expr->returning->typmod? */
 		default:
 			break;
 	}
@@ -954,6 +970,19 @@ exprCollation(const Node *expr)
 		case T_PlaceHolderVar:
 			coll = exprCollation((Node *) ((const PlaceHolderVar *) expr)->phexpr);
 			break;
+		case T_JsonValueExpr:
+			coll = exprCollation((Node *) ((const JsonValueExpr *) expr)->formatted_expr);
+			break;
+		case T_JsonConstructorExpr:
+			{
+				const JsonConstructorExpr *ctor = (const JsonConstructorExpr *) expr;
+
+				if (ctor->coercion)
+					coll = exprCollation((Node *) ctor->coercion);
+				else
+					coll = InvalidOid;
+			}
+			break;
 		default:
 			elog(ERROR, "unrecognized node type: %d", (int) nodeTag(expr));
 			coll = InvalidOid;	/* keep compiler quiet */
@@ -1161,6 +1190,21 @@ exprSetCollation(Node *expr, Oid collation)
 			/* NextValueExpr's result is an integer type ... */
 			Assert(!OidIsValid(collation)); /* ... so never set a collation */
 			break;
+		case T_JsonValueExpr:
+			exprSetCollation((Node *) ((JsonValueExpr *) expr)->formatted_expr,
+							 collation);
+			break;
+		case T_JsonConstructorExpr:
+			{
+				JsonConstructorExpr *ctor = (JsonConstructorExpr *) expr;
+
+				if (ctor->coercion)
+					exprSetCollation((Node *) ctor->coercion, collation);
+				else
+					Assert(!OidIsValid(collation)); /* result is always a
+													 * json[b] type */
+			}
+			break;
 		default:
 			elog(ERROR, "unrecognized node type: %d", (int) nodeTag(expr));
 			break;
@@ -1603,6 +1647,12 @@ exprLocation(const Node *expr)
 		case T_PartitionRangeDatum:
 			loc = ((const PartitionRangeDatum *) expr)->location;
 			break;
+		case T_JsonValueExpr:
+			loc = exprLocation((Node *) ((const JsonValueExpr *) expr)->raw_expr);
+			break;
+		case T_JsonConstructorExpr:
+			loc = ((const JsonConstructorExpr *) expr)->location;
+			break;
 		default:
 			/* for any other node type it's just unknown... */
 			loc = -1;
@@ -2334,6 +2384,28 @@ expression_tree_walker_impl(Node *node,
 					return true;
 			}
 			break;
+		case T_JsonValueExpr:
+			{
+				JsonValueExpr *jve = (JsonValueExpr *) node;
+
+				if (WALK(jve->raw_expr))
+					return true;
+				if (WALK(jve->formatted_expr))
+					return true;
+			}
+			break;
+		case T_JsonConstructorExpr:
+			{
+				JsonConstructorExpr *ctor = (JsonConstructorExpr *) node;
+
+				if (WALK(ctor->args))
+					return true;
+				if (WALK(ctor->func))
+					return true;
+				if (WALK(ctor->coercion))
+					return true;
+			}
+			break;
 		default:
 			elog(ERROR, "unrecognized node type: %d",
 				 (int) nodeTag(node));
@@ -2664,6 +2736,7 @@ expression_tree_mutator_impl(Node *node,
 		case T_RangeTblRef:
 		case T_SortGroupClause:
 		case T_CTESearchClause:
+		case T_JsonFormat:
 			return (Node *) copyObject(node);
 		case T_WithCheckOption:
 			{
@@ -3307,6 +3380,41 @@ expression_tree_mutator_impl(Node *node,
 				return (Node *) newnode;
 			}
 			break;
+		case T_JsonReturning:
+			{
+				JsonReturning *jr = (JsonReturning *) node;
+				JsonReturning *newnode;
+
+				FLATCOPY(newnode, jr, JsonReturning);
+				MUTATE(newnode->format, jr->format, JsonFormat *);
+
+				return (Node *) newnode;
+			}
+		case T_JsonValueExpr:
+			{
+				JsonValueExpr *jve = (JsonValueExpr *) node;
+				JsonValueExpr *newnode;
+
+				FLATCOPY(newnode, jve, JsonValueExpr);
+				MUTATE(newnode->raw_expr, jve->raw_expr, Expr *);
+				MUTATE(newnode->formatted_expr, jve->formatted_expr, Expr *);
+				MUTATE(newnode->format, jve->format, JsonFormat *);
+
+				return (Node *) newnode;
+			}
+		case T_JsonConstructorExpr:
+			{
+				JsonConstructorExpr *jve = (JsonConstructorExpr *) node;
+				JsonConstructorExpr *newnode;
+
+				FLATCOPY(newnode, jve, JsonConstructorExpr);
+				MUTATE(newnode->args, jve->args, List *);
+				MUTATE(newnode->func, jve->func, Expr *);
+				MUTATE(newnode->coercion, jve->coercion, Expr *);
+				MUTATE(newnode->returning, jve->returning, JsonReturning *);
+
+				return (Node *) newnode;
+			}
 		default:
 			elog(ERROR, "unrecognized node type: %d",
 				 (int) nodeTag(node));
@@ -3578,6 +3686,7 @@ raw_expression_tree_walker_impl(Node *node,
 		case T_ParamRef:
 		case T_A_Const:
 		case T_A_Star:
+		case T_JsonFormat:
 			/* primitive node types with no subnodes */
 			break;
 		case T_Alias:
@@ -4040,6 +4149,118 @@ raw_expression_tree_walker_impl(Node *node,
 		case T_CommonTableExpr:
 			/* search_clause and cycle_clause are not interesting here */
 			return WALK(((CommonTableExpr *) node)->ctequery);
+		case T_JsonReturning:
+			return WALK(((JsonReturning *) node)->format);
+		case T_JsonValueExpr:
+			{
+				JsonValueExpr *jve = (JsonValueExpr *) node;
+
+				if (WALK(jve->raw_expr))
+					return true;
+				if (WALK(jve->formatted_expr))
+					return true;
+				if (WALK(jve->format))
+					return true;
+			}
+			break;
+		case T_JsonConstructorExpr:
+			{
+				JsonConstructorExpr *ctor = (JsonConstructorExpr *) node;
+
+				if (WALK(ctor->args))
+					return true;
+				if (WALK(ctor->func))
+					return true;
+				if (WALK(ctor->coercion))
+					return true;
+				if (WALK(ctor->returning))
+					return true;
+			}
+			break;
+		case T_JsonOutput:
+			{
+				JsonOutput *out = (JsonOutput *) node;
+
+				if (WALK(out->typeName))
+					return true;
+				if (WALK(out->returning))
+					return true;
+			}
+			break;
+		case T_JsonKeyValue:
+			{
+				JsonKeyValue *jkv = (JsonKeyValue *) node;
+
+				if (WALK(jkv->key))
+					return true;
+				if (WALK(jkv->value))
+					return true;
+			}
+			break;
+		case T_JsonObjectConstructor:
+			{
+				JsonObjectConstructor *joc = (JsonObjectConstructor *) node;
+
+				if (WALK(joc->output))
+					return true;
+				if (WALK(joc->exprs))
+					return true;
+			}
+			break;
+		case T_JsonArrayConstructor:
+			{
+				JsonArrayConstructor *jac = (JsonArrayConstructor *) node;
+
+				if (WALK(jac->output))
+					return true;
+				if (WALK(jac->exprs))
+					return true;
+			}
+			break;
+		case T_JsonAggConstructor:
+			{
+				JsonAggConstructor *ctor = (JsonAggConstructor *) node;
+
+				if (WALK(ctor->output))
+					return true;
+				if (WALK(ctor->agg_order))
+					return true;
+				if (WALK(ctor->agg_filter))
+					return true;
+				if (WALK(ctor->over))
+					return true;
+			}
+			break;
+		case T_JsonObjectAgg:
+			{
+				JsonObjectAgg *joa = (JsonObjectAgg *) node;
+
+				if (WALK(joa->constructor))
+					return true;
+				if (WALK(joa->arg))
+					return true;
+			}
+			break;
+		case T_JsonArrayAgg:
+			{
+				JsonArrayAgg *jaa = (JsonArrayAgg *) node;
+
+				if (WALK(jaa->constructor))
+					return true;
+				if (WALK(jaa->arg))
+					return true;
+			}
+			break;
+		case T_JsonArrayQueryConstructor:
+			{
+				JsonArrayQueryConstructor *jaqc = (JsonArrayQueryConstructor *) node;
+
+				if (WALK(jaqc->output))
+					return true;
+				if (WALK(jaqc->query))
+					return true;
+			}
+			break;
 		default:
 			elog(ERROR, "unrecognized node type: %d",
 				 (int) nodeTag(node));