JSON_TABLE: Add support for NESTED paths and columns

A NESTED path allows to extract data from nested levels of JSON
objects given by the parent path expression, which are projected as
columns specified using a nested COLUMNS clause, just like the parent
COLUMNS clause.  Rows comprised from a NESTED columns are "joined"
to the row comprised from the parent columns.  If a particular NESTED
path evaluates to 0 rows, then the nested COLUMNS will emit NULLs,
making it an OUTER join.

NESTED columns themselves may include NESTED paths to allow
extracting data from arbitrary nesting levels, which are likewise
joined against the rows at the parent level.

Multiple NESTED paths at a given level are called "sibling" paths
and their rows are combined by UNIONing them, that is, after being
joined against the parent row as described above.

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: Andrew Dunstan <andrew@dunslane.net>
Author: Amit Langote <amitlangote09@gmail.com>
Author: Jian He <jian.universality@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, Álvaro Herrera, Jian He

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
Discussion: https://postgr.es/m/CA+HiwqE4XTdfb1nW=Ojoy_tQSRhYt-q_kb6i5d4xcKyrLC1Nbg@mail.gmail.com

1 files changed, 56 insertions, 4 deletions

diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 466d9576a21..52bf87ac55b 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -524,8 +524,13 @@ static char *flatten_reloptions(Oid relid);
 static void get_reloptions(StringInfo buf, Datum reloptions);
 static void get_json_path_spec(Node *path_spec, deparse_context *context,
 							   bool showimplicit);
-static void get_json_table_columns(TableFunc *tf, deparse_context *context,
+static void get_json_table_columns(TableFunc *tf, JsonTablePathScan *scan,
+								   deparse_context *context,
 								   bool showimplicit);
+static void get_json_table_nested_columns(TableFunc *tf, JsonTablePlan *plan,
+										  deparse_context *context,
+										  bool showimplicit,
+										  bool needcomma);
 
 #define only_marker(rte)  ((rte)->inh ? "" : "ONLY ")
 
@@ -11627,10 +11632,43 @@ get_xmltable(TableFunc *tf, deparse_context *context, bool showimplicit)
 }
 
 /*
+ * get_json_nested_columns - Parse back nested JSON_TABLE columns
+ */
+static void
+get_json_table_nested_columns(TableFunc *tf, JsonTablePlan *plan,
+							  deparse_context *context, bool showimplicit,
+							  bool needcomma)
+{
+	if (IsA(plan, JsonTablePathScan))
+	{
+		JsonTablePathScan *scan = castNode(JsonTablePathScan, plan);
+
+		if (needcomma)
+			appendStringInfoChar(context->buf, ',');
+
+		appendStringInfoChar(context->buf, ' ');
+		appendContextKeyword(context, "NESTED PATH ", 0, 0, 0);
+		get_const_expr(scan->path->value, context, -1);
+		appendStringInfo(context->buf, " AS %s", quote_identifier(scan->path->name));
+		get_json_table_columns(tf, scan, context, showimplicit);
+	}
+	else if (IsA(plan, JsonTableSiblingJoin))
+	{
+		JsonTableSiblingJoin *join = (JsonTableSiblingJoin *) plan;
+
+		get_json_table_nested_columns(tf, join->lplan, context, showimplicit,
+									  needcomma);
+		get_json_table_nested_columns(tf, join->rplan, context, showimplicit,
+									  true);
+	}
+}
+
+/*
  * get_json_table_columns - Parse back JSON_TABLE columns
  */
 static void
-get_json_table_columns(TableFunc *tf, deparse_context *context,
+get_json_table_columns(TableFunc *tf, JsonTablePathScan *scan,
+					   deparse_context *context,
 					   bool showimplicit)
 {
 	StringInfo	buf = context->buf;
@@ -11663,7 +11701,16 @@ get_json_table_columns(TableFunc *tf, deparse_context *context,
 		typmod = lfirst_int(lc_coltypmod);
 		colexpr = castNode(JsonExpr, lfirst(lc_colvalexpr));
 
-		if (colnum > 0)
+		/* Skip columns that don't belong to this scan. */
+		if (scan->colMin < 0 || colnum < scan->colMin)
+		{
+			colnum++;
+			continue;
+		}
+		if (colnum > scan->colMax)
+			break;
+
+		if (colnum > scan->colMin)
 			appendStringInfoString(buf, ", ");
 
 		colnum++;
@@ -11711,6 +11758,10 @@ get_json_table_columns(TableFunc *tf, deparse_context *context,
 		get_json_expr_options(colexpr, context, default_behavior);
 	}
 
+	if (scan->child)
+		get_json_table_nested_columns(tf, scan->child, context, showimplicit,
+									  scan->colMin >= 0);
+
 	if (PRETTY_INDENT(context))
 		context->indentLevel -= PRETTYINDENT_VAR;
 
@@ -11774,7 +11825,8 @@ get_json_table(TableFunc *tf, deparse_context *context, bool showimplicit)
 			context->indentLevel -= PRETTYINDENT_VAR;
 	}
 
-	get_json_table_columns(tf, context, showimplicit);
+	get_json_table_columns(tf, castNode(JsonTablePathScan, tf->plan), context,
+						   showimplicit);
 
 	if (jexpr->on_error->btype != JSON_BEHAVIOR_EMPTY)
 		get_json_behavior(jexpr->on_error, context, "ERROR");