Re: BUG #4434: Error inserting into view - unrecognized node type: 313

Tom Lane wrote:
> "Dean Rasheed" <dean_rasheed(at)hotmail(dot)com> writes:
>> CREATE TABLE foo(a int, b int);
>> CREATE VIEW foo_v AS SELECT * FROM foo;
>> CREATE RULE foo_r AS ON INSERT TO foo_v DO INSTEAD INSERT INTO foo
>> VALUES(NEW.a, NEW.b);
>> INSERT INTO foo_v VALUES ((SELECT 1), (SELECT 2)), ((SELECT 3), (SELECT 4));
>
>> ERROR: unrecognized node type: 313
>
> It looks like the parser's code path for multi-row VALUES is neglecting
> to detect sublinks and set pstate->p_hasSubLinks. I'm too tired to look
> closer tonight; anyone want to poke into it?

I think the parser is OK, but the p_hasSubLinks is lost in the rewrite
phase. In ResolveNew, we set p_hasSubLinks whenever a Var with a SubLink
is found. In case of Values RTE, however, there's no Vars, but plain
SubLink nodes. This patch seems to fix it:

--- src/backend/rewrite/rewriteManip.c
+++ src/backend/rewrite/rewriteManip.c
@@ -1112,6 +1112,12 @@ ResolveNew_mutator(Node *node, ResolveNew_context
*context)
context->sublevels_up--;
return (Node *) newnode;
}
+ else if (IsA(node, SubLink))
+ {
+ /* Report it if we are adding a sublink to query */
+ context->inserted_sublink = true;
+ /* fall through to copy the expr normally */
+ }
return expression_tree_mutator(node, ResolveNew_mutator,
(void *) context);
}

--
Heikki Linnakangas
EnterpriseDB http://www.enterprisedb.com