On Wed, Sep 10, 2025 at 9:31 PM Tender Wang <tndrwang@gmail.com> wrote:
>> PG Bug reporting form <noreply@postgresql.org> 于2025年9月10日周三 18:22写道:
>>> SELECT sub.c FROM
>>> (SELECT json_array(3, 2, t.c) AS c FROM t) AS sub
>>> RIGHT JOIN t ON FALSE;
> diff --git a/src/backend/optimizer/util/clauses.c b/src/backend/optimizer/util/clauses.c
> index 6f0b338d2cd..5ef364b7f7c 100644
> --- a/src/backend/optimizer/util/clauses.c
> +++ b/src/backend/optimizer/util/clauses.c
> @@ -1115,6 +1115,8 @@ contain_nonstrict_functions_walker(Node *node, void *context)
> return true;
> if (IsA(node, BooleanTest))
> return true;
> + if (IsA(node, JsonConstructorExpr))
> + return true;
>
> I added the above codes, then the query returned the correct result.
> I didn't dig more the details. Any thought?
Yeah, JsonConstructorExpr should not be treated as a non-strict
construct. This fix looks correct to me.
I'm wondering if this is the only case we've overlooked. How about
other Json-related expressions?
- Richard