On Wed, Jun 2, 2021 2:44 PM Dilip Kumar <dilipbal...@gmail.com> wrote:
> Attached patch fixes that, I haven't yet added the test case. Once
> someone confirms on the approach then I will add a test case to the
> patch.
key_tuple = heap_form_tuple(desc, values, nulls);
*copy = true;
...
key_tuple = toast_flatten_tuple(oldtup, desc);
heap_freetuple(oldtup);
}
+ /*
+ * If key tuple doesn't have any external data and key is not changed
then
+ * just free the key tuple and return NULL.
+ */
+ else if (!key_changed)
+ {
+ heap_freetuple(key_tuple);
+ return NULL;
+ }
return key_tuple;
}
I think "*copy = false" should be added before return NULL because we don't
return a modified copy tuple here. Thoughts?
Regards
Tang
