Hi Martin,
I'm assuming this generated an error for you on compile? Which os/compiler
did you use? On our various compiles this does not show up.
As for the issue, in PostgreSQL the *ereport *function causes an exit, so
execution will never reach the return that follows. Some compilers complain
if there isn't a final return in a function, so that is why there is one.
Although, it will never reach it.
I will correct it with a patch, though, so that it won't cause any other
issues.
John
On Wed, Jan 27, 2021 at 1:17 AM Martin Strunz <notificati...@github.com>
wrote:
> /src/backend/utils/adt/agtype_util.c line 1536 Non-boolean value returned
> from function returning bool
> static bool equals_agtype_scalar_value(agtype_value *a, agtype_value *b)
> {
> if (a->type == b->type)
> {
> switch (a->type)
> {
> case AGTV_NULL:
> return true;
> case AGTV_STRING:
> return length_compare_agtype_string_value(a, b) == 0;
> case AGTV_NUMERIC:
> return DatumGetBool(DirectFunctionCall2(
> numeric_eq, PointerGetDatum(a->val.numeric),
> PointerGetDatum(b->val.numeric)));
> case AGTV_BOOL:
> return a->val.boolean == b->val.boolean;
> case AGTV_INTEGER:
> return a->val.int_value == b->val.int_value;
> case AGTV_FLOAT:
> return a->val.float_value == b->val.float_value;
> case AGTV_VERTEX:
> {
> graphid a_graphid, b_graphid;
> a_graphid = a->val.object.pairs[0].value.val.int_value;
> b_graphid = b->val.object.pairs[0].value.val.int_value;
>
> return a_graphid == b_graphid;
> }
>
> default:
> ereport(ERROR, (errmsg("invalid agtype scalar type %d for equals",
> a->type)));
> }
> }
> ereport(ERROR, (errmsg("agtype input scalars must be of same type")));
> return -1; // <- should be bool
>
> }
>
> —
> You are receiving this because you are subscribed to this thread.
> Reply to this email directly, view it on GitHub
> <https://github.com/apache/incubator-age/issues/15>, or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/AALPO4EERTXHP5TUFZQGYS3S37KZ5ANCNFSM4WU47RPA>
> .
>
