Re: Question about duplicate JSONTYPE_JSON check - Mailing list pgsql-hackers

On Wed, Mar 12, 2025 at 7:09 PM Álvaro Herrera <alvherre@alvh.no-ip.org> wrote:
> On 2025-Mar-12, Amit Langote wrote:
>
> > I was able to construct a test case that crashes due to this bug:
> >
> > CREATE TYPE mood AS ENUM ('happy', 'sad', 'neutral');
> > CREATE FUNCTION mood_to_json(mood) RETURNS json AS $$
> >   SELECT to_json($1::text);
> > $$ LANGUAGE sql IMMUTABLE;
> > CREATE CAST (mood AS json) WITH FUNCTION mood_to_json(mood) AS IMPLICIT;
> >
> > SELECT JSON_OBJECT('happy'::mood: '123'::jsonb);
> > server closed the connection unexpectedly
>
> Good reaction time :-)  I see that that line shows as not even uncovered
> in the report, but as non-existant (no background color as opposed to
> red):
>
> https://coverage.postgresql.org/src/backend/utils/adt/jsonb.c.gcov.html#660
>
> Evidently the compiler must be optimizing it out as dead code.

Ah, I did wonder about the coverage, thanks for pointing it out.

Patch look good for committing?

--
Thanks, Amit Langote