Robert Haas <robertmhaas@gmail.com> writes:
> On Thu, Oct 7, 2010 at 2:53 PM, Pavel Stehule <pavel.stehule@gmail.com> wrote:
>> b) SRF functions must not be finished by RETURN statement - I know, so
>> there is outer default block, but it looks like inconsistency for SRF
>> functions, because you can use a RETURN NEXT without RETURN. It maybe
>> isn't bug - but I am filling it as inconsistency.
> I don't see what's wrong with this.
Back around 8.0 we intentionally changed plpgsql to not require a final
RETURN in cases where RETURN isn't used to supply the result value:
http://archives.postgresql.org/pgsql-hackers/2005-04/msg00152.php
http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=e00ee887612da0dab02f1a56e33d8ae821710e14
Even if there were a good argument for going back to the old way,
backwards-compatibility would win the day, I think. Being strict
about this --- in *either* direction --- would break a lot of code.
regards, tom lane