Re: [HACKERS] Should we still require RETURN in plpgsql?
Hello... On Tue, 05 Apr 2005 02:28:23 -0400, [EMAIL PROTECTED] (Tom Lane) wrote: How does Oracle's PL/SQL handle this? On ORACLE a FUNCTION MUST return a value. If the FUNCTION doesn't return a value Oracle give a 'hint' on FUNCTION compilation and error on SELECT function invocation: ORA-06503. When we don't want to return any result on ORACLE we must use PROCEDURE statement instead of FUNCTION. Example: SQL CREATE OR REPLACE FUNCTION F_test RETURN NUMBER IS 2 BEGIN 3NULL; 4 END F_TEST; 5 / Function created. SQL SELECT F_TEST FROM DUAL; SELECT TUZSA.F_TEST FROM DUAL * ERROR at line 1: ORA-06503: PL/SQL: Function returned without value ORA-06512: at F_TEST, line 3 SQL Greetings. ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
[HACKERS] Should we still require RETURN in plpgsql?
As of CVS tip, plpgsql handles output parameters, in the style CREATE FUNCTION sum_n_product(x int, y int, OUT sum int, OUT prod int) AS $$ BEGIN sum := x + y; prod := x * y; RETURN; END; $$ LANGUAGE plpgsql; The RETURN statement is kinda useless in this example, but it is still required, because we don't allow control to fall off the end of a plpgsql function without causing an error. I am thinking we should allow exit by falling off the end of the function when (a) it has output parameter(s), or (b) it is declared RETURNS void. Comments? How does Oracle's PL/SQL handle this? regards, tom lane ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] Should we still require RETURN in plpgsql?
I am thinking we should allow exit by falling off the end of the function when (a) it has output parameter(s), or (b) it is declared RETURNS void. Comments? I agree - makes sense. Chris ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Should we still require RETURN in plpgsql?
On Tue, 5 Apr 2005, Tom Lane wrote: CREATE FUNCTION sum_n_product(x int, y int, OUT sum int, OUT prod int) AS $$ BEGIN sum := x + y; prod := x * y; RETURN; END; $$ LANGUAGE plpgsql; The RETURN statement is kinda useless in this example, but it is still required, because we don't allow control to fall off the end of a plpgsql function without causing an error. I am thinking we should allow exit by falling off the end of the function when (a) it has output parameter(s), or (b) it is declared RETURNS void. Comments? The above code example do not have any RETURNS clause, does that mean that it defaults to RETURNS void? I don't see what (a) has to do with anything. The return value is independent of in/out:ness of the parameters, isn't it? -- /Dennis Björklund ---(end of broadcast)--- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] Should we still require RETURN in plpgsql?
Dennis Bjorklund [EMAIL PROTECTED] writes: On Tue, 5 Apr 2005, Tom Lane wrote: CREATE FUNCTION sum_n_product(x int, y int, OUT sum int, OUT prod int) AS $$ BEGIN sum := x + y; prod := x * y; RETURN; END; $$ LANGUAGE plpgsql; The above code example do not have any RETURNS clause, does that mean that it defaults to RETURNS void? No, it effectively RETURNS record, where the particular record type is implied by the set of output parameters. See my previous proposal. regards, tom lane ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] Should we still require RETURN in plpgsql?
On Tue, 2005-04-05 at 03:43, Tom Lane wrote:
Dennis Bjorklund [EMAIL PROTECTED] writes:
On Tue, 5 Apr 2005, Tom Lane wrote:
CREATE FUNCTION sum_n_product(x int, y int, OUT sum int, OUT prod int) AS
$$
BEGIN
sum := x + y;
prod := x * y;
RETURN;
END;
$$ LANGUAGE plpgsql;
The above code example do not have any RETURNS clause, does that mean that
it defaults to RETURNS void?
No, it effectively RETURNS record, where the particular record type is
implied by the set of output parameters. See my previous proposal.
While it is useless in this example, istm it only makes things more
confusing to require return in some cases but not in others. Is there
some technical advantage to dropping it?
Robert Treat
--
Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL
---(end of broadcast)---
TIP 6: Have you searched our list archives?
http://archives.postgresql.org
Re: [HACKERS] Should we still require RETURN in plpgsql?
Robert Treat [EMAIL PROTECTED] writes: While it is useless in this example, istm it only makes things more confusing to require return in some cases but not in others. Is there some technical advantage to dropping it? It's about the same either way as far as the code is concerned. But I've only written a dozen or so plpgsql functions using OUT parameters, and I've already found having to write a useless RETURN to be tedious (not to mention that I forgot it a couple times). I don't think I'll be the last one complaining if we leave in the requirement. Basically the requirement exists to make sure you don't forget to define the return value. But when you're using OUT parameters, the existence of a RETURN statement has nothing to do with defining the return value. regards, tom lane ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] Should we still require RETURN in plpgsql?
On Tue, 5 Apr 2005, Tom Lane wrote: Robert Treat [EMAIL PROTECTED] writes: While it is useless in this example, istm it only makes things more confusing to require return in some cases but not in others. Is there some technical advantage to dropping it? It's about the same either way as far as the code is concerned. But I've only written a dozen or so plpgsql functions using OUT parameters, and I've already found having to write a useless RETURN to be tedious (not to mention that I forgot it a couple times). I don't think I'll be the last one complaining if we leave in the requirement. Basically the requirement exists to make sure you don't forget to define the return value. But when you're using OUT parameters, the existence of a RETURN statement has nothing to do with defining the return value. what if not require RETURN iff OUT parameter is defined ? regards, tom lane ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED]) Regards, Oleg _ Oleg Bartunov, sci.researcher, hostmaster of AstroNet, Sternberg Astronomical Institute, Moscow University (Russia) Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/ phone: +007(095)939-16-83, +007(095)939-23-83 ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [HACKERS] Should we still require RETURN in plpgsql?
Oleg Bartunov oleg@sai.msu.su writes: what if not require RETURN iff OUT parameter is defined ? That's what I'm suggesting ;-) regards, tom lane ---(end of broadcast)--- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
