MIME-Version: 1.0
References: <CACX+KaPBaSW+0PyMC7dmY6W9W8mPSbWpkJRhAOQOC9Qb0TjCQA@mail.gmail.com>
In-Reply-To: <CACX+KaPBaSW+0PyMC7dmY6W9W8mPSbWpkJRhAOQOC9Qb0TjCQA@mail.gmail.com>
From: Ron Johnson <ronljohnsonjr@gmail.com>
Date: Tue, 18 Jun 2024 13:07:16 -0400
Message-ID: <CANzqJaA-Bmsq5npa1qCaQkDMb-4+9ZqZVUHqpVzURJ-RtSgS-w@mail.gmail.com>
Subject: Re: Seeking Clarification on Function Definitions in PostgreSQL Extensions
To: "pgsql-generallists.postgresql.org" <pgsql-general@lists.postgresql.org>
Content-Type: multipart/alternative; boundary="000000000000cc1173061b2d1dfc"
Archived-At: <https://www.postgresql.org/message-id/CANzqJaA-Bmsq5npa1qCaQkDMb-4%2B9ZqZVUHqpVzURJ-RtSgS-w%40mail.gmail.com>
Precedence: bulk

--000000000000cc1173061b2d1dfc
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

fun1 returns a table set just like any other SELECT statement.
fun2 puzzles me.  Why would you return parameters AND *a single record* (un=
less
it's an error status).
fun3 just returns two parameters.  Why isn't it a procedure?

fun2, returning parameters AND a function value, would have made my Comp
Sci professors very, very angry.  Definitely Bad Practice.

You choose which to use based on how much data you want to return.

On Tue, Jun 18, 2024 at 12:50=E2=80=AFPM Ayush Vatsa <ayushvatsa1810@gmail.=
com>
wrote:

> Hi PostgreSQL community,
> Recently I am exploring extensions in postgres and got a little confused
> regarding the function definition present in SQL file. For example consid=
er
> below three functions:
>
> CREATE FUNCTION fun1(integer)
> RETURNS TABLE(
>         col1 integer,
>         col2 text
> )
> AS 'MODULE_PATHNAME', 'fun1'
> LANGUAGE C;
>
> CREATE FUNCTION fun2(
>     IN  input integer,
>     OUT col1 integer,
>     OUT col2 text
> )
> RETURNS SETOF record
> AS 'MODULE_PATHNAME', 'fun2'
> LANGUAGE C;
>
> CREATE FUNCTION fun3(
>     IN  input integer,
>     OUT col1 integer,
>     OUT col2 text
> )
> AS 'MODULE_PATHNAME', 'fun3'
> LANGUAGE C;
>
> 1/ I wanted to know what's the difference between the above three
> definitions.
> As per my understanding, "fun1" and "fun2" look the same, taking one
> integer and returning two columns with multiple rows.
> Whereas "fun3" is used when we are returning only one row, but my doubt
> for "fun3" is that, Can the above definition (used for fun1 and fun2) cov=
er
> both single and multiple row scenarios.
>
> 2/ How does someone decide which type of definition is to be used?
>
> Regards
> Ayush Vatsa
>

--000000000000cc1173061b2d1dfc
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>fun1 returns a table set just like any other SELECT s=
tatement.</div><div>fun2 puzzles=C2=A0me.=C2=A0 Why would you return parame=
ters AND <i>a single record</i>=C2=A0(unless it&#39;s an error status).</di=
v><div>fun3 just returns two parameters.=C2=A0 Why isn&#39;t it a procedure=
?</div><div><br></div><div>fun2, returning parameters AND a function value,=
 would have made my Comp Sci professors very, very angry.=C2=A0 Definitely =
Bad Practice.</div><div><br></div><div>You choose which to use based on how=
 much=C2=A0data you want to return.<br></div><br><div class=3D"gmail_quote"=
><div dir=3D"ltr" class=3D"gmail_attr">On Tue, Jun 18, 2024 at 12:50=E2=80=
=AFPM Ayush Vatsa &lt;<a href=3D"mailto:ayushvatsa1810@gmail.com">ayushvats=
a1810@gmail.com</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" s=
tyle=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);pad=
ding-left:1ex"><div dir=3D"ltr">Hi PostgreSQL community,<br>Recently I am e=
xploring extensions in postgres and got a little confused regarding the fun=
ction definition=C2=A0present in SQL file. For example consider below three=
 functions:<br><br><div>CREATE FUNCTION fun1(integer)<br>RETURNS TABLE(<br>=
=C2=A0 =C2=A0 =C2=A0 =C2=A0 col1 integer,<br>=C2=A0 =C2=A0 =C2=A0 =C2=A0 co=
l2 text<br>)<br>AS &#39;MODULE_PATHNAME&#39;, &#39;fun1&#39;<br>LANGUAGE C;=
<br><br>CREATE FUNCTION fun2(<br>=C2=A0 =C2=A0 IN =C2=A0input integer,<br>=
=C2=A0 =C2=A0 OUT col1 integer,<br>=C2=A0 =C2=A0 OUT col2 text<br>)<br>RETU=
RNS SETOF record<br>AS &#39;MODULE_PATHNAME&#39;, &#39;fun2&#39;<br>LANGUAG=
E C;<br><br>CREATE FUNCTION fun3(<br>=C2=A0 =C2=A0 IN =C2=A0input integer,<=
br>=C2=A0 =C2=A0 OUT col1 integer,<br>=C2=A0 =C2=A0 OUT col2 text<br>)<br>A=
S &#39;MODULE_PATHNAME&#39;, &#39;fun3&#39;<br>LANGUAGE C;<br><br>1/ I want=
ed to know what&#39;s the difference between the above three definitions. <=
br>As per my understanding, &quot;fun1&quot; and &quot;fun2&quot; look the =
same, taking one integer and returning two columns with multiple rows. <br>=
Whereas=C2=A0&quot;fun3&quot; is used when we are returning only one row, b=
ut my doubt for &quot;fun3&quot; is that, Can the above definition=C2=A0(us=
ed for fun1 and fun2) cover both single and multiple row scenarios.<br><br>=
2/ How does someone decide which type of definition is to be used?<br><br>R=
egards<br>Ayush Vatsa</div></div>
</blockquote></div></div>

--000000000000cc1173061b2d1dfc--