Re: pl/perl problem - Mailing list pgsql-general

From FERREIRA William (COFRAMI)
Subject Re: pl/perl problem
Date
Msg-id 1904E3EB39448246A7ECB76DF34A70B00143B4A7@TOCOMEXC03
Whole thread Raw
In response to pl/perl problem  ("FERREIRA William (COFRAMI)" <william.ferreira@airbus.com>)
List pgsql-general

yes, it works
exactly what i needed, thanks a lot

-----Message d'origine-----
De : Richard Huxton [mailto:dev@archonet.com]
Envoyé : mardi 22 mars 2005 12:41
À : FERREIRA William (COFRAMI)
Cc : 'Sean Davis'; pgsql-general@postgresql.org
Objet : Re: [GENERAL] pl/perl problem

FERREIRA William (COFRAMI) wrote:
> my function is very long but i found an example with the same comportment :
> CREATE OR REPLACE FUNCTION adoc.totoTest()
>   RETURNS int4 AS
> $BODY$
>  my $var = '->>>';
>  &concat($var);
>
>  sub concat {
>   $var .= 'tagada';
>  }
>  elog NOTICE, $var;
>  return 4;
>
> $BODY$
>   LANGUAGE 'plperl' VOLATILE;

> first execution : ->>>tagada
> second execution : ->>>

In the example above $var in sub concat is NOT an argument provided to
the function. What you've done there is create a named closure (if I'm
getting my terms right) in which the inner $var is allocated on first
call but not afterwards. The second time you run totoTest() the outer
$var (my $var) is a new variable, whereas the inner one still refers to
the original.

If you actually want to return a concatenated string you'd want
something like:

sub concat {
  my $var = shift;
  return $var . 'tagada';
}

If you want to affect an outer variable you'll want something like

sub concat {
   my $var_ref = shift;
   $$var_ref .= 'tagada';
}

Does that help?
--
   Richard Huxton
   Archonet Ltd

This mail has originated outside your organization,
either from an external partner or the Global Internet.
Keep this in mind if you answer this message.

pgsql-general by date:

Previous
From: "FERREIRA William (COFRAMI)"
Date:
Subject: Re: pl/perl problem
Next
From: "Randy Samberg"
Date:
Subject: Re: postgres oracle emulation question