From: "Greg Sabino Mullane" <greg@turnstep.com>
To: pgsql-interfaces@postgresql.org
Cc: dbdpg-general@postgresql.org
Cc: dbdpg-general@postgresql.org
Subject: Re: DBD::Pg and $sth->{ParamValues}
In-Reply-To: <878ye2c2bv.fsf@openinformatics.com>
Date: Fri,  2 Jul 2004 23:32:30 -0000
Message-ID: <8419477399a5ea6e29452fbf21676c76@biglumber.com>


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
 
 
> What is the status of the this attribute? I would like to help
> implement the code (since I need it for my application).
 
Feel free to jump in. I'm pretty sure it is unsupported at the moment.
 
> But I'm a bit confused about what the status of prepared statements is
> in the code. I'm assuming the the dbd_st_execute() in dbdimp.c is the
> latest code used and that prescan_stmt.c isn't being used.
 
That is correct.
 
>               paramValues = calloc(imp_sth->numphs, sizeof(*paramValues));
>
> but it is never freed, isn't that a memory leak?
 
Yes, probably so. I don't like the use of calloc anyway, so if you
can find a way to use New() in a loop instead, I'd go that way. Feel
free to strip out the 'y' and 'z' ints as well, looks like those
are leftover debugging aids.
 
> this because in order to make the ParamValues attribute work, I'll
> need to make that array part of the statement handle - and I don't
> want to be keeping track of memory that is going to vanish out from
> underneath me.
 
Upon looking at the code again, I think everything you need is already
there, inside of imp_sth->seg->value. But getting rid of that calloc
would still be nice. :)
 
- --
Greg Sabino Mullane greg@turnstep.com
PGP Key: 0x14964AC8 200407021933
 
-----BEGIN PGP SIGNATURE-----
 
iD8DBQFA5fDvvJuQZxSWSsgRAqR7AKCbp97sb/Y1plTdJ0ucTOiNy2EXgACfQmzE
+X0SUFg+ImHdyUAobivwPoI=
=iCjA
-----END PGP SIGNATURE-----