On Thu, Jul 04, 2002 at 02:40:03PM +0200, Gerald Richter - ecos gmbh wrote:
> >
> > > The ability to share handles, is one of the reason to use
> > > threads, isn't it?
> >
> > It would certainly be nice - but very hard to get right and safe.
> >
>
> Tim, just a question, do you know if the C-Library of Orcale is threadsafe,
Yes, if initialised with the OCI_THREADED option (as per an earlier email).
> so that when DBI would be able to share handles (in the long term), this is
> usefull when using Orcale. I guess it is so, just would like to get a
> confirmation, before I put any thoughs or work into makeing DBI able to
> share handles. The basic goal is to have just one database connection, which
> is shared by all threads. (At least to be able to do so)
A reasonable way to do that for DBD::Oracle (which may apply in
principle to others) is to not try to share a DBI handle but rather
have a way to get/set the underlying C pointer values.
One thread would connect as normal then do
$shared_scalar = $dbh->{ora_oci_pointer_thingy};
and another thread could do
$dbh = DBI->connect(..., ora_oci_pointer_thingy => $shared_scalar);
in which case the DBD::Oracle connect wouldn't actually call OCI
functions to make a connection, it would just plug in the pointer
value[*] it's been passed.
This concept is very similar to calling dup() on a filehandle.
Tim.
[*] ora_oci_pointer_thingy would probably need to be a string
containing several integer pointer values separated by some character
like a colon.
