Henrik Tougaard wrote:
On Mon, Mar 03, 2003 at 10:16:07AM +1100, Stas Bekman wrote:

There should be no perl datastructures involved at all in the pool storage.


Maybe not 'involved' but there's a strong chance the imp_dbh contains SV*'s.

I thought that drivers are pure C.



Just a datapoint:


The upcoming version II DBD::Ingres driver is written in (nearly) pure
C, but contains several SV*'s in both the imp_dbh and the imp_sth
storage.  A SV* is just so much more handy than continously
(re)allocating storage for strings that (may) vary in size.

Why reinvent the wheel when you have a very nive round one at hand?
Especially as the garbage collection problems are neatly handled
by the Perl GC by just mortalising at the right time.

We are talking about SV*s being a constraint when they are to be used by interpreters that haven't created them. It's not that we don't want to handle SV*s, it's just that it's not trivial and in some cases would be impossible.



__________________________________________________________________ Stas Bekman JAm_pH ------> Just Another mod_perl Hacker http://stason.org/ mod_perl Guide ---> http://perl.apache.org mailto:[EMAIL PROTECTED] http://use.perl.org http://apacheweek.com http://modperlbook.org http://apache.org http://ticketmaster.com



Reply via email to