> > On Mon, 2004-02-23 at 14:36, Jeff Urlwin wrote: > > > > Which raises another issue: at present DBD::Teradata allocates a > > > $row > > array with a size equal to the sum of the number of > > columns for each > > statement. I've tried adjusting the NUM_OF_FIELDS on the fly to > > represent just the number of fields for > > the current statement, > > but DBI carps. And of course, it presents some issues for providing > > statement metadata. Any chance > the more_results() work > will provide > > a more elegant solution ? > > > > Actually, check out DBD::ODBC. If you need help, let me > know, but I > > did do some hacks that I got from DBD:Sybase which clears out > > NUM_OF_FIELDS, etc... > > Looks like we've all borrowed from one another in this respect :-) > > And yes - there's quite a bit of hackery going on...
And yes, I think we all agree that it's time for the DBI to specify this, so others don't go through the same pain ;) I'm not sure what the approach should be from here on, but documentation of some DBI spec and some DBI helper functions to reset the cached attributes would be a start. I can volunteer for some of it, but I certainly do not feel that I'm the expert covering all databases ;) Regards, Jeff
