Tim Bunce <[EMAIL PROTECTED]> writes: My head is spinning with all the sudden interest in native Oracle execute_array() ... I have Cc'ed dbi-dev, hope that's ok.
> Oh, hang on. I've just realised that I had been confusing these. > Are you saying you have developed your own patch separately to > (and effectively competing with) Kristian's? [CC'd] > > Why not work together? We did work together last summer. From what I understand Chris used the XS part of my patch and modified Oracle.pm to work with the (at that time) new execute_for_fetch() method. My Oracle.pm was based on old DBI which did not have execute_for_fetch(). I haven't seen Chris' patch, but I would expect the XS/C part of it to be identical to mine (feel free to correct me on this). Does Chris' patch work with named placeholders? I will now rewrite the XS part of my patch to be better integrated with the existing scalar bind/execute code. Hopefully this will simplify it somewhat; note though that there will still be a fair amount of code left, for example to get ArrayTupleStatus from Oracle. I understand that this is what Tim needs to integrate the code into the official DBD::Oracle. As to the Oracle.pm modifications I am happy to implement whatever Tim decides on or to use another implementation. If nobody come up with anything I will settle for direct implementation of both bind_param_array(), execute_array(), and execute_for_fetch() in Oracle.pm, mimicking the DBI implementations. I need this anyway for merging the newest DBI/DBD::Oracle into our own internal DBI tree. And now for me it's back to doing some coding. - Kristian. -- Kristian Nielsen [EMAIL PROTECTED] Development Manager, Sifira A/S
