Carlos, > We are going to need to implement the version 11 of the Firebird > Protocol and review the buffer handling code to make better checks on > buffer lengths and server resposnses. > > Now the Firebird server supports batching of messages ( not available > for all of them ), so we can send several messages in the same request > and retrieve all the answers at the same time.
These two examples that you posted, where did you get them? I ask, because I think that would be to our (.Net and Java drivers) benefit to document these changes somehow. Do you think it is worth the effort? Is it feasible at all? > For the new GDS implementation, my first thought is that we can make all > the public methods on the current GDS implementation virtual, and > override ( and rewrite ) them in the new GDS implementation ( i will try > to check if that approach will be possible or not ... i think it's but > ... better to be sure XD ) Looks good. In Jaybird I was thinking about introducing new methods into GDS interface that would do similar things to your Prepare() and Execute() methods. However I would need to do some refactorings to be able to handle different protocol versions. Also, as I remember from my discussions with Dmitry and Vlad, we need to have some buffer for the calls that can be delayed (e.g. isc_dsql_free), but it seems that it can be hidden somewhere in the GDS implementation (or maybe even in the stream class). Roman ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Firebird-net-provider mailing list Firebird-net-provider@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/firebird-net-provider
