On Sun, Feb 17, 2013 at 7:34 AM, Dick & Hannah <[email protected]> wrote: > Hi. Sorry this is late. > > Do you have more details on... > > * Changes to the way storage subsystems provide SPARQL Update > > ...please? > > As we have a heavily modified wrapper around the current code. :-) >
Hi Dick, The main change is that UpdateEngine is now geared towards *streaming* SPARQL Update operations. This means that any custom engines need to provide an object that can accept update operations as they are being parsed. Please take a look at UpdateEngineMain for the default ARQ implementation that is used for in-memory stores, SDB, and TDB. If you need to continue to receive a fully populated UpdateRequest ahead of time, please look at UpdateEngineNonStreaming for an example that is very much like the old style. Please let me know if you have any more questions as you port your previous implementation. -Stephen
