Hi Knut, If your subscriber is 'connected' to partition 'y' its dataReaders will read data that was written into that partition. If you want 'explicit' knowledge on what partition data was published in after being read, you'll need to model that (also) into the data-model itself (i.e. topic-attributes).
-Hans Hans van 't Hag OpenSplice DDS Product Manager PrismTech Netherlands Email: hans.vant...@prismtech.com Tel: +31742472572 Fax: +31742472571 Gsm: +31624654078 PrismTech is a global leader in standards-based, performance-critical middleware. Our products enable our OEM, Systems Integrator, and End User customers to build and optimize high-performance systems primarily for Mil/Aero, Communications, Industrial, and Financial Markets. -----Original Message----- From: developer-boun...@opensplice.org [mailto:developer-boun...@opensplice.org] On Behalf Of Knut Krause Sent: Saturday, May 07, 2011 5:02 PM To: OpenSplice DDS Developer Mailing List Subject: Re: [OSPL-Dev] Sending general data on opensplice Hi Hans, this means I can create partitions later on but cannot do something like "get the data of memeber x from partition y"? More specific: We use Qt and thought about creating a partition for every type of data and then do always somethin like "get data from robot x out of partition 'positions'". Since the data comes from the positions partition and we know it's from robot x so we can decode some kind of QVariant as a position. We then thought of doing the same for things like the distance sensor and so on. Is there another approach you would recommend? I mean I can of course add some model classes to my library and offer methods like "Position getPosition(Robot x)" but this would maybe be unnecessary overhead. regards Knut Am Samstag 07 Mai 2011, 16:54:49 schrieb Hans van't Hag: > Hi Knut, > > Be aware that although you can "connect" your subscriber to various > partitions (even with wildcards), there's no specific data that tells you > via which partition you've read your data (if you're connected to multiple > partitions). > > Typically people put enough data in the data-model to allow 'decoding' the > generic part of your message. > > -Hans > > On 7 May 2011 14:10, Knut Krause <knut.kra...@lagom.de> wrote: > > Hi. > > > > We try to send some data models on opensplice. Now we thought about about > > sending a generic model like a QVariant decoded to some string on a > > general IDL model (just id and message) using different partitions. Due > > to the specific > > partition we know what data we can expect in the generic model and decode > > it > > at the client again. > > > > Are there disadavantages with this approach? > > > > regards > > > > > > Knut > > > > _______________________________________________ > > OpenSplice DDS Developer Mailing List > > Developer@opensplice.org > > Subscribe / Unsubscribe > > http://dev.opensplice.org/mailman/listinfo/developer _______________________________________________ OpenSplice DDS Developer Mailing List Developer@opensplice.org Subscribe / Unsubscribe http://dev.opensplice.org/mailman/listinfo/developer _______________________________________________ OpenSplice DDS Developer Mailing List Developer@opensplice.org Subscribe / Unsubscribe http://dev.opensplice.org/mailman/listinfo/developer
