Am 22.08.2011 10:23, schrieb Egon Willighagen: > I was just wondering how this functionality testing is done, but your > answer suggests that is manual work. Now, I could imagine that KNIME > workflow are run in a test suite, with some validation of the outcome, > like number of objects in a few output nodes or so. (I'm just > interested in automated testing :) We have automated test flows, also for the CDK nodes. They essentially compare a "golden table" with the current output of a node. However, we cannot use the old testflows because the two CDK versions cannot be used simultaneously because the serialiazed objects are incompatible. I also have on my list (for quite some time) to set up a public testflow server where all community contributors can upload flows which will then be run automatically every night.
> Something else, which you surely must have been talking about with > Christoph... how do the CDK-KNIME workflows compare to the CDK-Taverna > versions? Will you be writing about that? I'd love to see a > Battle-of-the-CDK-workflows! :) :-) Best regards, Thorsten -- Dr.-Ing. Thorsten Meinl room: Z813 Nycomed Chair for Bioinformatics fax: +49 (0)7531 88-5132 and Information Mining phone: +49 (0)7531 88-5016 Box 712, 78457 Konstanz, Germany
signature.asc
Description: OpenPGP digital signature
------------------------------------------------------------------------------ uberSVN's rich system and user administration capabilities and model configuration take the hassle out of deploying and managing Subversion and the tools developers use with it. Learn more about uberSVN and get a free download at: http://p.sf.net/sfu/wandisco-dev2dev
_______________________________________________ Cdk-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/cdk-user
