If you are just using the Vernalis benchmarking loop nodes, please ignore this post!!!
Quick question, to which I assume the answer is ‘No’, but want to check…
Has anyone subclassed any of the Vernalis timing loop start / end node classes, or implemented the associated loop start / end interfaces?
That is the following interfaces:
- com.vernalis.knime.perfmon.MemoryPerformanceMonitoringLoopEnd
- com.vernalis.knime.perfmonMemoryPerformanceMonitoringLoopStart
- com.vernalis.knime.perfmon.PerformanceMonitoringLoopEnd
- com.vernalis.knime.perfmon.PerformanceMonitoringLoopStart
and the following classes:
- com.vernalis.knime.perfmon.nodes.timing.abstrct.AbstractMemMonPerLoopEndNodeModel
- com.vernalis.knime.perfmon.nodes.timing.abstrct.AbstractMemMonPerfLoopStartNodeModel
- com.vernalis.knime.perfmon.nodes.timing.abstrct.AbstractPerMonTimingEndNodeModel
- com.vernalis.knime.perfmon.nodes.timing.abstrct.AbstractPerfMonTimingStartNodeModel
The reason is that as of KNIME 4.2 some changes in the KNIME core mean that some of these nodes now break test cases, and fixing them is going to involve breaking the API…
Thanks
Steve