Sometimes people get so lost in the technology, that it’s tough to just do something pretty simple. I totally get that. Pentaho still has plenty of room for improvement on the usability, especially for people coming to the platform for the first time.
Well, here tis.
Unzip examplecsv.zip to pentaho-demo/pentaho-solutions/samples/etl/ in the sample server.
Basically, the idea is to turn a csv file (example.csv):
into this chart
The confusing part, I’m guessing from the thread, was how data gets from KETTLE to PENTAHO.
Not hard at all actually.
Pentaho initiates (ie, calls Kettles API) the Kettle transform and then “slurps” in memory records from a specified step. It’s the UNIX equivalent of the “tee” utility where you’re just watching data arriving at a certain place. In this example, I’ve made it even more explicit by naming the “dummy” step “for_pentaho” so that it’s clear the step that Pentaho is “slurping” the data from.
After that, it’s just a matter of building a chart like any other in the platform.