KNIME_repeatedly process wages with user-selection

Demo workflow for how we might record previous user-selection so that we don't repeatedly process already-processed records on subsequent invocations of the workflow.

We have an Excel spreadsheet consisting of Name, Date of Birth, Wage and a column for Wage Increase. The idea is that on each run of the workflow, a user could select one or more rows from the Table View and when selected, a 10% wage would be applied to the selected rows.

We want to present to the user the entire file of wages each time, but only allow them to process somebody if they haven't been processed before. Clearly we could just not allow processing if the "increase" column showed non-zero ;-) but let's ignore that minor detail and work on the idea that we only process a row if the "previously selected" flag that we've stored away on a csv file is not set to "Y".

On the Table View, the user can select any row they like (I couldn't see how to prevent that ) so after returning from the Table View, a Rule Engine 'de-selects' any row that was previous selected.

The Wage Increase action takes place on the required rows, and the "previously selected" flags are stored away in a CSV file for processing again on a future run of the workflow.

In the attached workflow, to I have used a couple of "home grown" components. These have blue-box annotations near them. They are here because it is easier for me to produce the working demo with these to simplify a few things, but they aren't really related to the mechanics of the demonstration. You are free to make use of them or replace with your own nodes and processes as you see fit.

The boxes in red indicate the nodes that are just used to "reset" the workflow to an initialised state. You can activate those nodes by configuring the CASE switch to use port 0. Set it back to 1 again to run (and re-run) the main branch of the workflow .