Dear all,
i am using the XLS formatting nodes from Continental. Sometimes I produce errors and get a message like this:
Execute failed: The provided input table is not a valid XLS control table. See log for details.
Now my question is: Where do I find the log file?
Dear @JLD, with this, the regular KNIME log file is meant.
Regarding the validity of an XLS control table: You can also visually inspect the table:
Hi @arbe, thanks a lot for your quick answer.
I was curious, about the ādetailsā from the log entry āSee log for details.ā
I checked the knime log and the error message which I found there seems to be the same as in the KNIME console, except that it has a timestamp. I also changed the log level to DEBUG in order to see every message, but I received no more relevant info.
Besides of this, thank you for this awesome extension. I am using it a lot and I really like the concept behind it.
I have another question, though: While it is possible to define cell background colors via XLS Background Colorizer with RGB values or Hex codes, it is not possible to do the same with the font color. Here I have to use the XLS Font Formatter which only accepts single tags. Letās say Iād like to have 5 different combinations of background and font color. Is there any way to do this with less than 5 XLS Font Formatter nodes?
I have been thinking about some kind of variable loop but I think the little green square (XLS Formatter output type) is a showstopper for this.
Hi @JLD,
your thought is absolutely right. Iād say thatās something for advanced KNIME users, but it is absolutely possible to add formatting instructions via chaining the ālittle green squareā within loops.
We actually have an example workflow for this:
The āModel to Cellā node does the trick.
Hi @JLD ,
Iām quite a fan of the Continental nodes, and also appreciate the effort that has gone into providing them.
With their power comes some complexity though, and to make life easier for āfuture meā, I have created a number of components that encapsulate some of the functionality of the Continental nodes.
Having hit a similar problem re loops that you described, I have a couple of components that make use of the functionality that @arbe has mentioned, but hopefully takes away some of the learning curve by hiding (some of) the complexity.
You place the ācollectorā inside the loop to collect the results and the āretrieverā after the loop to give you back the collected results .
Example of their usage can be found in this demo workflow:
I should add a note that my components donāt do really do anything in the demonstrated use case that is different to the way the node and component work in the Continental demo that @arbe posted, and are merely an alternative (and for me are easier to remember 
).
They do allow for separate formatting models to be collected and retrieved within the same loop, by configuring a unique name under which the different āmodelsā are stored, but this could also be achieved in the Continental demo by adding additional collection ports on the loop end.
