Netlogo Api Controller - Get Table View

Question

I am using Netlogo Api Controller With spring boot this my code (i got it from this link )

HeadlessWorkspace workspace = HeadlessWorkspace.newInstance();
        try {
             workspace.open("models/Residential_Solar_PV_Adoption.nlogo",true);

             workspace.command("set number-of-residences 900");
             workspace.command("set %-similar-wanted 7");
             workspace.command("set count-years-simulated 14");
             workspace.command("set number-of-residences 500");
             workspace.command("set carbon-tax 13.7");
             workspace.command("setup");
             workspace.command("repeat 10 [ go ]");
             workspace.command("reset-ticks");
             workspace.dispose();

             workspace.dispose();
        }
        catch(Exception ex) {
                  ex.printStackTrace();
        }

i got this result in the console:

But I want to get the table view and save to database. Which command can I use to get the table view ?

Table view:

any help please ?

Answer 1

If you can clarify why you're trying to generate the data this way, I or others might be able to give better advice.

There is no single NetLogo command or NetLogo API method to generate that table, you have to use BehaviorSpace to get it. Here are some options, listed in rough order of simplest to hardest.

Option 1

If possible, I'd recommend just running BehaviorSpace experiments from the command line to generate your table. This will get you exactly the same output you're looking for. You can find information on how to do that in the NetLogo manual's BehaviorSpace guide. If necessary, you can run NetLogo headless from the command line from within a Java program, just look for resources on calling out to external programs from Java, maybe with ProcessBuilder.

If you're running from within Java in order to setup and change the parameters of your BehaviorSpace experiments in a way that you cannot do from within the program, you could instead generate experiment XML files in Java to pass to NetLogo at the command line. See the docs on the XML format.

Option 2

You can recreate the contents of the table using the CSV extension in your model and adding a few more commands to generate the data. This will not create the exact same table, but it will get your data output in a computer and human readable format.

In pure NetLogo code, you'd want something like the below. Note that you can control more of the behavior (like file names or the desired variables) by running other pre-experiment commands before running setup or go in your Java code. You could also run the CSV-specific file code from Java using the controlling API and leave the model unchanged, but you'll need to write your own NetLogo code version of the csv:to-row primitive.

globals [
  ;; your model globals here

  output-variables
]

to setup
  clear-all

  ;;; your model setup code here

  file-open "my-output.csv"
  ; the given variables should be valid reporters for the NetLogo model
  set output-variables [ "ticks" "current-price" "number-of-residences" "count-years-simulated" "solar-PV-cost" "%-lows" "k" ]
  file-print csv:to-row output-variables

  reset-ticks
end

to go
  ;;; the rest of your model code here

  file-print csv:to-row map [ v -> runresult v ] output-variables
  file-flush
  tick
end

Option 3

If you really need to reproduce the BehaviorSpace table export exactly, you can try to run a BehaviorSpace experiment directly from Java. The table is generated by this code but as you can see it's tied in with the LabProtocol class, meaning you'll have to setup and run your model through BehaviorSpace instead of just step-by-step using a workspace as you've done in your sample code.

A good example of this might be the Main.scala object, which extracts some experiment settings from the expected command-line arguments, and then uses them with the lab.run() method to run the BehaviorSpace experiment and generate the output. That's Scala code and not Java, but hopefully it isn't too hard to translate. You'd similarly have to setup an org.nlogo.nvm.LabInterface.Settings instance and pass that off to a HeadlessWorkspace.newLab.run() to get things going.