Loading...
 

ABE debugging

 This page describes some tipps and tricks of using ABE, and some hints on efficient debugging of the ABE system.

Logging

You can control the amount of log messages that is created by ABE, and you can add user-defined debug messages. All log messages end in the iLand log file.
The log function fmengine.log adds to each log message the current stand and the current year. This allows efficient filtering of messages for stands and/or years.
The fmengine.verbose property controls the level of details for system generated messages. Note, that verbose can be enabled also only for a short time.
The level of detail of log messages can also be increased for individual stands by using the stand.trace property.

Running code snippets / interactive mode

Full reload of ABE

This triggers setting up the (ABE-related) landscape, loading of all Javascript source and running initial code (e.g., for selecting STPs for stands).

fmengine.reloadABE()

Load and run test code with iLand

Additional javascript code may be loaded within the iLand-viewer using the “load script” and “reload script” buttons in the Scripting tab. This usually just adds (or overwrites) code to the active Javascript environment, which can be called manually from iLand.
It is useful for testing to tell ABE to run code in the context of a stand – this enables the various links between Javascript and C++ (e.g., the stand, activity, stp, agent objects).

function test() {
   fmengine.standId = xxx; // set to some valid stand Id
   <code executing in the context of ABE, e.g.>
   fmengine.log(stand.volume); // print the volume of the stand ‘xxx’ 
}

Execute a management activity

Running an activity in the context of a specific stand is also possible:

fmengine.runActivity(standId, activity)
fmengine.runActivityEvaluate(standId, activity)

with standId a valid ID, and activity the name of an activity (as provided in setting up the STP). The runActivity function calls the “main” routine of the activity, effectively bypassing constraints and the scheduling of events. The runActivityEvaluate variant executes not the main execution slot, but the evaluate-function (for scheduled activities).

citation

Rammer, W., Seidl, R., 2015. Coupling human and natural systems: Simulating adaptive management agents in dynamically changing forest landscapes. Global Environmental Change, 35, 475-485.


Created by werner. Last Modification: Friday 23 of September, 2016 10:23:27 GMT by werner.