NumberFormatException // ArrayIndexOutOfBoundsException

[Holzhauer]

First, I had to switch the working directory to a directory below the project folder (e.g. 'src'), because otherwise a lot of files under "../data" cannot be found. Furthermore, after running the launcher I get a few NumberFormatException (java.lang.NumberFormatException: For input string: ""), but the GUI appears still. When initialise the model I get an ArrayIndexOutOfBoundsException (java.lang.ArrayIndexOutOfBoundsException: Array index out of range: 20), originating at uk.ac.dmu.iesd.cascade.agents.prosumers.HouseholdProsumer.initialiseSpaceHeatParams(HouseholdProsumer.java:2003). Could you assist me with these issues? Let me know in case further information is required. Thanks!

[rsnape]

Dear Sascha,

thanks for your interest! This project has been a bit dormant for some time. I will try to fix some of these issues. I think I know the root of them - it is with the default parameter values. When I originally wrote the framework the default value for ticksPerDay was 48, but I fear that the new Repast GUI selects the lowest available value (I think 1). There is probably some bad code in HouseholdProsumer.java that assumes the ticksPerDay where it should read the parameter. Also - the issue with the working directory is not nice, but perhaps due to a non-intuitive default. I will try to do an install and check it out. Can I confirm that if you change the working directory and set the parameter "ticksPerDay" to 48, the code runs for you?

[Holzhauer]

Dear Richard, thanks for your rapid reply! Yes, setting "ticksPerDay" to 48 lets the model run. However, I get "java.lang.RuntimeException: Projection 'economicNetwork' not found", and the number of measures remains zero. Best, Sascha

On 19.03.21 22:40, J. Richard Snape wrote: Dear Sascha, thanks for your interest! This project has been a bit dormant for some time. I will try to fix some of these issues. I think I know the root of them - it is with the default parameter values. When I originally wrote the framework the default value for ticksPerDay was 48, but I fear that the new Repast GUI selects the lowest available value (I think 1). There is probably some bad code in HouseholdProsumer.java that assumes the ticksPerDay where it should read the parameter. Also - the issue with the working directory is not nice, but perhaps due to a non-intuitive default. I will try to do an install and check it out. Can I confirm that if you change the working directory and set the parameter "ticksPerDay" to 48, the code runs for you? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAM5HE7MAORE6TZJFK773MLTEPAFLANCNFSM4ZPRBMQQ>.

-- :::: Dr. Sascha Holzhauer :::: Dipl.-Systemwissenschaftler Universität Kassel - FB Elektrotechnik/Informatik - FG: Integrierte Energiesysteme Wilhelmshöher Allee 73, Raum -1517/18 (Untergeschoss) 34121 Kassel Tel: +49 561 804-6513 Fax: +49 561 804-6027 http://ines.uni-kassel.de

[Holzhauer]

Dear Richard, were you able to look into the model yet? Thanks and best regards, Sascha

-------- Forwarded Message -------- Subject: Re: [rsnape/CASCADE] NumberFormatException // ArrayIndexOutOfBoundsException (#2) Date: Fri, 19 Mar 2021 22:51:30 +0100 From: Sascha Holzhauer ***@***.***> To: rsnape/CASCADE ***@***.***>, J. Richard Snape ***@***.***> Dear Richard, thanks for your rapid reply! Yes, setting "ticksPerDay" to 48 lets the model run. However, I get "java.lang.RuntimeException: Projection 'economicNetwork' not found", and the number of measures remains zero. Best, Sascha

On 19.03.21 22:40, J. Richard Snape wrote: Dear Sascha, thanks for your interest! This project has been a bit dormant for some time. I will try to fix some of these issues. I think I know the root of them - it is with the default parameter values. When I originally wrote the framework the default value for ticksPerDay was 48, but I fear that the new Repast GUI selects the lowest available value (I think 1). There is probably some bad code in HouseholdProsumer.java that assumes the ticksPerDay where it should read the parameter. Also - the issue with the working directory is not nice, but perhaps due to a non-intuitive default. I will try to do an install and check it out. Can I confirm that if you change the working directory and set the parameter "ticksPerDay" to 48, the code runs for you? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAM5HE7MAORE6TZJFK773MLTEPAFLANCNFSM4ZPRBMQQ>.

-- :::: Dr. Sascha Holzhauer :::: Dipl.-Systemwissenschaftler Universität Kassel - FB Elektrotechnik/Informatik - FG: Integrierte Energiesysteme Wilhelmshöher Allee 73, Raum -1517/18 (Untergeschoss) 34121 Kassel Tel: +49 561 804-6513 Fax: +49 561 804-6027 http://ines.uni-kassel.de

[rsnape]

Hi Sascha. I'm working with it right now, actually. I think what I will do shortly is check in my current working version so that it is baselined to that.

I have discovered a few awkward corners so am thinking of streamlining the code (removing some of the research ideas that didn't get used in the end). For example, some absolute paths are hard-coded that should be relative and similar. It didn't matter when the model was in constant use by the team 2010-2015, but returning to it now, I see how hard it is to use.

The runtime error you found seems to indicate that the economicNetwork for the model was never created, which might mean that you have tried to run a model with no prosumers in it. Can I ask which scenario directory you were trying? I suspect that there is an issue with the scenario setup if it is a runtime error you see. You could also paste the error text, which will help me track it.