Real-Time: Disabling code conditionally when an RT executable

[Mr._Jim]

Hello,

When I build an executable in LabVIEW for Windows, I often enable/disable certain code using the "Application:Kind" property.  For example, it's common to disable the "Quit LabVIEW" function during development, but not in an executable.

Is there something analogous in Real-Time?  I know that property nodes are hit or miss in RT and are generally a bad idea, so is there a way to set up the conditional disable structure?

Thanks in advance,

Jim

[GerdW]

Hi Jim,

atleast for the Quit or Stop-problem I just use a simple case structure...

[Mr._Jim]

Okay, GerdW, thanks... but that doesn't answer my question.  I'm not asking about the quit or stop function; I'm asking about Real-Time, specifically.

Is there a property that will work for what I'm trying to do?  (Keeping in mind that most properties in RT don't work)  I can't use a case structure if I don't have a property, which is why I'm wondering if I can use the conditional disable structure.  Of course, I could just manually disable stuff, but I really want to get away from that.



Thanks again,

Jim

[Ben]

Does "App.TargetOS" help?

Ben

Message Edited by Ben on 11-30-2007 09:43 AM

[Mr._Jim]

Thanks, Ben.

Hmmm... I think I tried that one already, and I'm pretty sure that the OS was PharLap whether the VI was compiled or not.  I could try it again, though.  Maybe in my effort to be succinct I wasn't articulate in explaining what I'm trying to do (sorry about that).

Here are the details:
I found out the hard way that I can't get the strings from an enum using a property node in the RT environment.  It seems to work in the development environment, but as soon as I compile, the property node just gives me an empty array.  So I figured I had to put in arrays of constants, but it would be nice to also keep the property nodes there for development.  I'd like to be able to switch between the two programatically.  Thus, if it's compiled, I want to use constants.  If it's not compiled, I'd like to use the property nodes.

Does that clear any of it up?

Thanks very much for the answers thus far.

Jim

[Ben]

Mr Jim wrote "I found out the hard way that I can't get the strings from an enum using a property node in the RT environment. "

Same here!

It drove me nuts for a while becuase if I had the FP open (loaded) the property node would work fine.

If you just want the equivelent string for the current value, then just wire the enum to a "format into string".

It was also possible to parse the type descriptor for the enum to get the list of strings.

The property I mentioned when combined with a check to see if it was an exe I thought could give you what you were after.

Just trying to help,

Ben

[Mr._Jim]

That's definitely a step in the right direction, Ben, and thanks for the suggestions. 

I use the "format to string" trick all the time with enums, but unfortunately I need the array of names this time.  I set up a server that accepts commands via TCP, and those commands are stored in an enum.  The way I check to see if the user entered a valid command is by doing a search on the 1D string array that I get from a property node.  It would have been super easy to change commands around through a typedef'ed enum and automatically update the array, but it doesn't look like there's any good way to do it in RT.

I digress... I'll have to try those properties you mentioned and see if those help.

Jim

[Ben]

This used to work back in the LV 6.1 days for getting the strings in RT without using a property node.

I believe I have used that as recently as LV 7.1. Since LV 8 changed some data repesentaion stuff, you may have to use the LV 7.1 version of the flatten to string to get the type descritor.

Ben

[Mr._Jim]

That worked perfectly! Like you said, I had to use the old-school 7.1 flatten to string.  It doesn't really answer the property node question, but it solved my problem!  Thanks for your help, Ben.

Jim