Why can't I write Variant back into a cluster control?

[jasonhill]

I have three clusters with about 21 controls of various types in each.  I would like to store the values of various configurations and be able to swap them out programatically.  I thought I could be clever and just write them out to a binary file and then read them back in, but I keep getting error 91 "The data type of the variant is not compatible with the data type wired to the type input."   I get the error even when I only have one cluster.

How did the datatype become incompatible?

Message Edited by jasonhill on 05-15-2006 04:27 PM

[chilly_charly]

I'm probably missing something, but I don't understand why, in your write vi, you initialize a variant cluster array with 3 empty clusters then, in the loop, append the read values at the end of the array, leaving the 3 empty clusters as first 3 elements.

[jasonhill]

DOH!

Yeah.  That would be the problem.  Then the empty variant does not fit back into the cluster.   I was so worried about how to treat the variants and which refnum was which that I completely ignored how I was storing the varaints. Thanks, tst.

[Ben]

????

"Thanks, tst"

????

Ben

[jasonhill]

I should just not bother logging in until i have had my first cup of coffee. 

Sorry, CC.  I will shower you with stars in compensation.

[chilly_charly]

Waiting with the soap in hand...

[MartyS]

One other thing: You should close the Controls[ ] refnums. LabVIEW may not (probably doesn't) close them automatically when you close the VI refnum. If not, you'll have the functional equivalent of a memory leak. This kind of thing has happened to me in the past. It's good policy always to close dynamically generated refnums, such as Controls[ ], or Pages[ ] of a tab control, etc.

::Marty

[jasonhill]

Ah.  Good idea.  Not strictly necessary for the current version of LabView (see exhaustive discussion here).  But who knows what future versions might hold.

[MartyS]

jasonhill: Ah.  Good idea [to close dynamically allocated refnums ASAP].  Not strictly necessary for the current version of LabView (see exhaustive discussion here).  But who knows what future versions might hold.

Yes, I would rather close ASAP than rely on LabVIEW's garbage collection or auto-deallocation or whatever you want to call it. There may be situations where a few kbytes matters for a little while. I agree with the posters in the cited thread who say it's good style to close a reference that's no longer needed. Maybe that's overkill, but it's a small price to pay for piece of mind.

::Marty