infinitenothing

No, the queue only contains elements of the child class. I guess I could switch my queue over to the base class. It just seems a little better to use the most specific class possible and I don't understand why the refnum input can't be made dynamic

I'm not clear how adding another dynamic dispatch input solves my problem. The queue ref is still statically dispatched and will only take a queue of the parent class.

In terms of intent, I'm looking to improve reuse. Rather than writing an enqueue function in each child class, I just want one VI to handle that.

I can post a project but I'm hoping this image makes clear what I want to do.

I want to be able to reuse my StartRun's Enqueue Serial Number in the same way that I'm able to use StartRun's Write Serial Number for its child's object.

I'm referring to the dynamic dispatch terminal setting being unavailable (see below). I was hoping that I could wire a queue of ChildA or a queue of ChildB into that terminal and it would autoadapt (assuming ChildA and ChildB inherited from the StartRun class

I want the children of the "start run" class to be able to use this method but I'm noticing it doesn't let me set the queue refnum input to dynamic. Is there some other way to accomplish this? 

Nothing sticks out to me. Have you tried the memory profiler to see what subVI is giving you problems?

http://zone.ni.com/reference/en-XX/help/371361P-01/lvdialog/profile/

To clarify the second image was posted before the forum update. Here's the thread:

https://lavag.org/topic/20671-programmatically-select-all-boolean-best-way-to-avoid-race-condition/?tab=comments#comment-125949

So, I'm guessing the forum is stripping the metadata on upload now. I reposted that image in this thread by copying the link into the editor box. The box "embedded" the image for me automatically.

This snippet works for me:

So does this one:

I want to clarify that as far as I know there's no standard way to store an image in a SQLite database. That is, @Gepponline, your question has nothing to do with SQLite much less this toolkit. As far as we know, SQLite Administrator uses a proprietary binary format for storing the images as blobs. You should really go to that community to determine what that format is. Unfortunately, that project looks like it was abandoned with no source code available so you might just be out of luck.

The "Data Changed?" property of the action output doesn't trigger the data change event. You have to do that manually. The data change event is triggered when the control is accessed as an indicator, property node, local, etc. 

I can't get snippets working. Did they die in the recent site update? Example:

I use the editor that shows up at the bottom of the thread and the "choose files" link. Using Chrome.

I suggest you break your question down into parts as different threads and forums will be able to answer each part better. It looks like your SQLite stuff is correct and your issue is just flattening and unflattening so the machine vision forum might be a better place to ask for help. In the interim, here's one way to do your flatten unflatten using the IMAQ toolkit.

The attached code shows how SQLite will interpret various types into an integer. It makes me a little nervous that there's no warnings or other way to determine that a value might not match what was requested. I guess I could check every value for its column type. Is there an easier way that I'm missing?

I believe they are in the reference array according to the tab order
You either have to know the names in advance or carry their names with their values.

No, it will break if you try and convert a U8 into U16 so it's a little conservative in that manner.  The programmer would be expected to do the U8 to U16 upstream to "get rid of the broken arrow". It would be really cool if there was a way to allow all the "safe cooerces".

Coerce to type is useful for getting nontypedeffed data into a typedef form but there's a small problem of type safety. For example, if you try and coerce an I32 to a U8, the value will silently wrap without even so much as a coercion dot.

But in recent versions of LV, we have something to ensure a little safety— "Assert Structural Type Match"

And fortunately, for this use case, it's not so sensitive as to disallow typedeffed data from matching it's plain counterpart so we can use it to enforce a runtime error if someone wires an I32 up. And, to make this new function more convenient, you can wrap it up in a VIM. 

Search the downloads page on their site or try this:

http://search.ni.com/nisearch/app/main/p/bot/no/ap/tech/lang/en/pg/1/sn/ssnav:dwl/q/vivado/

Cross posted:

https://forums.ni.com/t5/Real-Time-Measurement-and/LabVIEW-2018-and-SbRIO-9626/td-p/3830776

No, by current spec I mean:

Quote

±16 mA all channels maximum; ±4 mA per channel typical

 

By "how are you connecting", I meant connecting to the analog side of the module. For example do you have one of these: http://sine.ni.com/nips/cds/view/p/lang/en/nid/210164

I see that you have the module setup for differential, that's a little more tricky right? Can you try single ended? maybe with a battery or a potentiometer. Otherwise you have to watch out for things like common mode voltages and such.

Many people use variant attributes for trees.

I'm a little curious about your time sensitive use cases, it almost sounds like you want a sorted list instead of a tree which, I think there are some examples of using maleables in 2018. 

I think being able to use variants in the in place structure may have opened some doors:

This is pre 2012 but it's a implementation of some data structures using DVRs and OOP

I assume you've configured the 9205 for single ended vs differential correctly?

Are you within the current spec of the 9264?

How are you connecting the modules? A cable or one of NI's breakout boards?

Try and simplify things as much as possible. For example, use a battery as a voltage source directly to the input and disconnect the outputs to rule out loading issues.

Does this only affect users that get updates from the repositories? because that seems like a fairly small minority.

If you make an idea on the exchange, I'll vote for it.

I made a snippet to illustrate what I was thinking. I have a Basler dart camera that I was able to test with and it looked OK in theory. You'll notice it is in continuous mode. I'm hoping that since trigger and the exposure both go through property nodes, the settings will apply in that order. There might be a theoretical race condition that would apply the new exposure to a capture in progress, I guess for that, you could capture the exposure before and after and have some sort of indeterminate case but I think it would be nearly impossible.