LabVIEW
Question related to stopping sub vi loop
@crossrulz wrote:
@mcduff wrote:1 recommendation though, you probably do it already, but when sending string commands I like Case Insensitive Match, so STOP, Stop, StOp, ... will work.
Absolutely. I abuse that functionality. Do note you will have a slight performance hit. But the protection from miscapitalizations is worth it.
Sending an (type def) enum will solve that more elegantly, IMHO.
Of course, just to mention the option, not to recommend it, the main VI could also use VI Server to simply abort the VI... No way to build an application, but still a valid answer to the OP.
Search LabVIEW like a graph!
@crossrulz wrote:
As an aside...I often run into people who use the Producer/Consumer and stop the consumer when the queue is empty. This is BAD because you might be stopping prematurely. You should have some type of message in your queue to tell the consumer when to stop. Of course, Bob doesn't have that issue because it is built into Channel Wires.
You could however stop any loop when the queue is destroyed. Then it will return an error... Not the prettiest practice, but not really any harm in that. You won't get any elements over the queue anyway.
@Gaffinator wrote:
The program that I am writing is supposed to be a practice example for me to test the waters of data Acquisition. The requirements for this program was that 1) I needed to have a daq Engine sub VI that I can spawn using the Asynchronous start and wait functions. The DAQ engine will read in analog and digital voltages and will display them to a chart. as well as display the analog out.
That are silly requirements. Who would require that and why?
@Gaffinator wrote:
2) I need to be able to control this Sub VI from the main front panel application.
Now that is a sound requirement...
@Gaffinator wrote:Some things to note is that my boss would prefer me to not use global variables,
He seems like a smart guy...
@Gaffinator wrote:and not use a queue to drive the sub vi.
And now it's gone...
@Gaffinator wrote:
I have decided on using custom events to drive my DAQ engine as someone has recommended.
I have run into a couple of new problems. The main one being that I am able to start my DAQ engine from my front panel, but for some reason my DAQ engine doesnt recognize my Stop user event.
Any examples or insightful help would be awesome.
I appreciate it a lot guys
It would be more productive if you'd show some code. Pretty sure there are User Event examples shipped with LabVIEW.
@Gaffinator wrote:
My boss told me that I should be using the company template
My sincere condolences.
Knowing you can't change the situation (or don't\shouldn't want to try)...
I've seem an about box implemented as a producer\consumer (two while loops, event structure, two case structures and a queue) because that 'was the template'. Only the event structure was needed. That's just wrong.
Just my 2 cents.
wiebe@CARYA wrote:
@crossrulz wrote:
As an aside...I often run into people who use the Producer/Consumer and stop the consumer when the queue is empty. This is BAD because you might be stopping prematurely. You should have some type of message in your queue to tell the consumer when to stop. Of course, Bob doesn't have that issue because it is built into Channel Wires.
You could however stop any loop when the queue is destroyed. Then it will return an error... Not the prettiest practice, but not really any harm in that. You won't get any elements over the queue anyway.
That is how NI has always taught the Producer/Consumer. Yes, it does work. However, you will lose any data that has not been processed at the time of the queue being destroyed. That is a MAJOR issue in 95% of the projects I have done where P/C fit the need. So over the years I have learned that you almost always want your consumer to maintain the queue (create, read, destroy) while everybody else can only write to it. I do this via a library for the consumer with an Action Engine that is private and a public method for allowing other loops to write to the queue.
