Write to file Question

Question

I am doing some expirimental modeling with a VI that many members of this board greatly helped me get going.  Part of this modeling is the creation of an arbitrary periodic function.  I've accomplished this using 30 "Y-Value" Inputs, and a frequency to create sort of a "unit function", or basically, one period worth of the desired function, with a maximum amplitude of 1, so that it is easily scaled.
What I'm finding is that it would be nice to save these "Unit Functions" along with their corresponding amplitude and frequency to file, that I can later laod back into the program.  Thus, I need to both write the data to a file, (30 pairs of doubles, along with 2 independent doubles), and then have the ability to browse through all my saved functions, and load the desired one.
I've used the express vi "Write to Measurement File" before, but that seems like a clumsy method for this.  So I then played with the various "write to..." vi's available.
For a relatively small set like this, is there any benefit of one file type over another?  Obviously, much the data could go straight into an excel file, while it needs to be reformatted into strings to be stored in text.  I have little or no experience with bianary files.
Are any of the file types easier than others to then load from?
And is it possible to have the available (previously saved) functions selectable from a menu ring, or will I need to prompt the user?
That said, I'm just looking for some general pointers on the area.  I've read all the associated help files, and read quite a few threads on specific issue.  But being new, I'd like to hear how you professionals would approach this function.
Attached is my current code.
Solved!
Go to Solution.
Attachments:
Obtain Unit Periodic Input.vi ‏37 KB

Answer 1

Given your description of what you want to do I would define a typedefed cluster that contains all of your parametric data for your data sets. You might want to allow the user to name these data sets when they save them. You can use these names as the section name. If you pass in the cluster to the write VI with the section name the data will be saved. Using the section name you can later read the data back. You can use the built in config file VI to read the section names from the config file. This will allow you to initialize your ring control. One thing I generally do for my applications is allow the user to select the ini file that want to use. That way they can save multiple versions taliored to their particular task. You could do the same thing and therefore they could have multiple data sets tailored for particular uses. Just offering that as some added flexibility for your application.
Also, I generally don't include the handling of the config files in a state machine. This functionality is generally included in my UI task which is an event structure. Reading/writing/loading config files is more event driven by the UI than something in a processing state machine.
Mark Yedinak
"Does anyone know where the love of God goes when the waves turn the minutes to hours?"
Wreck of the Edmund Fitzgerald - Gordon Lightfoot