You can see the Test JSON Low-Level.vi as an example attached to the following post: https://forums.ni.com/t5/JSON-Toolkit-for-LabVIEW/JSON-Toolkit-Support-for-LabVIEW-2010/gpm-p/341358...
As recommended in that post, if you are able to use the LabVIEW flatten to JSON and unflatten from JSON primitives available in LabVIEW 2013 and up then you are likely to get much better performance.
Hi linna.ma and Milan,
as I commented in other post (i don’t remember where), it depends that native jain parser are better... for example in my case this is not true (and this is way I use i3 and developed some new vis to perform the libraries).
In my case (I’m italian) I can find italian keyboard on a pc, so the systems has “,” for decimal separator and “.” for 0’s grouping (for example 1.000.000,56. By this way, even if i uncheck the flag in the labview options, the jason native parsing interprets an array [1,3] as a dbl 1.3....
This is an old and well known bug not already fixed (this parsing error is the same in other vis that i do not remember which are)
I am trying to reproduce the behavior you describe in LabVIEW 2018 with the built-in Unflatten from JSON function but I am not seeing the behavior being described:
From that screenshot you can see the environment is configured to show numbers with ',' as the decimal separator and it appears that the JSON string "[1,3]" is being correctly turned into an array of length two with values one and three. This is using Windows 10 with LabVIEW 2018 and the following region settings:
I also searched internally and found CAR 400245 which describes the behavior you mention in the LabVIEW 2013 Beta but that appears to have been fixed before LabVIEW 2013 released, so I would not expect any release version of LabVIEW to have the described behavior.
Do you have a situation where you can reproduce the behavior being described? If you do see the behavior in LabVIEW can you provide more information about the configuration of machine and installed LabVIEW version?