drjdpowell

4 hours ago, Rolf Kalbermatter said:

And together with the SO_EXCLUSIVEADDRUSE flag this makes new requests to create a socket on the same port fail with an according error, since the port is technically still in use by that half dead socket. That socket gets eventually deleted and then a new Create Listener call on that port will succeed, unless someone else was able to grab it first.

Are Listener ports affected by this "half-dead" issue?  I would have thought this is just and issue of TCP Connections (with a connected remote party) rather than a Listener.

More details: I'm not using the "Internecine Avoider" and I'm using a "net address" of 127.0.0.1, which I beleive means I'm not going through any network card (all three apps are on the same computer).

11 hours ago, ShaunR said:

What's the error?

Sadly, I don't know as such an error would get lost in the code as written.  This is a rare error in code deployed code on a low-powered single-board computer.  Appears to be a loss of TCP connection, followed by the Client not being able to reconnect (which is why I suspect the Listener dying).  An added clue is that an additional third-party non-LabVIEW TCP server seems to fail and restart itself at the same time (according to entries in its log file.

What work does Type Cast need to do with arrays that it doesn't do with strings, which are just arrays of bytes?  Seems to me that the fastest cast would be something like array of U64 to Doubles, as those are the same size and don't require any length checks (unlike, say, a 9-byte string to array of 8-byte Double).  Is the problem not a missing optimization for U8, but instead nonperformance code for numeric arrays?

Has anyone ever encountered a case where a TCP Listener is created and used, but at some point becomes invalid, throwing an error at the "TCP Wait on Listener" node?  I'm trying to understand a rare issue on a deplyed system that might be explained by that.

In the actual VIM "Read Element by Index of Count" it is an I32, but I think when you connected a U32, and LabVIEW did type propagation through the code of the VIM.  Note sure why this caused your specific issue, but this needs fixing regardless, perhaps just by a convert-to-I32 node.  Thanks.

Again, I've never used PPLs, but that seems wrong.  Why can't you put a class in a library/PPL?

I have never used PPLs, but my understanding is that you would first need to put the PQ library in its own PPL, and have all other code call that PPL.  It looks like you have instead made independent copies of PQ in each PPL.

Here is a quick hack from some existing INI parsing code I had.  See if this works on NI-MAX files.

This VI converts the INI items to JSONpath notation, converts the values to JSON strings unless already a valid JSON value (such and numbers, true, etc.), and uses the JSONtext VIs "Unflatten JSONpath Array to Object" and "Pretty Print" to convert to JSON.

INI to JSON Example.vi

Good luck AQ,

Hope your new employer still lets you attend a conference or two in Europe.

6 minutes ago, drjdpowell said:

That menu should have been installed by VIPM, though. 

Ah, it's because I added submenus in last release, and this added '_1' onto teh menu name.  I'll fix that.

On 5/16/2022 at 3:38 PM, dannyt said:

A very minor comment, the Read Buffer context Help is incorrect and a duplication of the Read Element By Index Contents Help.

Fixed, thanks.

1 hour ago, dannyt said:

I had an issue building a executable using the package as it introduces a missing Item within Malleable Buffer.lvproj

That is the Project I use to develop the library; it isn't meant to be used after install (and doesn't update the various paths, such as to teh examples).   I should probably not include it in the VIPM.    If you'd like to fork the package and develop it, I'd recommend forking the Git repo. 

That menu should have been installed by VIPM, though.  

16 minutes ago, Mads said:

The reason I though the $.[0] notation should work was that if I tested it with https://jsonpath.com/ it (almost) seems to find what I wanted...

JSONpath doesn't have a specification, unfortunately, and so you can't really rely on different implementations matching on edge cases.  I also haven't implemented everything (just due to teh effort required that I don't have time for).

BTW, you should probably explore the "Advanced" subpallet of JSONtext, in particular "Difference.vi" and "Flatten JSON Object to JSONpath Array", as they are useful for this kind of stuff.

17 hours ago, Mads said:

I did successfully rewrite the demo to work with JSON using JSONtext, but then I just treated each array element as a separate JSON string and ran Set item on each of them with the path $.<Label.text>

That would be a perfectly good choice.  You could also use an array of paths like '$[n].Enabled' and call Set Item in a loop.  

Aside: Note that you could also use a JSON Object rather than an array to hold your list of targets.  Like this:

{
  "My Inst X":{"Enabled":true,"Hello":-2,"Hi":6},
  "My Inst Y":{"Enabled":true,"Hello":-2,"Hi":6},
  "My Inst Z":{"Enabled":true,"Hello":-2,"Hi":6},
  "My Other Instrument":{"Enabled":true,"Something Else":true}
}

I mention this as I have noticed that LabVIEW programmers often get stuck in a mental model of JSON Objects and Arrays mapping onto LabVIEW Clusters and Arrays.  In particular note that JSON Objects are not fixed type and number or elements at edit time (unlike Clusters), and JSON Arrays are not restricted to all elements having the same type (like LabVIEW Arrays).

I also made the last instrument in my example one with a differences in its config structure.  By using JSON, you can still let your User change common settings (like "Enable") even with instruments that cannot be represented as the same LabVIEW Cluster. This is an example of "Duck Typing", which can be difficult to do in LabVIEW otherwise.

17 hours ago, Mads said:

and try using Set item with the path $.[*].Enabled e.g. when that value changes, it returns an error. Accessing just one of the indexes ($.[0].Enabled e.g.) fails as well.

You should be using $[*] or $[0] to indicate Array elements; $.[*] indicates all items in a JSON Object and $.[0] is the Object item named "0".  Look at the detailed Help page for JSON Path notation in JSONtext.

Two suggestions:

1) Consider using JSON as your config-data format, rather than clusters.  Using JSONtext to manipulate JSON will be faster than using OpenG tools to manipulate clusters.  

2) Programmatically get an array of references to all your config-window controls and register a single Event case for Value Change of any one of them.  Then use their (hidden) labels to encode what config item they set.  For example, your control with the caption "Baud Rate" could have the hidden label "$.Serial Settings.Baud Rate" which is the JSONpath to set in your config JSON (or config clusters).

Tried it on LabVIEW 2021sp1, and it seems to have at least fixed the problem with a broken VI requiring manual recompile.  No working VI with apparently broken wires yet, either.  Possibly I should just up the base version of this package to 2021.

On 3/19/2022 at 2:08 PM, Dataflow_G said:

The LabVIEW 2021 SP1 bug fixes mention a fix (bug ID: 1596011) for vims and erroneously broken wires, but I've not tried it out yet. I really hope that's the end of the broken wire quirks with malleable VIs, as they're a really great feature.

Thanks. I see there is also this one that may be the problem I'm having:

1513139        Malleable VIs Are Sometimes Erroneously Broken When Operating on Cluster Data Value References

Put a new 0.5 version on VIPM.io    Remember, this is still beta code.

@AQ,  I've incorporated most of your improvements, and made some changes myself.   Question, though.  These VIMs seem plagued by (in LabVIEW 2020) strange recompile bugs: broken wires that are usually fixed by doing a manual force recompile on the VI.  Though often this still leaves some broken wires in a runable VI?!?  See image.

What do you think causes this and is there a way I can avoid it?  

18 hours ago, Виктор Зиновьев said:

I tried to add &cache=shared or with ? to the file's path but it does not work

Did you add "file:" to the front to make a URI path.  And what do you mean by "does not work"?  What are you expecting shared-cache mode to do exactly?  Shared Cache has nothing to do with multiple computers accessing a db (and yes, that is something you would be better off using Postgres).

I have never tried shard cache mode.  Are you sure you actually need it?  If so, try using a URI path to set it following this:

https://www.sqlite.org/c3ref/open.html#urifilenameexamples

On the off chance anyone does use this, I have a beta with improvements: https://forums.ni.com/t5/JDP-Science-Tools/Beta-of-Cyth-SQLite-Logger-1-7-2/m-p/4208206

Biggest improvement might be a highlighting mode, as an alternative to the usual filtering, so entries matching the pattern show up in blue: