All Activity

😂😂😂

Heathen My go. Last Mod.vi

If you are in a Windows environment, and have many files to process, this is probably going to be faster. There probably are several factors in determining when doing this in .NET is the better solution.

I also realised I messed up my benchmark and the final High Precision Time should be after the sorting. I meant to do this just forgot!

In addition to the LV native method, there are options with .NET and command prompt: Get Recently Modified Files.

That's how I'd do it. Then combine that with the Foreign Key Sort from my Array package, putting the Time Stamps into the Keys, then paths into the Arrays, and it will sort the paths from oldest to newest. Reverse the array and index at 0, or use Delete From Array to get the last element, which would be the newest file.

I believe to have read somewhere that he eventually conceded that it was basically an unsolvable problem for a generic, easy to use solution. But that may be my mental interpretation of something he said in the past and he may not agree with that.

Done some simple testing. On a directory containing 838 files it took 60 ms.

Have you tried this? The `last mod` output should hopefully give you the timestamp of the last modification, and it would then be pretty simple to find the latest. I have no idea what the performance of this would be if you loop over 10000 files. That is something you would just have to try.

Hi everyone, I have a question: Suppose there's a folder containing a large number of PNG images — say, around 10,000 files. What's the best way to get the name of the most recently created or modified image in that folder? Is there a faster method?

We are on a hiding to nothing as we can't create objects. I abandoned those thoughts over a decade ago (maybe even 2 decades ago ). It is feasible with scripting but so slow and won't work in built applications. The half-way house is to use scripting to make the labview prototypes for typdefs and handlers then load dynamically as plugins but it's a lot of infrastructure just to propagate what are hopefully rare changes. I did play with that (hence my question to hooovahh) but in the end went for a string based solution to avoid it altogether. What happened to AQ's behemoth of a serializer? Did he ever get that working?

The popular serializer/deserializer problem. The serializer is never really the hard part (it can be laborious if you have to handle many data types but it's doable) but the deserializer gets almost always tricky. Every serious programmer gets into this problem at some point, and many spend countless hours to write the perfect serializer/deserializer library, only to abandon their own creation after a few attempts to shoehorn it into other applications. 🙂

That's the easy bit. It's much easier getting stuff into other forms than it is reconstituting them because we can't create objects and primitives. This is why JSON libraries have very straightforward encoders that can take any type but you have all sorts of awkward VI's for getting them out into LabVIEW again. If you are going to use JSON strings you might as well not use LabVIEW types at all . Add the Network Stream endpoint to it and you're good to go. Getting it back out again is where you will find the problems unless, of course, the device uses strings too (SCPI)

No but that is a great suggestion to think about for future improvements. At the moment I could do the reverse though. Given the Request/Reply type defs, generate the JSON strings describing the prototypes. Then replacing the Network Streams with HTTP, or TCP could mean other applications could more easily control these remote systems.

Have you played with scripting event prototypes and handlers from JSON strings?

Yeah I tried making it as elegant as possible, but as you said there are limitations, especially with data type propagation. I was hoping to use XNodes, or VIMs to help with this, but in practice it just made things overly difficult. I do occasionally get variant to data conversion issues, if say the prototype of a request changes, but it didn't get updated on the remote system. But since I only work in LabVIEW, and since I control the release of all the builds, it is fairly manageable. Sometimes to avoid this I will make a new event entirely, to not break backwards compatibility with older systems, or I may write version mutation code, but this has performance hits that I'd rather not have. Like you said, not always elegant.

LabVIEW network streams are fine. I wouldn't worry about the transport too much. Network Streams have a nice way of integrating into API's. What I was looking at were network events where you don't have to synchronise prototypes throughout a system on different machines (changing an event prototype usually breaks an Event Structure). Everything for my events is a string so this is not a problem but it makes parsing tricky. This was one of the reasons I wanted "Named Events" (events can be named like queues can) but they botched the downstream polymorphism. I was wondering whether you had found an elegant way of serialising events (a bit like protocol buffers but without the compiler).

Yes. The transport mechanism could have been anything, and as I mentioned I probably should have gone with pure TCP but it was the quickest way to get it working.

IC. so you have created a cloning mechanism of User Events - reconstruct pre-defined User Event primitives locally with the data sent over the stream?

Variants and type defs. There is a type def for the request, and a type def for the reply, along with the 3 VIs for performing the request, converting the request, and sending the response. All generated with scripting along with the case to handle it. Because all User Events are the same data type, they can be registered in an array at once, like a publisher/subscriber model. Very useful for debugging since a single location can register for all events and you can see what the traffic is. There is a state in a state machine for receiving each request for work to be done, and in there is the scripted VI for handling the conversion from variant back to the type def, and then type def back to variant for the reply. When you perform a remote request, instead of sending the user event to the Power Supply actor, it gets sent to the Network Streams actor. This will get the User Event, then send the data as a network stream, along with some other house keeping data to the remote system. The remote system has its Network Stream actor running and will get it, then it will pull out the data, and send the User Event, to its own Power Supply actor. That actor will do work, then send a user event back as the reply. The remote Network Stream actor gets this, then sends the data back to the host using a Network Stream. Now my local Network Stream actor gets it, and generates the user event as the reply. The reason for the complicated nature, is it makes using it very simple.

Yes. It is the "send user event" that I'm having difficulty with. User events are always local and require a prototype so how do you serialise a user event to send it over a stream?

I have a network stream actor (not NI's actor but whatever) that sits and handles the back and forth. When you want work to happen like "Set PSU Output" you can state which instance you are asking it to (because the actor is reentrant), and which network location you want. The same VI is called, and can send the user event to the local instance, or will send a user event to the Network Stream loop, which will send the request for a user event to be ran on the remote system, and then reverse it to send the reply back if there is one. I like the flexibility of having the "Set PSU Output" being the same VI I call if I am running locally, or sending the request to be done remotely. So when I talked about running a sequence, it is the same VIs called, just having its destination settings set appropriately.

How does this work?

I work in a battery test lab and was the architect for the testing platform we use. Our main hardware is an embedded cDAQ, or cRIO running Linux RT. This is the hardware that runs the actual sequence, and talks to the various hardware. We use the RT platform more for reliability, and less for its determinism. Timing is of course important, but we don't have any timed loops. mS precision is really all we care about. The architecture is built around User Events with asynchronous loops dedicated to specific tasks. These events can be triggered from the RT application, or from a device on the network using Network Streams. A low level TCP would probably be better, but the main sequence itself isn't sent, step by step from Windows. Instead the sequence file is downloaded via WebDAV, then told to run it. The RT then reads the file and performs the step one at a time. Windows from this point is just for monitoring status, and processing logs. This is important to us since Windows PCs might restart, or update, or blue screen, or have any other number of weird situations, and we wanted our test to just keep going along. This design does mean that any external devices that use Windows DLLs for the communication can't be used. But network controlled power supplies, loads, cyclers, chambers, and chillers all are fine. Same with RS-232/485. Linux RT supports VISA, and we plugged in a single USB cable to a device that gives us 8 serial ports. If money were no object I suppose we would have gone with PXI but it is pretty over kill for us. On the Windows side we do also have a sequence editor. The main reason we didn't go with TestStand is because we want that sequence to run entirely on Linux RT. Because the software communicates over User Events, and Network Streams sending User Events, we can configure the software to run entirely on Windows if we want. This of course takes out many of the safety things I talked about, but to run a laptop to gather some data, control a chamber, battery, cycler, or chiller for a short test is very valuable for us. As for the design, we don't use classes everywhere. Mostly just for hardware abstraction. The main application is broken up into Libraries, but not Classes. If I were to start over maybe I would use classes just to have that private data for each parallel loop, but honestly is isn't important and a type def'd cluster works fine for the situation we use it in.

You can script with text files with #1. Just adding a feature to delay N ms gets you most of the way to a full scripting language (conditionals and for-loops are what's left but a harder proposition). However, for a more general solution I use services in #2 with queue's for inputs and events for outputs. There was a discussion ages ago at whether queues were needed at all since LabVIEW events have a hidden queue but you can't push to the front of them (STOP message ) and, at the time, you couldn't clear them so I opted for proper queues. So, architecturally, I use "many to one" for inputs and "one to many" for outputs.