hooovahh

There are complications with this. On Windows the first thing it does it figure out where the controls are, and creates the template HTML pages and web code. Then when it runs it can send the values of controls to the HTML pages, or send images of what the controls look like. If I'm remembering right, on embedded devices, some of the needed functions won't work. I think the get an image of a control doesn't work, even if the embedded UI is enabled. And Writing to PNG using compression doesn't work. But the good news is the values of controls can still be sent and received. At some point I did have this working on devices with an embedded UI as a test and showed it here It looks like some controls are in image mode, so maybe I'm wrong about it not working. The code as it is today is already a bit of a mess, when it comes to examples and demoing the features it has. So that code was an even bigger hack. I do wish I had time to work on this because it is one of the cooler projects I've worked on, and the easiest 80% of the project is done. But the last 20% would take a bunch of work. So it sorta sits here in this sorta works, but needs some attention, stage.

First I'll mention that snippets aren't working recently on LAVA. I messaged the admin but haven't heard anything. Second I'm wondering why this can't be a native LabVIEW feature. Has anyone made an idea on the Idea Exchange for probes to work on parallel For loops, where the P terminals can constant fold to a 1? I'm also not sure why this works only in the enabled structure. I wouldn't expect the structure to be a non-reentrant type function. And it doesn't, when I put a wait in there as a test.

Okay it looks like you'll get an Out of Memory error on building. https://knowledge.ni.com/KnowledgeArticleDetails?id=kA00Z0000019No1SAE&l=en-US

I have seen others on the forums state that this can be an issue. What are the symptoms of hitting this limit? Should I just set this value on all development machines just in case?

There is an offline help option, but I would much prefer to the more straight forward CHM files. https://forums.ni.com/t5/LabVIEW/Offline-Help-window-in-LabVIEW-2022/m-p/4251163#M1236962 NI's website has many deficiencies that have been well known for years. So the effort to force things to go online, is crazy to me. Maybe trying to drive up network traffic to look better for investors or buyers?

I'll ping the site admin.

Sorry I can't get them to work either. Sometimes LAVA's servers will do weird things with compression or resizing, and in the process it might strip out the meta data that is the VI file. Once we noticed this, these settings were turned off but it is possible these were back when that was an issue.

NI-XNet is newer then NI-CAN. I believe its first release supported LabVIEW 2010 or so. It really is a superior API with much more flexible features. I cover some of the differences in NI-CAN and NI-XNet in Part 3 of my CAN blog. Part 6 goes into XNet in more detail. Newer CAN hardware support either API, but some old stuff does need to use NI-CAN. One thing you didn't mention was if you installed NI-CAN 17 after installing LabVIEW or not. During the install it will detect versions of LabVIEW installed and add support for it. So if you install NI-CAN then LabVIEW, you won't see the supporting palettes. You can just run the NI-CAN installer again and it will add any missing components. This will add Example Finder examples from the Help >> Find Examples menu, and add the NI-CAN palette, I believe it goes under the Measurement I/O on the functions palette, but I don't have it installed at the moment.

More information is needed. Is this a response to an ISO 15765 request? UDS, KWP2000, etc. Or is this periodically broadcasted frames? If it is UDS or something similar you'll need a CDD file to define how it should interpret the response. If it is a periodic message then a DBC file will define the structure of the data being sent, and what IDs belong to what data. I cover the Signal API and how bits can be interpreted in Part 5 of my CAN blog, and I talk about multi frame messaging in Part 8. If you are trying to reverse the CAN bus without this information, then I suspect you will have a very difficult time. There is an IEEE Float Data Type in the XNet Database editor which uses 4 bytes so it is possible this will do what you want. I'd make a database using the XNet Database editor specifying where in the frame the bits are and try using the conversion API with it. EDIT: Also I think Shaun is incorrect. 2 8 D5 0 0 0 0 B 5 70 I'm unsure what the 2 means but 8 means there are 8 bytes to read and the 8 bytes are D5 00 00 00 00 0B 05 70. 8 bytes is a common payload length for CAN. Double Edit: What are your expected values? I ran them through the IEEE conversion and tried Big/Little Endian but ended up with really large numbers both negative and positive.

Write code that detects if the error is thrown, and if it is then close and reconnect. Honestly that shouldn't happen under normal use so I'm wondering if a reference is closed or lost somehow.

I know right? I was looking for a 2D barcode generator not that long ago which is why I mentioned Darin's post that I found, which I presumed you had seen. I searched the forums a bunch and the only other code I found would send it to a website to get generated. I also saw a semi-hacky method using a Zebra printer API, but I never found that NI post either.

Yeah sorry I haven't heard anything about a BBQ, and won't be attending.

I've been using a few AIs in help for writing things. Professional sounding emails of sensitive subjects, work references, and a few other things to help get past a writing block. Image and animation stuff I see cool new things all the time, but haven't done much myself lately.

Oh I get that you're welcome to post it here, I just wanted to link there so anyone here can see the conversation on the subject that has already taken place.

Crosspost

Very cool stuff. I don't know much about 2D barcode encoding, but how does this differ from the code posted by Darin here? Yours looks more polished, but I didn't know if one encoding is better or if both are the same. My users generally don't care at all and just want something that if they scan types what is in the barcode and it seems my QR Code scanning app works with either.

Thanks I figured the information I had was dated, or incomplete.

If the data is accessible by the VI, I tend to stick with an unbundle. I heard that using a property node forces a switch to the UI thread, and it wasn't clear to me if that wasn't the case for class data or not, so I just avoid them if I can. That is one place that NXG did improve on the design of classes and clusters. A cluster was accessed with a property node like it was a class. This made the conversion from a cluster to a class easier, since the methods to access the data were the same.

I posted my Image Manipulation code over on VIPM.IO https://www.vipm.io/package/hooovahh_image_manipulation/ Here is a video demonstrating some of its functionality, including loading an image into a picture box and manipulating it.

Love it, thanks for the shoutout. There is plenty of room for improvement if you ever want to make this into a more polished API. I could see it reading the column width information, and working with the resizing of them. Thanks for sharing, I know I've seen a few requests for things like this in the past.

I just moved it to Code In Development. I haven't tested this but native TIFF is pretty neat. When you feel development is in a good place you should post your package on VIPM.IO. Then it will show up in VIPM. The process for uploading packages is pretty straight forward, and you can link back to the Git for the source. In my Image Manipulation package there is a Load Image function which accepts TIFF. It also allows you to select which image from within the file if there are multiple in there. This all uses .Net image library and is less native, being restricted to Windows.

Crap, it sure looks like it. I did lots of unit testing, but it didn't test for that apparently. Also note that if Indicies Ascending is true then it has the correct result. Version 3.0.1.15 has this fix on VIPM.IO. Thanks a lot for finding issues like this.

I participated in the 2022 Q3 beta. And I see Christina mentioned preparing for the 2023 Public Beta on March 9th of 2023. Which is a bit odd since the mentioned release date of 2023 Q1 was January 2023, with the last update on the document being March 16th of 2023. I never saw a 2023 beta before it was released.

This idea was implemented. https://forums.ni.com/t5/LabVIEW-Idea-Exchange/Same-Height-of-Unbundle-by-Name-Terminal-Local-Variable/idi-p/1298608 Opening existing code will create bends in wires where there weren't any before.

I just went to 2022Q3 64-bit about a month ago with a fairly basic program. USB cDAQ chassis with DAQmx and XNet on Windows 10. It had several classes, reentrant-ness, subpanels and data logging. I haven't seen any weirdness yet. My next step is to dive in upgrading our Linux RT remote controllers and our sequence editor, and report generation. This is a much larger project and one more likely to find issues.