LabVIEW Idea Exchange

Community Browser
cancel
Showing results for 
Search instead for 
Did you mean: 
Post an idea

Here's how we currently make cluster element labels appear, and a proposed idea for improvement.

ClusterLabels.png

When using the insert quick-drop option, it works great when wiring something into a single wire.

 

However, when working with multiple wires, it just inserts a new VI on each wire. This could be improved upon to better support the standard LVOOP workflow where you'd want to pass two wires through each VI.

_carl_2-1718229032665.png

I find that this is the scenario where the quick-drop option could be most-useful, however the current implementation really isn't very useful in this scenario at the moment.

 

In the spirit of Ctrl + double-clicking a subVI to open it straight to the block diagram, Ctrl + double-clicking a type def control / constant / indicator should open the type definition. Just a bit quicker and more convenient than right-click -> Open Type Def.

 

ctrl_dbl-click_typedef.png

I sometimes use arrays of classic frameless colorboxes to quickly display a color table.

 

Unfortunately, the minimum size of these elements is 5 pixels :(, making a 256 element color ramp 1280+pixels wide, showing redundant information in 5 adacent pixel columns each.  If we display the color ramp vertically, it would not even fully fit on a common 1080p display!

 

This makes it too wide to correctly display them in the forum and forces a size reduction that makes all other text and diagram code illegible.

(Yes, we can right-click...open image in new tab, and click the image to display it 1:1 (in chrome), but that's more work and not everybody knows that)

 

It would be perfectly OK if we could make these elements only 2 pixels wide (or high), making the array display a much more reasonable 512 pixels and keeping the front panel to a more reasonable size.

 

 

Idea summary:

The minimum size of the classic colorbox (frameless) should be 2x2 pixels (or even 1x1?) Currently it is 5x5, i.e. too big.

 

 

For restructuring a larger application I created a VI that shall recursively replaces all missing VIs and type defs based on a lookup table.
For VIs this works fine but it does not work with missing *.ctl files (type def) for controls or constants because there is no "MissingTypeDef" property available.

I'd like to suggest to add something a "MissingXYName" property to each class type that refers to a file - corresponding to what the VI class is already offering.

In my personal opinion it would be even better to have a file-pathname property (which shall be valid even if the file is missing) plus a second property (enum) that shows if this file is valid or not (0: all fine; 1: there but somehow invalid; 2: missing).

 

Andi_S_0-1583930751196.png        Andi_S_1-1583930838129.png

 

 

I am struggling (yet again) with LabVIEW installation problems that appear to involve LabVIEW 2017 (and possibly LabVIEW 2016 f5 patches).  After having systems with multiple (sometimes only 2) LabVIEW Versions installed "go south" (typically by having MAX stop working and Block Diagrams with DAQmx code fail to load), I've tried to "Remove All" NI software, only to discover that "bits and pieces" still remain, both on Disk and (especially) scattered throughout the Registry.

 

I've been working with NI Support for 2-3 weeks trying to "recover" from a LabVIEW corruption probably caused by installing the 2016 f5 patch.  We finally decided to do the "Uninstall/Reinstall" route.  Although I got 2012 SP1 installed, 2014 SP1 failed (could not install NI Network Discovery 14.0, "Verify you have sufficient privileges to install Services").

 

My concern is that, short of reformatting my hard drive and reinstalling Windows (which I was forced to do on two of my PCs), there appears to be no way to fully uninstall all NI Software.  I would like to propose that NI develop an "Eraser" utility (like Eraser for Microsoft Office) that searches out all files that NI puts on the C: drive (not in User Space) during installation and all Registry entries that it scatters throughout the Registry, allowing the PC to be "rolled back" to a "pre-NI" state.  Such a tool might want to be restricted to Full or Professional licenses, or maybe provided on an "As Needed" basis by the NI Support Team, but I really don't want to have to rebuild yet a third PC ...

 

Bob Schor

The Logical Shift, Rotate, Boolean Array to number and Number to Boolean Array primatives should work with 64 bit values. Currently they are restricted to 32 bit values.

Have you ever auto indexed out of a while loop but had to use "Delete From Array" because you didn't want the last element?  Sometimes I don't want the element from the iteration with the stop condition (behavior similar to a true "while loop" rather than LabVIEW's "do while".

 

Delete Last Element.png

 

I'd like a right-click option on the autoindex terminal to not include the last element.

 

Auto Delete Last Element.png

 

Smiley Wink Only single click to concatinate 1D array out from For Loop:

Concatenat For Loop output tunnel.png

 

 Usually we can make it like below Smiley Sad.

 

But which easier, faster and most elegant Smiley Happy.

 

Concatenat 1D Array via shift register + Build_Array in For Loop.png

If you configure a plot to use both left and right axes, the little dash marks can be mistaken for negative signs on the right axis:

 

plots.PNG

 

In the above example, the right axis ranges from +80deg to +120deg. But it sure doesn't look like it!

 

Can we have an option to turn off those dashes? Or change the symbol?

I spend time making the size of my front panel and block diagrams fit my code well.  Unfortunately everyone once in a while I will hit ctrl+T when I'm trying to press ctrl+R to run my code, and it of course changes the dimensions of my windows.  I try to undo the operation, but it does not work properly (or as I would want it to).  It moves it back to its position, but the dimensions are all wrong still.  Very frustrating.

 

I propose that Undo reverts the panels to their pre-tiling size.

An Image worth a thousand words 🙂

 

ErrorRingImprovement.png

 

Thank you

 

Dany

 

Please provide option to select display style in string probing.

mrk_ramesh_0-1614662185638.png

 

Imediately after posting this I found the "find all instances" for a typedef sort of right there in front of me. Smiley Embarassed

 

But finding all instances returns the typedef used as a control. It would be useful to be able to find all instances of a typedef constant from the right click menu.

 

 

Auto wire is very convenient most of the time. Sometimes, however, something like this happens:

To avoid that, you either need to place the second node very far away and move it back, or manually delete the wires, which can be a pain. Especially considering you see the wires it's going to add while you're dragging it, but have no way to tell LabVIEW you don't want them other than deleting them manually. It would be nice to be able to hold a key down to temporarily disable this feature.

 

But what if you don't notice it's going to place a wire until you've already placed it? There should be a key for that too. If there haven't been any edits since the last time something was auto-wired, you could just press a key and the last auto-wire would go away. A good way to do this would be to just have the auto-wiring count as a separate action in the undo buffer, so you can hit Ctrl+Z once to get rid of the wires, and again if you don't want the object you placed at all.

I remember this in previous (pre-8.2) versions of LabVIEW - not sure why it was removed.  I have a use case to use projects as templates (like when someone wants to write a plugin for a utility I've written, I want to be able to send them a zip containing a project, methods, etc).  The project includes installer settings (so their files go into the right place under my util's plugins folder, but when they build and try to install their plugin, they get an error if another plugin bult using the same template has already been installed.  This is because the "Upgrade Code" (stored in the lvproj file) is the same (it tells Windows that the two products are the same, so subsequent installs are seen as upgrades or replacements, not new installs.

 

Upgrade Code.gif

 

My memory tells me that I used to be able to hit a "Generate" button somewhere which would give me build a new code - all I'm asking for is that back (I can add a step in my work instruction to hit that button before you build).

 

I don't currently have a workaround for this (other than having engineers manually edit the lvproj file) - if anyone has a better idea, I'd love to hear it for the interim!

While browsing through Property Nodes, I would like a visual clue that a node is Scripting. It's great that the nodes are light blue, but only once they are placed on the BD do we know that they belong to Scripting.

 

Ideas might include:

1) Glyphs next to each node in the list, but I am concerned how these adornments might slow down the pop-ups.

2) Ordering the Scripting nodes at the bottom of the list, below a line that reads "Scripting Nodes".

3) Easiest: Indicate that a node is for Scripting in the Context Help, as shown below.

 

 

 

ScripNoce.png 

Remark: This example is using the numerical-double-data-type but it is valid for all other data-types, too. I'm using LabView 2011SP1 (german)

 

Creating a reference from a numerical-control is not new and very easy to do via the context menu. But I'm missing a similar function to create a double-control from a double-reference / double-reference-control.

This would be very helpful especially when there is a VI that requires a reference to a special data type as input and you need to create a control of exactly this datatype.

 

As already suggested in an other idea here in this board it would also be helpful to have a context menu entry that allows to create a reference control from a numerical control without make the detour with a reference in the block diagram.

 

references.png

Something that I think got lost in the newer versions:

You copied a code snipped , switch to your block diagram and paste the code. 

Currently (8.6.1) the code will show up highlighted in the center of the screen and you have to move it where you think it belongs.

 

I want to (right) click on a position in a diagram, hit Ctrl-V  and the code show up at the click position.

 

 

I think it was there in the past, however I didn't found something in the options to turn that back on.

I would often like to directly compare numbers within an array control stretched vertically and with fixed-width font and have tried in vain in the past to include the sign of the number automatically so that alignment is not lost between psotive and negative numbers.

 

I'd love to be able to display the sign always (+ and -) so that I can maintain constant number width in my displays.

 

nonuniform number formatting.png

 

The digits under the long fgreen line SEEM to be the same decimal palce, but they are not.  The shift due to the sign is hard to spot and makes a cursory glance insufficient to distinguish numbers from each other.

 

I know I can get this done using minimum field width and padding with spaces but why can't I just show the sign?

 

That's all.

 

Shane.