I think that the "Maximum Undo Step by VI"
default value (Tools>Options>Environment) should be changed form 8 to the maximum of 99. For every new
LabVIEW version I install, I end up changing this (from 8 to 99).
I'm not sure why it's even configurable... I know someone once mentioned that the development environment performance might decrease as you get up there in number of undos, but I've never had an issue.
I fought for this exact feature to be implemented in the next LabVIEW release. But I guess 99 is way too scary a number for some people, so keep your eyes peeled for a sorta-kinda-not-really solution in a future LabVIEW release. 🙂
And yes, I think the <99 value is a legacy of the days of wimpier computers.
Frankly, I rarely run into the 8 undo limitations, even though I have never increased the default undo buffer size.
Most often I run into a situation where the undo buffer has been reset and I wish it hadn't. So, increasing the size of the undo history is not very useful if we can actually never fill the buffer because it gets emptied regularly. (One example is the "clear on save" as discussed in another idea).
I am sure there are some other actions that clear the undo buffer because once in a while I try to undo and not even a single step works. Does anyone have a good list of such operations?
Still, I agree that the default number of undo steps should probably be increased.
I addition, the undo history is a bit too course-grained, because certain relatively important editing events don't force an entry into the undo history.
Here's one scenario where undo fails miserably, no matter how many steps are defined: 😞
(0) Open a VI that has a cleary defined small FP size and contains a few controls and indicators outside the visible area (e.g. an xcontrol). (1) Open the BD and double-click the terminal of one of the controls that are outside the visible FP area. (2) The FP automatically pans to bring the selected control into view, shifting everything important outside the frame. (3) Of course we don't like that and press ctrl+z to undo the panning. (4) nothing happens. The shift cannot be undone for some reason.
---> (To undo, we must do a "file...revert". Undesirable because we possibly already made some code changes that we want to keep. Manually panning the FP back is tedious).
altenbach wrote: Frankly, I rarely run into the 8 undo limitations, even though I have never increased the default undo buffer size.
Woa, this is a surprise. When I have to show something on a customer machine that has this setting left at 8, I discover this within the first few min [since I extensively use undo/redo (CTRL+Z, SHIFT+CTRL+Z)].
Note: I even have run into the 99 limit (although extremely infrequently).
altenbach wrote: Well, I learned LabVIEW with version 4.0 where undo did not exists..
Thanksfully, I started with LabVIEW 5.0 so I never had the "pleasure" of the no undo (although at the time I was using a 3D software for rapid prototyping that did not had any undo either, so I can appreciate the "experience").
I just want to say I was a vocal supporter of Darren's campaign to raise the undo limit. It was like watching one of those strongman muscle competitions on TV where the guy is dragging a semitruck and everyone cheers when it rolls a few feet. Tune in next version when Darren will attempt to drag it... gasp... another two feet!