Hadn't really paid attention to that - thank you for pointing it out!

Well, with a lot of help of ChatGPT (and a ton of patience too figuring out and fixing what the AI was doing wrong) I managed to come up with code that can dynamically build that type of image/icon based on full/free percentage. It's not 100$% perfect but it's pretty similar.

This said, I'm not sure this would be my first choice for one of the disk monitor iconic module styles.

I think that if I were to use either I would probably go with the drive shell.

But of the pie variants I liked this one as it is quick (obvious) which aspect relates to the empty portion. As you say the original view looks more complete / finished, however your variant is pretty close!

Just ran this and it tells me that problems were found - 6 Orphan Stacks and 6 Crosslinked stacks.

Pressing OK to correct the problems caused Workshelf to go into an unresponsive state, which I had to terminate from Task Manager. A restart of Workshelf enabled my shelf to return to normal. However running the Integrity Check again still shows the same problems.

Tried correcting again, and again Workshelf stopped responding. Left it a bit longer this time to see if it eventually sorts itself out - using between 8-10% CPU. After 7 mins terminated it again. Restarted Workshelf OK, back to where I was. Integrity Check still reports same fault!

It's just about the ugliest of an ugly lot IMO, sorry.

Not good, but glad you caught that.

Please make a backup of the settings and send it over to Winstep support.



AHAHAHA! Which is why, as they say, opinions are like arses (everyone has one)

Ok, so the idea would be to use something like this.

The round "button" is actually an activity LED. Having a led blinking with drive activity can actually be turned on or off by the user. If the user chooses NOT to display the activity LED, the idea is that the used capacity bar "grows" to occupy its space, so it becomes more visible at lower icon sizes too.

Also, had forgotten to add the white floor to the pie chart version. Colors can, of course, be defined by the user (except perhaps the floor which would always be white?)

LOLOL! Well, if they didn't we'd all be in serious trouble! As Burns was said to have remarked once when challenged by a cop for vomiting over a bridge, "Well Sir, I'm shitting through my teeth as I've lost the keys to my arse!" - we'd all be doing that literally!

Yeah, the black and silver one (for dark and light) seem like a reasonable choice. (Simple bar graph would have been preferable I still think, but what the heck.)


[Edit] Hey, just realised the 'Dark Theme' thread's gone AWOL! What happened, Jorge? [/Edit]

Everyone has their likes / dislikes.


Sent


I like the look of these (particularly the copper one) - will the colours be configurable?
Of the two black ones side by side, the one on the left looks better - probably due to the lighter graded shading on the top

Since the forum has a fully functioning dark mode now, it didn't make sense to keep that thread as "Sticky". It still exists, but it is now buried somewhere in time eheh

Just for info - deleting the launch pads / stacks / sub-dock from my test workshelf resulted in the Integrity Check showing: 11 Orphan Stacks, 2 Crosslinked Stacks.

Opting to correct these resulted in a message stating the problems had been corrected this time. Running the Integrity Check again still showed - 1 Orphan Stack, 1 Crosslinked Stack.

Opting to correct again gave message that problems had been corrected, but running the Integrity Check a further time resulted in - 6 Orphan Stacks, 3 Crosslinked Stacks

Repeating the process kept changing the reported values - but at least Workshelf didn't lock up.

I did wonder if the reason it didn't lockup these times was due to the fact that I had my Test Workshelf open (which I hadn't during the previous set of tests), but closing it and repeating the test didn't cause lockup - just continued giving varying problem results.

I'm glad you run into this because it allowed me to fix oversights when duplicating grid stacks, etc, which would result in cross links, and also the possibility of infinite recursion:

When a cross link was found, instead of deleting the cross-linked pointer I tried to fix the problem by duplicating the cross-linked grid stack/launch pad (let's call them sub-shelves because that is what they really are)(. Problem was, if this sub-shelf had itself crosslinked items and other issues, it was easy to get into an infinite recursion loop.

What was Going Wrong:

Shelf 17 was being duplicated.

During its duplication, it tried to duplicate sub-shelf 24.

Sub-shelf 24 contains sub-shelf 23, which completes and returns.

Then Shelf 17 continues duplicating sub-shelf 25, 14, 27, 28... all fine.

But then it hits sub-shelf 30, which contains a cross-linked pointer to sub-shelf 17 again — and boom, we're back at the start.

This created a recursive loop: 17 → 30 → 17 → 22 → 19 → ... → 17 → 24 → 23 → ... It never ended because there was no check to prevent re-duplicating a sub-shelf that’s already in the duplication chain.

I fixed it by adding this check.

This cross-links, etc, probably happened as you experimented because of a bug in the Duplicate Shelf routine that failed to copy nested grid stacks and launch pads (!).

So, the Check for Integrity with these fixes finally completed and now reports no erros.

Thing is, despite the check reporting no errors, in truth after the duplications you now have something like 90 Grid Stacks and Launch Pads that are nowhere to be found. Still looking into this.

So, there is nothing you can do there, it has to fixed here with a change to the code itself.

OK, let me know when you think you've fixed it and I will verify.