llDetectedGrab in link set attached to HUD?

A while back I finally figured out how llDetectedGrab works in child prims and posted up a freebie.



This shows how to use llDetectedGrab in a link set's root prim to handle dragging both the root prim and child prims. Notice the kicker was you had to multiply the prim's local position by 3 before doing some other math... it all seems fuzzy now.

So what I'm wondering is if anyone has made any strides in figuring out how to make sense of llDetectedGrab in child prims of a link set attached to HUD.

I've posted a simpler and more straightforward example if it will help push research forward: There's a script in the root prim that handles grabs on itself and on the unscripted child prims, and there's a scripted child prim that handles its own drags.



I know you're out there! Let's get together and feel alright!

Just for clarity: Is the question strictly about *child* prims in HUDs? I found the behavior of llDetectedGrab() for HUD root prims unintuitive enough that I begged (and most graciously received) help disentangling specifically how this is a function of cam rotation. I'd just guess that some TBD combination of that magic and your magic would be necessary, but you may be far beyond that point already.

Sorry dude, thanks for the response. I think I did make sense of the HUD root prim behavior but didn't care, haha... I had my mind set on scrollbars and stuff!

We should be able to finally get rid of llDetectedGrab() in HUDs for sliders when the functions of SVC-1092 become available. I'm probably going to hold off deployment of a HUD I'm working on till those arrive, as llDetectedGrab() has too many associated problems.

Link to SVC-1092:
http://jira.secondlife.com/browse/SVC-1902