Physics commands in attachments - BUG

This was driving me absolutely yesterday today... I couldn't figure out for the life of me why the collar I make wasn't working... I'm doing a complete rewrite of the entire system of scripts, and I was making changes to the leash script.

The leash script makes several calls to llMoveToTarget. The problem was... that it wasn't working! The attachment was no longer moving the avatar anywhere. So, I got out the old version of my collar, looked at the old leash script, and noticed that the movement parts were called nearly identically. So I saved the script, and behold... now my old collar leash functions no longer worked either!

After about two hours of frantically trying everything I could, I detached the collar and went to sleep. Today, I put the collar on to work with it, and on a whim used the leash function to tie myself to an object... and it magically worked. O.o

I figured maybe I was going crazy the previous day. Lack of sleep. So I looked in the script, couldn't find anything out of the ordinary... and I made the same mistake - I saved it. Tada, non-working again.

So it hit me. I detached the collar and put it back on. I was automatically pulled to the correct target.

After some testing, it appears that all the physics commands that I tested are disabled on an object after recompilation INSIDE an attachment until that attachment is detached and worn again. This bug (or feature? o.O) could drive a person ABSOLUTELY insane if they didn't understand why it was happening. I have no idea when this bug started appearing, but it can't have been too long ago because this was functioning as expected when I was creating that version of the collar (in 1.7, I believe).

Just a heads up for anyone messing around with physics in an attachment... craziness!

My apologies if this is known... I searched the forums, but couldn't find anything about it.

File a bug report. Just make it less wordy. Good job on finding this.

Something like: Physics functions disabled in script after it is compiled, untill the attachment, that the script was compiled in, is rerezzed.
Functions effected: llMoveToTarget

Sure it's a run on sentance but it gets the facts across.
Try to come up with instructions to replicate.

Oh and this is in the wrong forum (Script Tips isn't a bug report forum) so I'm going to move it too...

But it was a tip... to people scripting. O.o

Yeah, this one drove me nuts a few times even though I knew all about it! It's so easy to forget.

And if I were a QA person, your run-on sentence would confuse the heck out of me, strife. Maybe something like this:

1. Wear an attachment with a script inside it that uses llMoveToTarget().
2. Recompile the script.
3. Try to use it, and the llMoveToTarget() won't work.
4. Detach it and then put it back on.
5. Now the llMoveToTarget() works.

Aha!

I think I've seen this with llSetForce() and llApplyImpulse() as well.

I know for a fact I've seen this. It too was driving me nuts.

I even started using MASSIVE impulses to see what the frell was going on... Later on that night, I unquipped the item, re-equipped it and then shot into the sky...

Oh man. X_X

I think I've seen the same problem in flight scripts as well, after a teleport. After the teleport, the physics stops working in an attached object, until you detach and reattach the object.

It's a bug, but not quite the way you think.

Those types of functions use energy, which is a guide to how well they work.

When a script with something like movetotarget is compiled, it depletes the object's energy. The bug here is that the object doesn't regain it's energy as it is supposed to.

I just wanted to bump this thread up to let the developers know this bug still exists in 1.10.2.

I have a similar problem, yet whatever I do, detach, attach ect. there's no way to get llMoveToTarget() to work again in my case ...

/54/08/113596/1.html