particle systems changed

I noticed that particle systems no longer stop if you put a llParticleSystem([]); I now have ot enter the count to zero to make it work.



also, target omega does not stop if you set all vectors to 0, but if you set it to a teeny number like 0.0000000000000000000001 it will stop..............


Was the particle system thing intensional?

I noticed this as well.

It seems to be unpredictable. Repeating the llParticleSystem([]) sometimes fixes it, which made me think it was packetloss... but setting a "do-nothing" list always works, so it's probably a bug and I'll bug-report it.

I was able to successfully use llParticleSystem([ ]); (note the space between the brackets)

This seems to work reliably for me.

I have this code in two different products. It's pretty simple. Both use the exact same code.
But one works and the other does not.
It's a bit odd how the same code can get different effects.

Have the Lindens even acknowledged this issue?
Or are they still ablivious to the fact because they test it and see a working version.
I haven't seen anything posted in the known issues or anything about a fix yet.

It sure is costing some people a lot of money.

Here's all the code that parses that:

In indra.l:

"]"                                     { count(); return(']'); }
"["                                     { count(); return('['); }

So the only tokens are '[' and ']'. There is no '[]' token.

In indra.y:

list_constant
        : '[' list_entries ']'
        {
                $$ = new LLScriptSAList(gLine, gColumn, $2);
                gAllocationManager->addAllocation($$);
        }
        | '[' ']'
        {
                $$ = new LLScriptSAList(gLine, gColumn, NULL);
                gAllocationManager->addAllocation($$);
        }
        ;

list_entries
        : list_entry
        {
                $$ = $1;
        }
        | list_entry ',' list_entries
        {
                $$ = $1;
                $1->addAssignable($3);
        }
        ;

list_entry
        : simple_assignable_no_list
        {
                $$ = $1;
        }
        ;       

So there's no '[' space ']' sequence.

Consequently, there is no way for the parser to tell "[ ]" from "[]". Period.

I'm curious as to why this would even be changed? What part of the update has anything to do with particals?
And has anyone found a way around it yet?

Ok I did a little test and this is what I found.

Use a simple touch switch to turn on and off particals. For me it seems to work fine if it's left alone as is.

But now make that prim a child in a link set.
Make the parent send a link message to turn on and off particals and put a listener in the child prim that contains the partical script.
Using this code llParticleSystem([]); will no longer work for either the link messages OR the touch switch.
Why?