| 01-23-2007, 02:21 PM | #1 |
Code:
function ThreadCrush takes nothing returns nothing
local integer I
set I = I + 1
endfunctionwill crush thread this is complete useless, but this is only way to instant stop running thread |
| 01-23-2007, 02:25 PM | #2 |
This bug is a serious pain in the ass. It always takes me ages to figure out the problem when I do this. I know this happens with integers (reals too presumeably) but does it happen with booleans and strings? I've never actually tested it. |
| 01-23-2007, 02:35 PM | #3 | |
This is as old as the hills. Was there really any need to repost it? Quote:
It happens with any type of variable. If an uninitialised variable is used in anything but an assignment call, it will crash the thread. |
| 01-23-2007, 02:42 PM | #4 |
Thanks, I guess I should have known that. |
| 01-23-2007, 03:05 PM | #5 |
What is pretty odd is that arrays don't need initializing, they get default values already and don't crash the thread. This annoying bug is a non-issue when you run grimoire's war3err |
| 01-23-2007, 08:15 PM | #6 | |
Quote:
|
| 01-24-2007, 09:40 AM | #7 | |
Quote:
|
| 01-24-2007, 10:43 AM | #8 |
Yes, there are special exits that the VM follows when those things are going to happen, which is what war3err hooks into. Edit: BTW, grimoire only hooks about half of them, so there should be a few more with unknown purpose... |