Yeah, I know that. As the manual says:
Event functions are actually executed during the engine function that caused the event, like a c_move, c_scan, or c_trace instruction of another entity. The event function itself should be simple. It normally should only transfer information to the entities' main function - it shouldn't perform instructions that can trigger events itself, displace entities, start particles or change anything else in the level. Thus instructions like c_move, ent_create, ptr_remove, c_trace etc. must not be performed. If the event function must perform such 'critical instructions', precede them by a wait(1) for delaying them to the next frame. Then it's safe.
I guess that's cause those c_ functions and ent_create as well, toy around with 'you' pointer. Which is also used in events. That's why it requires wait(1) in order to 'seperate' pointers from eachother. But I've done c_traces, c_moves, ent_creates before in events without facing any troubles (using seperate function and wait(1) before does work pretty well, if I would add 'wait' in test function it won't help anyway), so I guess it's not related to that here either. It's probably has to do with the way I used structure, maybe I've done something wrong?
