On Mar 24, 12:48am, Larry Troxler wrote:

I've had instances of segmentation faults when resizing an array.

I suspect this happens when an array is downsized while a tabread4~ object is running, resulting in newly invalid refrences when the tabread4~ runs its next slice.

I must admit to not not having narrowed this down with test cases, but a quick scan of the code shows no evidence that this situation is not handled. If I'm wrong I'll buy a beer ...

Well, there's probably a bug in there somewhere, otherwise it wouldn't crash (tautology!) but the situation was supposed to be handled by the

if (x->x_usedindsp) canvas_checkdsp();

in array_resize(). Perhaps the check doesn't actually do anything though...

Miller