Topic: On the CS_Assert (n < count)

[Rosencrantz]

Well, after some tracing in gdb I discovered that the reason the assert is popping up is because both n and count are zero. Which fits with the assert, but leaves me feeling uneasy.

1) Count isn't being set right for whatever reason and should be > 0.
2) Count is fine because there's nothing to count, but the CS_ASSERT isn't taking that into account.

3) Something else entirely.

[Morgul]

It's actually #3.

Lynx looked through it, and easily saw that we're getting a CS_Assert because the PcSimpleCamera isn't in a region. Why not? I personally don't remember any of the details (if I knew them) but the CS_Assert is completely valid; our entities are currently the ones in an invalid state. In order to repair this, Lynx needs to make some CEL fixes, and the fix our code accordingly. So unfortunately you and I were barking up the wrong tree with this bug.

It should be fixed in a week... that's assuming everything goes to hell, and Lynx has absolutely no time... which is par for the course currently.

[Rosencrantz]

Excellent! I'm glad we figured out what it was. Lynx, how did you figure that one out? Your comment earlier today about iRegion* makes much more sense now, but I'd love to know how you tracked that one down.

--Rosencrantz

[whitelynx]

I figured it out using my Amazing Powers of Observation^TM.

Actually, it was #1 and #3 at the same time. The array in question was the array of sectors that the camera was currently "in", and it was empty because the way they name regions in CEL changed, and we hadn't fixed our code to take that into account yet. So now it's fixed.