getSignalDensityAlongAxis and getPopulationDensityAlongAxis assertions

[MrSlinkyman]

I saw in a previous issue where these two assertions were added. I'm curious, however, if the program asserts that the direction is not CENTER, how do you prevent creatures from actually using that direction when these sensors are called?

biosim4/src/getSensor.cpp

Line 22 in 33d6832

assert(dir != Compass::CENTER); // require a defined axis

biosim4/src/getSensor.cpp

Line 120 in 33d6832

assert(dir != Compass::CENTER); // require a defined axis

In the getSensor() call I would imagine that the creatures would have a last move direction as CENTER just by chance.

biosim4/src/getSensor.cpp

Line 311 in 33d6832

sensorVal = getPopulationDensityAlongAxis(loc, lastMoveDir);

In my implementation in Java, these assertions are preventing the program from moving forward so I changed them to early returns where the method just returns 0.5 if the direction supplied is CENTER. But I am not sure if that is the right interpretation.

Would love some insight.

[davidrmiller]

From a quick look, I think that the lastMoveDir member variable will not be CENTER because it is initialized to Dir::random8() (which can't be CENTER) and modified only in one place in Peeps::drainMoveQueue(). The lastMoveDir member is only modified there if the individual is alive and if the newLoc grid location is empty, which prevents it from being set on the CENTER location which is where the individual resides.