#include	<stdio.h>
#include	"smaze.h"
#define	SmoothMove	0

char	World[VGrids][HGrids] = {
	"       OG",
	"  O    O ",
	"S O    O ",
	"  O      ",
	"     O   ",
	"         "
};
Agent	A;
short	Startx, Starty;
unsigned long	Step, GoalCount;

draw_view(flag)
{
	extern	int	DisplayOn;
	if (DisplayOn) { set_step_label(Step); draw_world(flag); }
	else if (Step % ShowStepInterval == 0) set_step_label(Step);
}
domain_reset()
{
	int	x, y;
	Step = GoalCount = 0;
	for (y = 0; y < VGrids; y ++) for (x = 0; x < HGrids; x ++)
		if (World[y][x] == W_Start) { Startx = x; Starty = y; break; }
	A.x = Startx; A.y = Starty;
	brain_reset();
	draw_view(0);
}
static void new_position(x,y,a,newx,newy)
short	x, y, a, *newx, *newy;
{
	*newx = x; *newy = y;
	switch (a) {
		case ActionUp: if (y > 0) (*newy)--; break;
		case ActionDown: if (y < VGrids-1) (*newy)++; break;
		case ActionLeft: if (x > 0) (*newx)--; break;
		case ActionRight: if (x < HGrids-1) (*newx)++;
	}
}
one_step()
{
	policy();
	new_position(A.x,A.y,A.a,&A.newx,&A.newy);
	A.reward = 0.;
	switch (World[A.newy][A.newx]) {
		case W_Obstacle: A.newx = A.x; A.newy = A.y; break;
		case W_Goal: GoalCount ++; displayMonitor();
		A.reward = 1.; A.newx = Startx; A.newy = Starty;
	}
	learn();
	Step ++; draw_view(SmoothMove);
	A.x = A.newx; A.y = A.newy;
}