makingSTWv_Adj.m

% makingSTWv_Adj
% this function will create all s, t, weights, and v_Adj based off of the disntace and not have an
% edge to every battery level
% INPUTS
% area = is one dimension of the total area where the points are made (row
% x col = total area, it is row)
% x = the original points
% y = the original points that correspond to x
% numPoints = the amount of original points
% numLevels = the amount of levels
% v_Adj = this is the adjacency matrix that gives the costs
% OUTPUTS


function [sNew, tNew, weights, v_AdjNew, allDistances] =  makingSTWv_Adj(area, x, y, numPoints, numLevels, v_Cluster)

maxDistance = sqrt((area(1)-area(2))^2+(area(3)-area(4))^2);
maxDistancePerLevel = maxDistance/numLevels;
tempLevelArray = [];

for i = 1:numLevels
    tempLevelArray(end+1) = i;
end

allDistances = [];
for i = 1:numPoints
    for j = 1:numPoints
        comparingPoints = [x(i), y(i); x(j), y(j)];
        allDistances(i, j) = pdist(comparingPoints, 'euclidean');
    end
end

allDistancesRounded = ceil(allDistances./maxDistancePerLevel);
% outputv_Adj = v_Adj;

totalPoints = numPoints * numLevels;
tempv_AdjFinal = Inf(totalPoints);
tempv_Adj(1:totalPoints, 1:numPoints) = 0;
for i = 1:numPoints
    fillArray = allDistancesRounded(i, :);
    for j = 1:numLevels
        tempv_Adj((i-1)*numLevels+j, 1:numPoints) = fillArray;
    end
end

for i = 1:numPoints
    fillArray = tempv_Adj(:, i);
    for j = 1:numLevels
        tempv_AdjFinal(1:totalPoints, (i-1)*numLevels+j) = fillArray;
    end
end

pointA = [];
pointB = [];
sNew = [];
tNew = [];
weights = [];
matv_Cluster = cell2mat(v_Cluster);
for i = 1:numPoints
    pointA = find(matv_Cluster == i);
    for j = 1:numPoints
        node = find(matv_Cluster == i, 1);
        numOfIterations = allDistancesRounded(i, j);
        % pointB = find(v_Cluster == j);
        weightIJ = allDistances(i, j);
        for k = 1:(numLevels-numOfIterations+1)
            if numOfIterations == 0 || i == j % you don't need both conditions because they do the same thing, but it helps me remember how things work
                break;
            end
            pointB = find(matv_Cluster == j);
            sNew(end+1) = node;
            pointB = circshift(pointB, -(numOfIterations+k-1));
            tNew(end+1) = pointB(1);
            node = node+1;
            weights(end+1) = weightIJ;
        end
    end
end

numOfEdges = numel(sNew);
v_AdjNew = zeros(totalPoints);
for i = 1:numOfEdges
    v_AdjNew(sNew(i), tNew(i)) = weights(i);
end


end