static void Main(string[] args)
{
Console.WriteLine("You are runnning the Algorithms tutorial.");
Console.WriteLine("======================================================");
Console.WriteLine();
#region Sorting
{
Console.WriteLine(" Sorting Algorithms----------------------");
Console.WriteLine();
int[] dataSet = new int[] { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
Console.Write(" Data Set:");
Console.Write(dataSet[0]);
for (int i = 1; i < dataSet.Length; i++)
{
Console.Write(", " + dataSet[i]);
}
Console.WriteLine();
// if you want to sort non-array types, see the overloads using Get<int> and Assign<int>
// Delegates
//Get<int> get = (int index) => { return dataSet[index]; };
//Assign<int> assign = (int index, int value) => { dataSet[index] = value; };
// Shuffling (Randomizing)
Sort <int> .Shuffle(dataSet);
Console.Write(" Shuffle (Randomizing): ");
Console.Write(dataSet[0]);
for (int i = 1; i < dataSet.Length; i++)
{
Console.Write(", " + dataSet[i]);
}
Console.WriteLine();
// Bubble
Sort <int> .Bubble(dataSet);
Console.Write(" Bubble: ");
Console.Write(dataSet[0]);
for (int i = 1; i < dataSet.Length; i++)
{
Console.Write(", " + dataSet[i]);
}
Console.WriteLine();
Console.WriteLine(" shuffling dataSet...");
Sort <int> .Shuffle(dataSet);
// Selection
Sort <int> .Selection(dataSet);
Console.Write(" Selection: ");
Console.Write(dataSet[0]);
for (int i = 1; i < dataSet.Length; i++)
{
Console.Write(", " + dataSet[i]);
}
Console.WriteLine();
Console.WriteLine(" shuffling dataSet...");
Sort <int> .Shuffle(dataSet);
// Insertion
Sort <int> .Insertion(dataSet);
Console.Write(" Insertion: ");
Console.Write(dataSet[0]);
for (int i = 1; i < dataSet.Length; i++)
{
Console.Write(", " + dataSet[i]);
}
Console.WriteLine();
Console.WriteLine(" shuffling dataSet...");
Sort <int> .Shuffle(dataSet);
// Quick
Sort <int> .Quick(dataSet);
Console.Write(" Quick: ");
Console.Write(dataSet[0]);
for (int i = 1; i < dataSet.Length; i++)
{
Console.Write(", " + dataSet[i]);
}
Console.WriteLine();
Console.WriteLine(" shuffling dataSet...");
Sort <int> .Shuffle(dataSet);
// Merge
Sort <int> .Merge(Compute <int> .Compare, dataSet);
Console.Write(" Merge: ");
Console.Write(dataSet[0]);
for (int i = 1; i < dataSet.Length; i++)
{
Console.Write(", " + dataSet[i]);
}
Console.WriteLine();
Console.WriteLine(" shuffling dataSet...");
Sort <int> .Shuffle(dataSet);
// Heap
Sort <int> .Heap(Compute <int> .Compare, dataSet);
Console.Write(" Heap: ");
Console.Write(dataSet[0]);
for (int i = 1; i < dataSet.Length; i++)
{
Console.Write(", " + dataSet[i]);
}
Console.WriteLine();
Console.WriteLine(" shuffling dataSet...");
Sort <int> .Shuffle(dataSet);
// OddEven
Sort <int> .OddEven(Compute <int> .Compare, dataSet);
Console.Write(" OddEven: ");
Console.Write(dataSet[0]);
for (int i = 1; i < dataSet.Length; i++)
{
Console.Write(", " + dataSet[i]);
}
Console.WriteLine();
//Sort<int>.Shuffle(get, set, 0, dataSet.Length);
//// Slow
//Sort<int>.Slow(Logic.compare, get, set, 0, dataSet.Length);
//Console.Write("Slow: ");
//Console.Write(dataSet[0]);
//for (int i = 1; i < dataSet.Length; i++)
// Console.Write(", " + dataSet[i]);
//Console.WriteLine();
Sort <int> .Shuffle(dataSet);
// Bogo
//Sort<int>.Bogo(Logic.compare, get, set, 0, dataSet.Length);
Console.Write(" Bogo: Enable (uncomment) in Source Code");
//Console.Write(dataSet[0]);
//for (int i = 1; i < dataSet.Length; i++)
// Console.Write(", " + dataSet[i]);
//Console.WriteLine();
Console.WriteLine();
Console.WriteLine();
}
#endregion
#region Graph Search
{
Console.WriteLine(" Graph Searching----------------------");
Console.WriteLine();
// make a graph
Graph <int> graph = new GraphSetOmnitree <int>(
Compare.Default,
Hash.Default);
// add nodes
graph.Add(0);
graph.Add(1);
graph.Add(2);
graph.Add(3);
// add edges
graph.Add(0, 1);
graph.Add(0, 2);
graph.Add(1, 3);
graph.Add(2, 3);
//// represent a graph
//// Note: can be any type (doesn't have to be int?[,])
//int?[,] adjacencyMatrix =
//{
// { null, 1, 2, null },
// { null, null, null, 5 },
// { null, null, null, 1 },
// { null, null, null, null }
//};
// make a delegate for finding neighbor nodes
Action <int, Step <int> > neighbors =
(int current, Step <int> step_function) =>
{
//for (int i = 0; i < 4; i++)
// if (adjacencyMatrix[current, i] != null)
// step(i);
graph.Neighbors(current, step_function);
};
// make a delegate for computing heuristics
Func <int, int> heuristic =
(int node) =>
{
switch (node)
{
case 0:
return(3);
case 1:
return(6);
case 2:
return(1);
case 3:
return(0);
default:
throw new NotImplementedException();
}
};
// make a delegate for computing costs
Func <int, int, int> cost =
(int from, int to) =>
{
if (from == 0 && to == 1)
{
return(1);
}
if (from == 0 && to == 2)
{
return(2);
}
if (from == 1 && to == 3)
{
return(5);
}
if (from == 2 && to == 3)
{
return(1);
}
if (from == 0 && to == 3)
{
return(99);
}
throw new Exception("invalid path cost computation");
};
// make a delegate for determining if the goal is reached
Func <int, bool> goal =
(int node) =>
{
if (node == 3)
{
return(true);
}
else
{
return(false);
}
};
// run A* the algorithm
Stepper <int> aStar_path = Search <int> .Graph <int> .Astar(
0,
graph,
new Search <int> .Graph <int> .Heuristic(heuristic),
new Search <int> .Graph <int> .Cost(cost),
new Search <int> .Graph <int> .Goal(goal));
// run the Greedy algorithm
Stepper <int> greedy_path = Search <int> .Graph <int> .Greedy(
0,
graph,
new Search <int> .Graph <int> .Heuristic(heuristic),
new Search <int> .Graph <int> .Goal(goal));
Console.Write(" A* Path: ");
if (aStar_path != null)
{
aStar_path((int i) => { System.Console.Write(i + " "); });
}
else
{
Console.Write(" none");
}
Console.WriteLine();
Console.Write(" Greedy Path: ");
if (greedy_path != null)
{
greedy_path((int i) => { System.Console.Write(i + " "); });
}
else
{
Console.Write(" none");
}
Console.WriteLine();
Console.WriteLine();
}
#endregion
#region Graph Search (Vector Game-Style Example)
// Lets say you are coding enemy AI and you want the AI to find a path towards the player
// in order to attack them. Here are their starting positions:
Vector <float> enemy_location = new Vector <float>(-100, 0, -50);
Vector <float> player_location = new Vector <float>(200, 0, -50);
float enemy_attack_range = 2; // enemy has a melee attack with 2 range
// Lets say most of the terrain is open, but there is a big rock in between them that they
// must go around.
Vector <float> rock_location = new Vector <float>(15, 0, -40);
float rock_radius = 20;
// So, we just need to validate movement locations (make sure the path finding algorithm
// ignores locations inside the rock)
Func <Vector <float>, bool> validateMovementLocation = location =>
{
if ((location - rock_location).Magnitude() <= rock_radius)
{
return(false); // inside rock (not valid)
}
return(true); // not inside rock (valid)
// NOTE:
// This function will normally be handled by your physics engine if you are running one.
};
// Make sure we don't re-use locations (must be wiped after running the algorithm)
Set <Vector <float> > already_used = new SetHashList <Vector <float> >();
// Now we need the neighbor function (getting the neighbors of the current location).
Search <Vector <float> > .Graph <float> .Neighbors neighborFunction = (currentLocation, neighbors) =>
{
// lets make a simple neighbor function that returns 4 locations (directly north, south, east, and west)
// and the distance of each node in the graph will be 1
Vector <float>
north = new Vector <float>(currentLocation.X + 1, currentLocation.Y, currentLocation.Z),
east = new Vector <float>(currentLocation.X, currentLocation.Y, currentLocation.Z + 1),
south = new Vector <float>(currentLocation.X - 1, currentLocation.Y, currentLocation.Z),
west = new Vector <float>(currentLocation.X, currentLocation.Y, currentLocation.Z - 1);
// validate the locations (not inside the rock) and make sure we have not already traversed the location
if (validateMovementLocation(north) && !already_used.Contains(north))
{
already_used.Add(north); // mark for usage so we do not use this location again
neighbors(north);
}
if (validateMovementLocation(east) && !already_used.Contains(east))
{
already_used.Add(east); // mark for usage so we do not use this location again
neighbors(east);
}
if (validateMovementLocation(south) && !already_used.Contains(south))
{
already_used.Add(south); // mark for usage so we do not use this location again
neighbors(south);
}
if (validateMovementLocation(west) && !already_used.Contains(west))
{
already_used.Add(west); // mark for usage so we do not use this location again
neighbors(west);
}
// NOTES:
// - This neighbor function has a 90 degree per-node resolution (360 / 4 [north/south/east/west] = 90).
// - This neighbor function has a 1 unit per-node resolution, because we went 1 unit in each direction.
// RECOMMENDATIONS:
// - If the path finding is failing, you may need to increase the resolution.
// - If the algorithm is running too slow, you may need to reduce the resolution.
};
// Now we need the heuristic function (how close are we to the goal).
Search <Vector <float> > .Graph <float> .Heuristic heuristicFunction = currentLocation =>
{
// The goal is the player's location, so we just need our distance from the player.
return((currentLocation - player_location).Magnitude());
};
// Lets say there is a lot of mud around the rock, and the mud makes our player move at half their normal speed.
// Our path finding needs to find the fastest route to the player, whether it be through the mud or not.
Vector <float> mud_location = new Vector <float>(15, 0, -70);
float mud_radius = 30;
// Now we need the cost function
Search <Vector <float> > .Graph <float> .Cost costFunction = (location1, location2) =>
{
// If either locations are in the mud, lets increase the cost of moving to that spot.
if ((location1 - mud_location).Magnitude() <= mud_radius)
{
return(2);
}
if ((location2 - mud_location).Magnitude() <= mud_radius)
{
return(2);
}
// neither location is in the mud, it is just a standard movement at normal speed.
return(1);
};
// Now we need a goal function
Search <Vector <float> > .Graph <float> .Goal goalFunction = currentLocation =>
{
// if the player is within the enemy's attack range WE FOUND A PATH! :)
if ((currentLocation - player_location).Magnitude() <= enemy_attack_range)
{
return(true);
}
// the enemy is not yet within attack range
return(false);
};
// We have all the necessary parameters. Run the pathfinding algorithms!
Stepper <Vector <float> > aStarPath = Search <Vector <float> > .Graph <float> .Astar(
enemy_location,
neighborFunction,
heuristicFunction,
costFunction,
goalFunction);
// NOTE:
// if the "Astar" function returns "null" there is no valid path. (in this example there
// are valid paths, so I didn't add a nul check)
// Here is the path converted to an array (easier to read while debugging)
Vector <float>[] aStarPath_array = aStarPath.ToArray();
// flush the duplicate locations checker before running the Greedy algorithm
already_used.Clear();
Stepper <Vector <float> > greedyPath = Search <Vector <float> > .Graph <float> .Greedy(
enemy_location,
neighborFunction,
heuristicFunction,
goalFunction);
// Here is the path converted to an array (easier to read while debugging)
Vector <float>[] greedyPath_array = greedyPath.ToArray();
// lets calculate the movement cost of each path
float total_cost_astar = Compute <float> .Add(step =>
{
for (int i = 0; i < aStarPath_array.Length - 1; i++)
{
float cost = costFunction(aStarPath_array[i], aStarPath_array[i + 1]);
step(cost);
}
});
float total_cost_greedy = Compute <float> .Add(step =>
{
for (int i = 0; i < greedyPath_array.Length - 1; i++)
{
float cost = costFunction(greedyPath_array[i], greedyPath_array[i + 1]);
step(cost);
}
});
// Notice that that the A* algorithm produces a less costly path than the Greedy,
// meaning that it is faster. The Greedy path went through the mud, but the A* path
// took the longer route around the other side of the rock, which ended up being faster
// than running through the mud.
#endregion
#region Random Generation
Console.WriteLine(" Random Generation---------------------");
Console.WriteLine();
int iterationsperrandom = 3;
Action <Random> testrandom = (Random random) =>
{
for (int i = 0; i < iterationsperrandom; i++)
{
Console.WriteLine(" " + i + ": " + random.Next());
}
Console.WriteLine();
};
Arbitrary mcg_2pow59_13pow13 = new Arbitrary.Algorithms.MultiplicativeCongruent_A();
Console.WriteLine(" mcg_2pow59_13pow13 randoms:");
testrandom(mcg_2pow59_13pow13);
Arbitrary mcg_2pow31m1_1132489760 = new Arbitrary.Algorithms.MultiplicativeCongruent_B();
Console.WriteLine(" mcg_2pow31m1_1132489760 randoms:");
testrandom(mcg_2pow31m1_1132489760);
Arbitrary mersenneTwister = new Arbitrary.Algorithms.MersenneTwister();
Console.WriteLine(" mersenneTwister randoms:");
testrandom(mersenneTwister);
Arbitrary cmr32_c2_o3 = new Arbitrary.Algorithms.CombinedMultipleRecursive();
Console.WriteLine(" mersenneTwister randoms:");
testrandom(cmr32_c2_o3);
Arbitrary wh1982cmcg = new Arbitrary.Algorithms.WichmannHills1982();
Console.WriteLine(" mersenneTwister randoms:");
testrandom(wh1982cmcg);
Arbitrary wh2006cmcg = new Arbitrary.Algorithms.WichmannHills2006();
Console.WriteLine(" mersenneTwister randoms:");
testrandom(wh2006cmcg);
Arbitrary mwcxorsg = new Arbitrary.Algorithms.MultiplyWithCarryXorshift();
Console.WriteLine(" mwcxorsg randoms:");
testrandom(mwcxorsg);
#endregion
Console.WriteLine();
Console.WriteLine("============================================");
Console.WriteLine("Example Complete...");
Console.ReadLine();
}
