// Cloned by Eamon Cullimore on 10 Oct 2023 from World "Breadth-first search (Kevin Bacon)" by "Coding Train" project
// Please leave this clone trail here.
// Breadth-first search
// Port from:
// https://github.com/nature-of-code/NOC-S17-2-Intelligence-Learning/tree/master/week1-graphs/P2_six_degrees_kevin_bacon
// https://www.youtube.com/watch?v=piBq7VD0ZSo
// Hard to visualise graph
AB.headerRHS();
AB.newDiv("selectdiv");
$("#selectdiv").css({
"margin": "30",
});
$("#selectdiv").html(" Pick an actor: <span id='actorlist'></span> ");
// Daniel Shiffman
// Nature of Code: Intelligence and Learning
// https://github.com/shiffman/NOC-S17-2-Intelligence-Learning
// Graph object
function Graph()
{
// store nodes by key and also in an array
this.graph = {};
this.nodes = [];
// Start and end
this.start = null;
this.end = null;
}
// Set the start
Graph.prototype.setStart = function(node)
{
this.start = node;
}
// Set the end
Graph.prototype.setEnd = function(node)
{
this.end = node;
}
// Add a node
Graph.prototype.addNode = function(label)
{
var n = new Node(label);
// Add to both the graph object and the nodes array
this.graph[label] = n;
this.nodes.push(n);
return n;
}
// Clear all searched and parent values
Graph.prototype.clear = function()
{
for (var i = 0; i < this.nodes.length; i++)
{
this.nodes[i].searched = false;
this.nodes[i].parent = null;
}
}
// Daniel Shiffman
// Nature of Code: Intelligence and Learning
// https://github.com/shiffman/NOC-S17-2-Intelligence-Learning
// Node object
function Node(label)
{
// Has a label
this.label = label;
// zero or more edges
this.edges = [];
// No parent and not searched by default
this.parent = null;
this.searched = false;
}
// Connect one or more neighbors
Node.prototype.connect = function(neighbor) // one Node (movie) has a connection to another Node (person)
{
// This is a fancy way of having a function
// that can accept a variable number of arguments
for (var i = 0; i < arguments.length; i++)
{
this.edges.push(arguments[i]); // increase the size of the array
// Connect both ways
arguments[i].edges.push(this);
}
}
// Daniel Shiffman
// Nature of Code: Intelligence and Learning
// https://github.com/shiffman/NOC-S17-2-Intelligence-Learning
// The data
var data;
// The graph
var graph;
// A lookup table for actors
// (redundant, the graph could handle this)
var actors;
// Dropdown menu for actors
var actorlist;
function preload()
{
data = loadJSON('/uploads/codingtrain/bacon.json');
}
function setup()
{
noCanvas();
// Create the graph
graph = new Graph();
// A separate lookup table for actors
actors = {};
// For all movies
var movies = data.movies;
for (var i = 0; i < movies.length; i++)
{
// Title and castlist
var movie = movies[i].title;
var cast = movies[i].cast;
// Add the movie to the graph
var movieNode = graph.addNode(movie); // Node movie
// Go through all the cast list
for (var j = 0; j < cast.length; j++)
{
var name = cast[j];
var actorNode;
// Does the actor exist already?
if (actors[name]) actorNode = actors[name];
// If not add a new node
else
{
actorNode = graph.addNode(name); // Node person
actors[name] = actorNode;
}
// Connect movie and actor
movieNode.connect(actorNode); // movie connects to person
}
}
// Create dropdown
actorlist = createSelect(); // https://p5js.org/reference/#/p5/createSelect
actorlist.parent('actorlist');
var allactors = Object.keys(actors);
// Add all the actors
for (var i = 0; i < allactors.length; i++)
actorlist.option(allactors[i]);
// Set up an event
actorlist.changed(findbacon);
}
function findbacon()
{
// Clear everyone from having been searched
graph.clear();
// Start and end
var start = actors[actorlist.value()];
var end = actors['Kevin Bacon'];
graph.setStart(start);
graph.setEnd(end);
// Run the search!
bfs();
}
function bfs()
{
// Create a queue ad path
var queue = [];
var path = [];
// Get started
queue.push(graph.start);
while (queue.length > 0) // inside the loop we keep adding things to queue
{
var node = queue.shift(); // remove first item
console.log ("======" );
console.log ("start of search of " + node.label);
// Are we done?
if (node == graph.end)
{
console.log ("found goal node");
// Figure out the path
path.push(node);
var next = node.parent;
while (next)
{
path.push(next);
next = next.parent;
}
break;
}
else
{
console.log ("checking neighbours of " + node.label);
// Check all neighbors
var next = node.edges;
// console.log ("checking " + next.length + " neighbours");
for (var i = 0; i < next.length; i++)
{
var neighbor = next[i];
// Any neighbor not already searched add to queue
if (!neighbor.searched)
{
console.log ("adding unsearched neighbour " + neighbor.label + " to queue");
queue.push(neighbor);
// Update the parent
neighbor.parent = node;
}
}
// Mark node as searched
node.searched = true;
console.log ("end of search of " + node.label);
// console.log ("======" );
}
} // end of while loop
// now output path
console.log('finished search for path');
// path = bacon + movie + x + movie + x ... + movie + x
var s = '<h1> Path length ' + path.length + '</h1>';
for (var i = path.length-1; i >= 0; i--)
{
s = s + path[i].label + "<br>";
}
AB.newDiv("theoutput");
$("#theoutput").css({
"margin": "30",
"padding": "30",
"display": "inline-block",
"border": "1px solid black"
});
$("#theoutput").html(s);
}