I'm trying to implement a remove method for my binary heap implementation.
class Node {
constructor(priority) {
this.priority = priority;
}
}
class PriorityQueue {
constructor() {
this.heap = [null];
}
remove() {
const toRemove = this.heap[1];
this.heap[1] = this.heap.pop();
let currentIdx = 1;
let [left, right] = [2*currentIdx, 2*currentIdx + 1];
let currentChildIdx = this.heap[right] && this.heap[right].priority >= this.heap[left].priority ? right : left; //Assess which child node has higher priority
while (this.heap[currentChildIdx] && this.heap[currentIdx].priority <= this.heap[currentChildIdx].priority) {
let currentNode = this.heap[currentIdx]
let currentChildNode = this.heap[currentChildIdx];
this.heap[currentChildIdx] = currentNode;
this.heap[currentIdx] = currentChildNode;
currentIdx = this.heap.indexOf(currentNode);
}
return toRemove;
}
}
However, I'm not sure how to properly update the value of currentIdx and currentChildIdx when I'm running my while loop. In fact, the code seems to stop working when I try to update the value of currentIdx
currentIdx = this.heap.indexOf(currentNode);
Any tips on what I'm doing wrong?
Full code here: https://repl.it/#Stylebender/Binary-Heap-Naive
In the loop, once you swap the values for currentIdx and currentChildIdx, then you should assign currentIdx = currentChildIdx.
And once you change currentIdx, you need to recompute the left and right child indexes and a new currentChildIdx.
The basic idea is:
while currentIdx < heap_length
currentChildIdx = index of largest child
if (heap[currentIdx] >= heap[currentChildIdx])
break; // node is now in the right place
swap(heapCurrentIdx, heap[currentChildIdx)
currentIdx = currentChildIdx
My suggestion is that you build that basic structure, and then single-step it in the debugger to make sure it's working as expected.
Related
I am working on a compiler and one aspect currently is how to wait for interpolated variable names to be resolved. So I am wondering how to take a nested interpolated variable string and build some sort of simple data model/schema for unwrapping the evaluated string so to speak. Let me demonstrate.
Say we have a string like this:
foo{a{x}-{y}}-{baz{one}-{two}}-foo{c}
That has 1, 2, and 3 levels of nested interpolations in it. So essentially it should resolve something like this:
wait for x, y, one, two, and c to resolve.
when both x and y resolve, then resolve a{x}-{y} immediately.
when both one and two resolve, resolve baz{one}-{two}.
when a{x}-{y}, baz{one}-{two}, and c all resolve, then finally resolve the whole expression.
I am shaky on my understanding of the logic flow for handling something like this, wondering if you could help solidify/clarify the general algorithm (high level pseudocode or something like that). Mainly just looking for how I would structure the data model and algorithm so as to progressively evaluate when the pieces are ready.
I'm starting out trying and it's not clear what to do next:
{
dependencies: [
{
path: [x]
},
{
path: [y]
}
],
parent: {
dependency: a{x}-{y} // interpolated term
parent: {
dependencies: [
{
}
]
}
}
}
Some sort of tree is probably necessary, but I am having trouble figuring out what it might look like, wondering if you could shed some light on that with some pseudocode (or JavaScript even).
watch the leaf nodes at first
then, when the children of a node are completed, propagate upward to resolving the next parent node. This would mean once x and y are done, it could resolve a{x}-{y}, but then wait until the other nodes are ready before doing the final top-level evaluation.
You can just simulate it by sending "events" to the system theoretically, like:
ready('y')
ready('c')
ready('x')
ready('a{x}-{y}')
function ready(variable) {
if ()
}
...actually that may not work, not sure how to handle the interpolated nodes in a hacky way like that. But even a high level description of how to solve this would be helpful.
export type SiteDependencyObserverParentType = {
observer: SiteDependencyObserverType
remaining: number
}
export type SiteDependencyObserverType = {
children: Array<SiteDependencyObserverType>
node: LinkNodeType
parent?: SiteDependencyObserverParentType
path: Array<string>
}
(What I'm currently thinking, some TypeScript)
Here is an approach in JavaScript:
Parse the input string to create a Node instance for each {} term, and create parent-child dependencies between the nodes.
Collect the leaf Nodes of this tree as the tree is being constructed: group these leaf nodes by their identifier. Note that the same identifier could occur multiple times in the input string, leading to multiple Nodes. If a variable x is resolved, then all Nodes with that name (the group) will be resolved.
Each node has a resolve method to set its final value
Each node has a notify method that any of its child nodes can call in order to notify it that the child has been resolved with a value. This may (or may not yet) lead to a cascading call of resolve.
In a demo, a timer is set up that at every tick will resolve a randomly picked variable to some number
I think that in your example, foo, and a might be functions that need to be called, but I didn't elaborate on that, and just considered them as literal text that does not need further treatment. It should not be difficult to extend the algorithm with such function-calling features.
class Node {
constructor(parent) {
this.source = ""; // The slice of the input string that maps to this node
this.texts = []; // Literal text that's not part of interpolation
this.children = []; // Node instances corresponding to interpolation
this.parent = parent; // Link to parent that should get notified when this node resolves
this.value = undefined; // Not yet resolved
}
isResolved() {
return this.value !== undefined;
}
resolve(value) {
if (this.isResolved()) return; // A node is not allowed to resolve twice: ignore
console.log(`Resolving "${this.source}" to "${value}"`);
this.value = value;
if (this.parent) this.parent.notify();
}
notify() {
// Check if all dependencies have been resolved
let value = "";
for (let i = 0; i < this.children.length; i++) {
const child = this.children[i];
if (!child.isResolved()) { // Not ready yet
console.log(`"${this.source}" is getting notified, but not all dependecies are ready yet`);
return;
}
value += this.texts[i] + child.value;
}
console.log(`"${this.source}" is getting notified, and all dependecies are ready:`);
this.resolve(value + this.texts.at(-1));
}
}
function makeTree(s) {
const leaves = {}; // nodes keyed by atomic names (like "x" "y" in the example)
const tokens = s.split(/([{}])/);
let i = 0; // Index in s
function dfs(parent=null) {
const node = new Node(parent);
const start = i;
while (tokens.length) {
const token = tokens.shift();
i += token.length;
if (token == "}") break;
if (token == "{") {
node.children.push(dfs(node));
} else {
node.texts.push(token);
}
}
node.source = s.slice(start, i - (tokens.length ? 1 : 0));
if (node.children.length == 0) { // It's a leaf
const label = node.texts[0];
leaves[label] ??= []; // Define as empty array if not yet defined
leaves[label].push(node);
}
return node;
}
dfs();
return leaves;
}
// ------------------- DEMO --------------------
let s = "foo{a{x}-{y}}-{baz{one}-{two}}-foo{c}";
const leaves = makeTree(s);
// Create a random order in which to resolve the atomic variables:
function shuffle(array) {
for (var i = array.length - 1; i > 0; i--) {
var j = Math.floor(Math.random() * (i + 1));
[array[j], array[i]] = [array[i], array[j]];
}
return array;
}
const names = shuffle(Object.keys(leaves));
// Use a timer to resolve the variables one by one in the given random order
let index = 0;
function resolveRandomVariable() {
if (index >= names.length) return; // all done
console.log("\n---------------- timer tick --------------");
const name = names[index++];
console.log(`Variable ${name} gets a value: "${index}". Calling resolve() on the connected node instance(s):`);
for (const node of leaves[name]) node.resolve(index);
setTimeout(resolveRandomVariable, 1000);
}
setTimeout(resolveRandomVariable, 1000);
your idea of building a dependency tree it's really likeable.
Anyway I tryed to find a solution as simplest possible.
Even if it already works, there are many optimizations possible, take this just as proof of concept.
The background idea it's produce a List of Strings which you can read in order where each element it's what you need to solve progressively. Each element might be mandatory to solve something that come next in the List, hence for the overall expression. Once you solved all the chunks you have all pieces to solve your original expression.
It's written in Java, I hope it's understandable.
import java.util.ArrayList;
import java.util.Collections;
import java.util.List;
import java.util.Objects;
public class StackOverflow {
public static void main(String[] args) {
String exp = "foo{a{x}-{y}}-{baz{one}-{two}}-foo{c}";
List<String> chunks = expToChunks(exp);
//it just reverse the order of the list
Collections.reverse(chunks);
System.out.println(chunks);
//output -> [c, two, one, baz{one}-{two}, y, x, a{x}-{y}]
}
public static List<String> expToChunks(String exp) {
List<String> chunks = new ArrayList<>();
//this first piece just find the first inner open parenthesys and its relative close parenthesys
int begin = exp.indexOf("{") + 1;
int numberOfParenthesys = 1;
int end = -1;
for(int i = begin; i < exp.length(); i++) {
char c = exp.charAt(i);
if (c == '{') numberOfParenthesys ++;
if (c == '}') numberOfParenthesys --;
if (numberOfParenthesys == 0) {
end = i;
break;
}
}
//this if put an end to recursive calls
if(begin > 0 && begin < exp.length() && end > 0) {
//add the chunk to the final list
String substring = exp.substring(begin, end);
chunks.add(substring);
//remove from the starting expression the already considered chunk
String newExp = exp.replace("{" + substring + "}", "");
//recursive call for inner element on the chunk found
chunks.addAll(Objects.requireNonNull(expToChunks(substring)));
//calculate other chunks on the remained expression
chunks.addAll(Objects.requireNonNull(expToChunks(newExp)));
}
return chunks;
}
}
Some details on the code:
The following piece find the begin and the end index of the first outer chunk of expression. The background idea is: in a valid expression the number of open parenthesys must be equal to the number of closing parenthesys. The count of open(+1) and close(-1) parenthesys can't ever be negative.
So using that simple loop once I find the count of parenthesys to be 0, I also found the first chunk of the expression.
int begin = exp.indexOf("{") + 1;
int numberOfParenthesys = 1;
int end = -1;
for(int i = begin; i < exp.length(); i++) {
char c = exp.charAt(i);
if (c == '{') numberOfParenthesys ++;
if (c == '}') numberOfParenthesys --;
if (numberOfParenthesys == 0) {
end = i;
break;
}
}
The if condition provide validation on the begin and end indexes and stop the recursive call when no more chunks can be found on the remained expression.
if(begin > 0 && begin < exp.length() && end > 0) {
...
}
Following code works for sorting of the list (Peter,10) (John,32) (Mary,50) (Carol,31)
Ordered lists:
List 1: (Carol,31) (Carol,31) (John,32) (Mary,50)
however the peter is lost and carol is getting repeated, please help to suggest where Iam going wrong. WHat do I need to change in the loop to get this correct
LinkedList& LinkedList::order()
{
int swapped;
Node *temp;
Node *lptr = NULL;
temp=head;
// Checking for empty list
do
{
swapped = 0 ;
current = head;
while (current->get_next() != lptr)
{
if (current->get_data() > current->get_next()->get_data())
{
temp->set_Node(current->get_data());
current->set_Node(current->get_next()->get_data());
current->get_next()->set_Node(temp->get_data());
swapped = 1;
}
current = current->get_next();
}
lptr = current;
}
while (swapped);
return *this;
}
I have a List of a custom CallRecord objects
public class CallRecord {
private String callId;
private String aNum;
private String bNum;
private int seqNum;
private byte causeForOutput;
private int duration;
private RecordType recordType;
.
.
.
}
There are two logical conditions and the output of each is:
Highest seqNum, sum(duration)
Highest seqNum, sum(duration), highest causeForOutput
As per my understanding, Stream.max(), Collectors.summarizingInt() and so on will either require several iterations for the above result. I also came across a thread suggesting custom collector but I am unsure.
Below is the simple, pre-Java 8 code that is serving the purpose:
if (...) {
for (CallRecord currentRecord : completeCallRecords) {
highestSeqNum = currentRecord.getSeqNum() > highestSeqNum ? currentRecord.getSeqNum() : highestSeqNum;
sumOfDuration += currentRecord.getDuration();
}
} else {
byte highestCauseForOutput = 0;
for (CallRecord currentRecord : completeCallRecords) {
highestSeqNum = currentRecord.getSeqNum() > highestSeqNum ? currentRecord.getSeqNum() : highestSeqNum;
sumOfDuration += currentRecord.getDuration();
highestCauseForOutput = currentRecord.getCauseForOutput() > highestCauseForOutput ? currentRecord.getCauseForOutput() : highestCauseForOutput;
}
}
Your desire to do everything in a single iteration is irrational. You should strive for simplicity first, performance if necessary, but insisting on a single iteration is neither.
The performance depends on too many factors to make a prediction in advance. The process of iterating (over a plain collection) itself is not necessarily an expensive operation and may even benefit from a simpler loop body in a way that makes multiple traversals with a straight-forward operation more efficient than a single traversal trying to do everything at once. The only way to find out, is to measure using the actual operations.
Converting the operation to Stream operations may simplify the code, if you use it straight-forwardly, i.e.
int highestSeqNum=
completeCallRecords.stream().mapToInt(CallRecord::getSeqNum).max().orElse(-1);
int sumOfDuration=
completeCallRecords.stream().mapToInt(CallRecord::getDuration).sum();
if(!condition) {
byte highestCauseForOutput = (byte)
completeCallRecords.stream().mapToInt(CallRecord::getCauseForOutput).max().orElse(0);
}
If you still feel uncomfortable with the fact that there are multiple iterations, you could try to write a custom collector performing all operations at once, but the result will not be better than your loop, neither in terms of readability nor efficiency.
Still, I’d prefer avoiding code duplication over trying to do everything in one loop, i.e.
for(CallRecord currentRecord : completeCallRecords) {
int nextSeqNum = currentRecord.getSeqNum();
highestSeqNum = nextSeqNum > highestSeqNum ? nextSeqNum : highestSeqNum;
sumOfDuration += currentRecord.getDuration();
}
if(!condition) {
byte highestCauseForOutput = 0;
for(CallRecord currentRecord : completeCallRecords) {
byte next = currentRecord.getCauseForOutput();
highestCauseForOutput = next > highestCauseForOutput? next: highestCauseForOutput;
}
}
With Java-8 you can resolved it with a Collector with no redudant iteration.
Normally, we can use the factory methods from Collectors, but in your case you need to implement a custom Collector, that reduces a Stream<CallRecord> to an instance of SummarizingCallRecord which cotains the attributes you require.
Mutable accumulation/result type:
class SummarizingCallRecord {
private int highestSeqNum = 0;
private int sumDuration = 0;
// getters/setters ...
}
Custom collector:
BiConsumer<SummarizingCallRecord, CallRecord> myAccumulator = (a, callRecord) -> {
a.setHighestSeqNum(Math.max(a.getHighestSeqNum(), callRecord.getSeqNum()));
a.setSumDuration(a.getSumDuration() + callRecord.getDuration());
};
BinaryOperator<SummarizingCallRecord> myCombiner = (a1, a2) -> {
a1.setHighestSeqNum(Math.max(a1.getHighestSeqNum(), a2.getHighestSeqNum()));
a1.setSumDuration(a1.getSumDuration() + a2.getSumDuration());
return a1;
};
Collector<CallRecord, SummarizingCallRecord, SummarizingCallRecord> myCollector =
Collector.of(
() -> new SummarizinCallRecord(),
myAccumulator,
myCombiner,
// Collector.Characteristics.CONCURRENT/IDENTITY_FINISH/UNORDERED
);
Execution example:
List<CallRecord> callRecords = new ArrayList<>();
callRecords.add(new CallRecord(1, 100));
callRecords.add(new CallRecord(5, 50));
callRecords.add(new CallRecord(3, 1000));
SummarizingCallRecord summarizingCallRecord = callRecords.stream()
.collect(myCollector);
// Result:
// summarizingCallRecord.highestSeqNum = 5
// summarizingCallRecord.sumDuration = 1150
You don't need and should not implement the logic by Stream API because the tradition for-loop is simple enough and the Java 8 Stream API can't make it simpler:
int highestSeqNum = 0;
long sumOfDuration = 0;
byte highestCauseForOutput = 0; // just get it even if it may not be used. there is no performance hurt.
for(CallRecord currentRecord : completeCallRecords) {
highestSeqNum = Math.max(highestSeqNum, currentRecord.getSeqNum());
sumOfDuration += currentRecord.getDuration();
highestCauseForOutput = Math.max(highestCauseForOutput, currentRecord.getCauseForOutput());
}
// Do something with or without highestCauseForOutput.
I just started a project for my Java2 class and I've come to a complete stop. I just can't get
my head around this method. Especially when the assignment does NOT let us use any other DATA STRUCTURE or shuffle methods from java at all.
So I have a Deck.class in which I've already created a linked list containing 52 nodes that hold 52 cards.
public class Deck {
private Node theDeck;
private int numCards;
public Deck ()
{
while(numCards < 52)
{
theDeck = new Node (new Card(numCards), theDeck);
numCards++;
}
}
public void shuffleDeck()
{
int rNum;
int count = 0;
Node current = theDeck;
Card tCard;
int range = 0;
while(count != 51)
{
// Store whatever is inside the current node in a temp variable
tCard = current.getItem();
// Generate a random number between 0 -51
rNum = (int)(Math.random()* 51);
// Send current on a loop a random amount of times
for (int i=0; i < rNum; i ++)
current = current.getNext(); ******<-- (Btw this is the line I'm getting my error, i sort of know why but idk how to stop it.)
// So wherever current landed get that item stored in that node and store it in the first on
theDeck.setItem(current.getItem());
// Now make use of the temp variable at the beginning and store it where current landed
current.setItem(tCard);
// Send current back to the beginning of the deck
current = theDeck;
// I've created a counter for another loop i want to do
count++;
// Send current a "count" amount of times for a loop so that it doesn't shuffle the cards that have been already shuffled.
for(int i=0; i<count; i++)
current = current.getNext(); ****<-- Not to sure about this last loop because if i don't shuffle the cards that i've already shuffled it will not count as a legitimate shuffle? i think? ****Also this is where i sometimes get a nullpointerexception****
}
}
}
Now I get different kinds of errors
When I call on this method:
it will sometimes shuffle just 2 cards but at times it will shuffle 3 - 5 cards then give me a NullPointerException.
I've pointed out where it gives me this error with asterisks in my code above
at one point I got it to shuffle 13 cards but then everytime it did that it didn't quite shuffle them the right way. one card kept always repeating.
at another point I got all 52 cards to go through the while loop but again it repeated one card various times.
So I really need some input in what I'm doing wrong. Towards the end of my code I think my logic is completely wrong but I can't seem to figure out a way around it.
Seems pretty long-winded.
I'd go with something like the following:
public void shuffleDeck() {
for(int i=0; i<52; i++) {
int card = (int) (Math.random() * (52-i));
deck.addLast(deck.remove(card));
}
}
So each card just gets moved to the back of the deck in a random order.
If you are authorized to use a secondary data structure, one way is simply to compute a random number within the number of remaining cards, select that card, move it to the end of the secondary structure until empty, then replace your list with the secondary list.
My implementation shuffles a linked list using a divide-and-conquer algorithm
public class LinkedListShuffle
{
public static DataStructures.Linear.LinkedListNode<T> Shuffle<T>(DataStructures.Linear.LinkedListNode<T> firstNode) where T : IComparable<T>
{
if (firstNode == null)
throw new ArgumentNullException();
if (firstNode.Next == null)
return firstNode;
var middle = GetMiddle(firstNode);
var rightNode = middle.Next;
middle.Next = null;
var mergedResult = ShuffledMerge(Shuffle(firstNode), Shuffle(rightNode));
return mergedResult;
}
private static DataStructures.Linear.LinkedListNode<T> ShuffledMerge<T>(DataStructures.Linear.LinkedListNode<T> leftNode, DataStructures.Linear.LinkedListNode<T> rightNode) where T : IComparable<T>
{
var dummyHead = new DataStructures.Linear.LinkedListNode<T>();
DataStructures.Linear.LinkedListNode<T> curNode = dummyHead;
var rnd = new Random((int)DateTime.Now.Ticks);
while (leftNode != null || rightNode != null)
{
var rndRes = rnd.Next(0, 2);
if (rndRes == 0)
{
if (leftNode != null)
{
curNode.Next = leftNode;
leftNode = leftNode.Next;
}
else
{
curNode.Next = rightNode;
rightNode = rightNode.Next;
}
}
else
{
if (rightNode != null)
{
curNode.Next = rightNode;
rightNode = rightNode.Next;
}
else
{
curNode.Next = leftNode;
leftNode = leftNode.Next;
}
}
curNode = curNode.Next;
}
return dummyHead.Next;
}
private static DataStructures.Linear.LinkedListNode<T> GetMiddle<T>(DataStructures.Linear.LinkedListNode<T> firstNode) where T : IComparable<T>
{
if (firstNode.Next == null)
return firstNode;
DataStructures.Linear.LinkedListNode<T> fast, slow;
fast = slow = firstNode;
while (fast.Next != null && fast.Next.Next != null)
{
slow = slow.Next;
fast = fast.Next.Next;
}
return slow;
}
}
Just came across this and decided to post a more concise solution which allows you to specify how much shuffling you want to do.
For the purposes of the answer, you have a linked list containing PlayingCard objects;
LinkedList<PlayingCard> deck = new LinkedList<PlayingCard>();
And to shuffle them use something like this;
public void shuffle(Integer swaps) {
for (int i=0; i < swaps; i++) {
deck.add(deck.remove((int)(Math.random() * deck.size())));
}
}
The more swaps you do, the more randomised the list will be.
I am trying to implement a simple singly linked list of integers which are to be sorted upon insertion in Visual Studio c++ 2010 express.
The problem is that when I create a new node and call the .getValue() function on it, the correct number is returned, however somehow that is being lost when I try calling getValue() on a node already in the list. The node might not be inserted into the list correctly, however I can't find why that would be the case. Some other value which looks like a reference value or something is displayed instead of the correct value.
I added current to the watch window when debugging but was still unable to see any of my variables other than the give value to be inserted. I am new to visual studio so I'm not sure if I'm missing something there. Here is my code:
#include "Node.h";
#include <iostream>
//namespace Linked{
//The first two constructors would be the first in the linked list.
Node::Node(void){
value = 0;
next = 0;
}
Node::Node(int setValue){
value = setValue;
next = 0;
}
Node::Node(int setValue,Node *nextNode){
value = setValue;
next = nextNode;
}
Node * Node::getNext(){
return next;
}
void Node::setNext(Node newNext){
next = &newNext;
}
int Node::getValue(){
return value;
}
bool Node::isEqual(Node check){
return value==check.getValue()&&next == check.getNext();
}
/*
int main(){
int firstInt, secondInt;
std::cin>>firstInt;
Node first = Node(firstInt);
std::cout<<"Enter second int: ";
std::cin>>secondInt;
Node second = Node(secondInt, &first);
std::cout<<"Second: "<<second.getValue()<<"\nFirst: "<<(*second.getNext()).getValue();
system("pause");
}*/
Here is the linked list:
//LinkedList.cpp
LinkedList::LinkedList(void)
{
head = 0;
size = 0;
}
LinkedList::LinkedList(int value)
{
head = &Node(value);
size = 1;
}
void LinkedList::insert(int value){
if(head == 0){
Node newNode = Node(value);
head = &newNode;
std::cout<<"Adding "<<(*head).getValue()<<" as head.\n";
}else{
std::cout<<"Adding ";
Node current = *head;
int numChecked = 0;
while(size<=numChecked && (((*current.getNext()).getValue())<value)){
current = (*(current.getNext()));
numChecked++;
}
if(current.isEqual(*head)&¤t.getValue()<value){
Node newNode = Node(value, ¤t);
std::cout<<newNode.getValue()<<" before the head: "<<current.getValue()<<"\n";
}else{
Node newNode = Node(value,current.getNext());
current.setNext(newNode);
std::cout<<newNode.getValue()<<" after "<<current.getValue()<<"\n";
}
}
size++;
}
void LinkedList::remove(int){
}
void LinkedList::print(){
Node current = *head;
std::cout<<current.getValue()<<" is the head";
int numPrinted = 0;
while(numPrinted<(size-1)){
std::cout<<(current.getValue())<<", ";
current = (*(current.getNext()));
numPrinted++;
}
}
int main(){
int a[5] = {30,20,25,13,2};
LinkedList myList = LinkedList();
int i;
for(i = 0 ; i<5 ; i++){
myList.insert(a[i]);
}
myList.print();
system("pause");
}
Any guidance would be greatly appreciated!
When you create nodes in insert, you're allocating them off the stack, which means that they'll be lost after the function returns.
Get them off the heap with:
Node * newNode=new Node(value);
When you use:
Node newNode=Node(value);
You're allocating that object on the stack, which means that pointers:
&newNode
to it are only valid until that function returns. If you use heap memory this is no longer an issue, but it does mean that you have to implement a destructor for your list which goes through and deletes each node.