Java Stacks/Cannot Find Logic Error - java

For my CS class, I had to create a boolean function isBalanced(String x) that takes a string and evaluates the amount of brackets/parentheses and returns true if the brackets match up to its pair (e.g; { is a pair of }, ( is a pair of ), [ is a pair of ], etc.). The function would return true if the brackets correctly matched up, false if otherwise. For clarification, MyStack() is my own implementation of the Java stack interface if you are wondering what that Object was.
Examples of how the code would work and return:
{A(B[C])D} would return true.
{A(B[C)]D} would return false.
The problem in my code is a logic error. For some reason, my function is returning true if there is a missing bracket, which should return false.
{A(B)C would return false, but my code reads it as true. Do you have any solutions that would help my code work properly? Thanks!
Balancer.java
public static boolean isBalanced(String x) {
MyStack<String> stack = new MyStack();
if (x.substring(0,1).equals("}") || x.substring(0,1).equals(")") || x.substring(0,1).equals("]")) {
return false;
}
for (int i=0; i<x.length(); i++) {
if (x.substring(i,i+1).equals("{") || x.substring(i,i+1).equals("(") || x.substring(i,i+1).equals("[")) {
stack.add(x.substring(i,i+1));
}
if (x.substring(i,i+1).equals("}") || x.substring(i,i+1).equals(")") || x.substring(i,i+1).equals("]")) {
if (x.substring(i,i+1).equals("}") && stack.peek().equals("{")) {
stack.pop();
} else if (x.substring(i,i+1).equals(")") && stack.peek().equals("(")) {
stack.pop();
} else if (x.substring(i,i+1).equals("]") && stack.peek().equals("[")) {
stack.pop();
} else {
return false;
}
}
}
return true;
}
This file, labeled Main.java, is just a tester. I have omitted the other cases where the code works. The reason why the function should return false is that there is a missing } which should be at the end, but there is none, yet my function returns true for some reason.
Main.java
public static void main(String[] args) {
...
String test4 = "{AA[B(CDE{FG()T})V]";
System.out.println("Missing final close (empty stack case)");
System.out.println("Should be false, is: " + Balancer.isBalanced(test4)); // does not work
}

You have a series of conditions that only pop the stack if a matching pair of brackets are found. I think you're overcomplicating things - if an opening bracket is found, push it to the stack. If a closing parenthesis is found, pop the stack and make sure that they match. E.g.:
private static final Map<Character, Character> CLODSE_TO_OPEN = new HashMap<>();
static {
CLODSE_TO_OPEN.put(')', '(');
CLODSE_TO_OPEN.put(']', '[');
CLODSE_TO_OPEN.put('}', '{');
}
public static boolean isBalanced(String x) {
Stack<Character> stack = new Stack<>();
for (int i = 0; i < x.length(); ++i) {
char c = x.charAt(i);
if (CLODSE_TO_OPEN.containsValue(c)) {
stack.push(c);
} else if (CLODSE_TO_OPEN.containsKey(c)) {
try {
if (!CLODSE_TO_OPEN.get(c).equals(stack.pop())) {
return false;
}
} catch (EmptyStackException e) {
return false;
}
}
}
return stack.isEmpty();
}

Related

Almost increasing sequence coding problem

I am trying to solve a coding problem. The problem is following:
Given a sequence of integers as an array, determine whether it is possible to obtain a strictly increasing sequence by removing no more than one element from the array.
For example:
[1,3,2,1] is false
[1,3,2] is true
I implemented it in Java. The code is as follows:
boolean almostIncreasingSequence(int[] sequence) {
int count =0;
for(int i =0; i < sequence.length; i++){
if (sequence[i] <= sequence[i-1]){
count++;
}
if(count>1){
return false;
}
if(sequence[i] <= sequence[i-2] && sequence[i+1] <= sequence[i-1]){
return false;
}
}
return true;
}
This is the following error:
Execution error on test 1: Your program had a runtime error.
Any help will be appreciated. It seems a small problem but I can't resolve it.
One implementation can be based on remove just 1 element when strictly ascending condition is not achieved.
public class TestAlmostIncreasingSequence {
public static boolean almostIncreasingSequence(int[] sequence)
{
if(sequence==null) return false;
//mandatory to remove just 1 element, if no one(or more) removed then false
boolean flag_removed=false;
for(int i=1, prev=sequence[0];i<sequence.length;i++)
{
if(prev>=sequence[i] && flag_removed==false)
{
//mark removed
flag_removed=true;
}
//if element was removed then false
else if(prev>=sequence[i] && flag_removed==true)
{
return false;
}
else
{
//change only if element removed is not the current
//comparisons will not be done with removed element
prev=sequence[i];
}
//System.out.println(prev);
}
//could have a strictly increased arr by default which will return false [1,2,3]
return flag_removed;
}
public static void main(String[] args)
{
//only for printing purpose
String arr="";
int s1[] = {1,2,3,1};
arr=Arrays.stream(s1).mapToObj(t->String.valueOf(t)).
collect(Collectors.joining(",","[","]"));
System.out.println(arr+"\n"+almostIncreasingSequence(s1)+"\n");
int s2[] = {1,2,3};
arr=Arrays.stream(s2).mapToObj(t->String.valueOf(t)).
collect(Collectors.joining(",","[","]"));
System.out.println(arr+"\n"+almostIncreasingSequence(s2)+"\n");
int s3[] = {1,2,3,1,2};
arr=Arrays.stream(s3).mapToObj(t->String.valueOf(t)).
collect(Collectors.joining(",","[","]"));
System.out.println(arr+"\n"+almostIncreasingSequence(s3)+"\n");
int s4[] = {1};
arr=Arrays.stream(s4).mapToObj(t->String.valueOf(t)).
collect(Collectors.joining(",","[","]"));
System.out.println(arr+"\n"+almostIncreasingSequence(s4)+"\n");
int s5[] = {1,1};
arr=Arrays.stream(s5).mapToObj(t->String.valueOf(t)).
collect(Collectors.joining(",","[","]"));
System.out.println(arr+"\n"+almostIncreasingSequence(s5)+"\n");
int s6[] = null;
arr="null";
System.out.println(arr+"\n"+almostIncreasingSequence(s6)+"\n");
}
}
Output
[1,2,3,1]
true
[1,2,3]
false
[1,2,3,1,2]
false
[1]
false
[1,1]
true
null
false
Note: The implementation have a case when the result is wrong [1,5,2,3], just update with one more branch with removed element=the previous one(not the current) and check both branched (one true means true)
This should fix the case
//method name is misguided, removePrev is better
public static boolean removeCurrent(int[] sequence)
{
if(sequence==null) return false;
//mandatory to remove just 1 element, if no one remove then false
boolean flag_removed=false;
for(int i=1, prev=sequence[0];i<sequence.length;i++)
{
if(prev>=sequence[i] && flag_removed==false)
{
//mark removed
flag_removed=true;
}
//if element was removed then false
else if(prev>=sequence[i] && flag_removed==true)
{
return false;
}
//compared element will be the current one
prev=sequence[i];
//System.out.println(prev);
}
//could have a strictly increased arr by default which will return false [1,2,3]
return flag_removed;
}
and use
int s1[] = {1,5,2,3};
arr=Arrays.stream(s1).mapToObj(t->String.valueOf(t)).
collect(Collectors.joining(",","[","]"));
boolean result= (almostIncreasingSequence(s1)==false) ? removeCurrent(s1) : true;
System.out.println(arr+"\n"+result +"\n");
Output
[1,5,2,3]
true (from removeCurrent_branch)
Seems one more case is wrong [5,6,3,4], means need to see if element[i-2](only after remove element) is not greater then current and 'prev' on last branch.
6>3 remove 6 (prev=3, 3<4 but [5>4 or 5>3] so false)
public static boolean removeCurrent(int[] sequence)
{
if(sequence==null) return false;
//mandatory to remove just 1 element, if no one remove then false
boolean flag_removed=false;
for(int i=1, prev=sequence[0], twoprev=Integer.MIN_VALUE;i<sequence.length;i++)
{
if(prev>=sequence[i] && flag_removed==false)
{
//mark removed
flag_removed=true;
if(i>=2) twoprev=sequence[i-2];
}
//if element was removed then false
else if(prev>=sequence[i] && flag_removed==true)
{
return false;
}
else if(twoprev>=sequence[i] || twoprev>=prev)
{
return false;
}
//compared element will be the current one
prev=sequence[i];
//System.out.println(prev);
}
//could have a strictly increased arr by default which will return false [1,2,3]
return flag_removed;
}
Output
[5,6,3,4]
false
Now, as far as I see all cases seems covered.
Brute force can also generate a solution but will be less optimal.(use a loop to remove an element, sort the result and compare with base)
public class TestInc {
public static void main(String[] args)
{
int s1[] = {1,1,2,3};
System.out.println(checkInc(s1));
}
public static boolean checkInc(int[] arr)
{
if(arr==null || arr.length==1) return false;
List<Integer> lst = Arrays.stream(arr).boxed().collect(Collectors.toList());
//remove this check if requirement is other(or return true)
if(checkIfAlreadySortedAsc(lst))
{
return false;
}
for(int i=0;i<lst.size();i++)
{
List<Integer> auxLst = new ArrayList<Integer>(lst);
auxLst.remove(i);
List<Integer> sorted = new ArrayList<Integer>(auxLst);
sorted = sorted.stream().distinct().sorted().collect(Collectors.toList());
if(auxLst.equals(sorted))
{
// System.out.println("=");
return true;
}
else
{
// System.out.println("!=");
}
}
return false;
}
//any ascending sorted list will be the same type if remove one element
//but as requirement on this case will return false
//(or don't use method in want other)
public static boolean checkIfAlreadySortedAsc(List<Integer> lst)
{
List<Integer> auxLst = new ArrayList<Integer>(lst);
auxLst = auxLst.stream().distinct().sorted().collect(Collectors.toList());
if(auxLst.equals(lst))
{
return true;
}
return false;
}
}
Output
[1,1,2,3]
true
This line would produce an ArrayIndexOutOfBoundsException when i == 0 because it will attempt to access sequence[-1]
if (sequence[i] <= sequence[i-1]){

unable to correctly validate balanced parenthesis parsing in java method

I have a method that's supposed to validate accurate opening and closing parenthesis in a string using java. This method will be used to parse mathematical expressions so it's important for the parenthesis to be balanced. For some reason, it is returning false in both of these runs:
System.out.println(parChecker("(()")); // returns false :-)
System.out.println(parChecker("((()))")); // returns false :-( WHY??
Here is the method that uses a stack to solve the problem. Something is wrong here becuase it's returning false for a balanced set of parenthesis as well. What's the problem? Thank you in advance.
public static boolean parChecker(String str) {
String[] tokens = str.split("");
int size = tokens.length;
Stack theStack = new Stack(size);
int index = 0;
boolean balanced = true;
while ((index < size) && balanced) {
String symbol = tokens[index];
if (symbol.equals("(")) {
theStack.push(symbol);
} else {
if (theStack.isEmpty()) {
balanced = false;
} else {
theStack.pop();
}
}
index++;
}
if (balanced && theStack.isEmpty()) {
return true;
} else {
return false;
}
}
Here is my stack class that I'm using:
public class Stack {
private Object [] stack;
private int maxSize = 0;
private int top;
public Stack(int size){
maxSize = size;
stack = new Object[maxSize];
top = -1;
}
public void push(Object obj){
top++;
stack[top] = obj;
}
public Object pop(){
return stack[top--];
}
public Object peek(){
return stack[top];
}
public boolean isEmpty(){
return (top == -1);
}
public boolean isFull(){
return (top == maxSize -1);
}
}
The immediate problem is this
String[] tokens = str.split("");
Gives you first char = "" if you use java 1.7 or less, so you will exit your loop since stack is empty...
Note: this has been changed in java 1.8 split difference between java 1.7 and 1.8
change to:
char[] tokens = str.toCharArray();
I guess however that you need to consider the fact that there can be chars before your first ( and that you may have other chars then ( and )
As far as I can tell, there is no problem with the code (turns out it's a Java 7 specific issue..).
I would like to offer a replacement method though, for educational purposes, that is shorter, and and is tolerant of other characters being present:
public static boolean parChecker(String str) {
Stack stack = new Stack(str.length());
for (char c : str.toCharArray())
switch (c) {
case '(':
stack.push(c);
break;
case ')':
if (stack.isEmpty() || stack.pop() != Character.valueOf('('))
return false;
}
return stack.isEmpty();
}
As requested, here's another solution that doesn't use a stack:
public static boolean parChecker(String str) {
str = str.replaceAll("[^()]", "");
while (str.contains("()"))
str = str.replace("()", "");
return str.isEmpty();
}
And one more for the road: #FredK's algorithm:
public static boolean parChecker(String str) {
int depth = 0;
for ( char c : str.toCharArray() )
if ( ( depth += c == '(' ? 1 : c == ')' ? -1 : 0 ) < 0 )
return false;
return depth == 0;
}

Ignore a letter in isPalindrome() method - Java

I am wondering how the following method could be modified to ignore a certain letter and have it as a wildcard when checking if a string is a palindrome...
example: "wows", in this case the method should return false but
"pat" , the 't' can be a wildcard (considered as a p) and therefore it returns true
"job" , again the b can be a wildcard and considered as a j, therefore method returns true.
this is what i have so far, i have a seperate method ignoring special characters and spaces so that does not need to be considered in this post.
private static boolean checkPalindrome2(String word) {
if(word.length() < 2) {
return true;
}
char first = word.charAt(0);
char last = word.charAt(word.length()-1);
if(first != last) {
return false;
}
else {
return checkPalindrome2(word.substring(1,word.length()-1));
}
}
this is my test class,
public class testPalindromes {
public static void main(String[] args) {
//if (Palindromes.isPalindrome("a") == true) {
// System.out.println("true");
//} else {
// System.out.println("false");
//}
// block above is the same as this
// isPalindrome already returns true or false,
// and true and false can be printed as strings
System.out.println(isPalindrome("a"));
if (Palindromes.isPalindrome("cat") == true) {
System.out.println("true");
} else {
System.out.println("false");
}
if (Palindromes.isPalindrome("w o w") == true) {
System.out.println("true");
} else {
System.out.println("false");
}
if (Palindromes.isPalindrome(" a ") == true) {
System.out.println("true");
} else {
System.out.println("false");
}
if (Palindromes.isPalindrome("mom!") == true) {
System.out.println("true");
if (Palindromes.isPalindrome2("cat")==true){
System.out.print("true");
} else {
System.out.println("false");
}
}
}
}
isPalindromes2 is a method calling the checkPalindrome2 method above, the last case in my test class (word cat) should return true because the t will be the wildcard letter (wildcard again as stated above, replaced with c making cat, cac which is a palindrome)
thanks in advance for all help / input!!!!
ps, i purposely implemented a recursive method.
Just add extra conditions to your base case:
// both first and last have to NOT be the special character
// and first has to not equal last for this to return false
if(first != special && last != special && first != last)
return false;
else
return checkPalindrome(word.substring(1,word.length()-1));

java homework(recursion)

This is the question:
Problem I.
We define the Pestaina strings as follows:
ab is a Pestaina string.
cbac is a Pestaina string.
If S is a Pestaina string, so is SaS.
If U and V are Pestaina strings, so is UbV.
Here a, b, c are constants and S,U,V are variables. In these rules,
the same letter represents the same string. So, if S = ab, rule 3
tells us that abaab is a Pestaina string. In rule 4, U and V represent
Grandpa strings, but they may be different.
Write the method
public static boolean isPestaina(String in)
That returns true if in is a Pestaina string and false otherwise.
And this is what i have so far which only works for the first rule, but the are some cases in which doesnt work for example "abaaab":
public class Main {
private static boolean bool = true;
public static void main(String[] args){
String pestaina = "abaaab";
System.out.println(pestaina+" "+pestainaString(pestaina));
}
public static boolean pestainaString(String p){
if(p == null || p.length() == 0 || p.length() == 3) {
return false;
}
if(p.equals("ab")) {
return true;
}
if(p.startsWith("ab")){
bool = pestainaString(p, 1);
}else{
bool = false;
}
return bool;
}
public static boolean pestainaString(String p, int sign){
String letter;
char concat;
if("".equals(p)){
return false;
}
if(p.length() < 3){
letter = p;
concat = ' ';
p = "";
pestainaString(p);
}else if(p.length() == 3 && (!"ab".equals(p.substring(0, 2)) || p.charAt(2) != 'a')){
letter = p.substring(0, 2);
concat = p.charAt(2);
p = "";
pestainaString(p);
}else{
letter = p.substring(0, 2);
concat = p.charAt(2);
pestainaString(p.substring(3));
}
if(letter.length() == 2 && concat == ' '){
if(!"ab".equals(letter.trim())){
bool = false;
//concat = 'a';
}
}else if((!"ab".equals(letter)) || (concat != 'a')){
bool = false;
}
System.out.println(letter +" " + concat);
return bool;
}
}
Please tell me what i have done wrong.
I found the problem i was calling the wrong method.
You are describing a Context Free Language, which can be described as a Context Free Grammer and parsed with it. The field of parsing these is widely researched and there is a lot of resources for it out there.
The wikipedia page also discusses some algorithms to parse these, specifically - I think you are interested in the Early Parsers
I also believe this "language" can be parsed using a push down automaton (though not 100% sure about it).
public static void main(String[] args) {
// TODO code application logic here
String text = "cbacacbac";
System.out.println("Is \""+ text +"\" a Pestaina string? " + isPestaina(text));
}
public static boolean isPestaina(String in) {
if (in.equals("ab")) {
return true;
}
if (in.equals("cbac")) {
return true;
}
if (in.length() > 3) {
if ((in.startsWith("ab") || in.startsWith("cbac"))
&& (in.endsWith("ab") || in.endsWith("cbac"))) {
return true;
}
}
return false;
}
That was fun.
public boolean isPestaina(String p) {
Set<String> existingPestainas = new HashSet<String>(Arrays.asList(new String[]{"ab", "cbac"}));
boolean isP = false;
int lengthParsed = 0;
do {
if (lengthParsed > 0) {
//just realized there's a touch more to do here for the a/b
//connecting rules...I'll leave it as an excersize for the readers.
if (p.substring(lengthParsed).startsWith("a") ||
p.substring(lengthParsed).startsWith("b")) {
//good connector.
lengthParsed++;
} else {
//bad connector;
return false;
}
}
for (String existingP : existingPestainas) {
if (p.substring(lengthParsed).startsWith(existingP)) {
isP = true;
lengthParsed += existingP.length();
}
}
if (isP) {
System.err.println("Adding pestaina: " + p.substring(0, lengthParsed));
existingPestainas.add(p.substring(0, lengthParsed));
}
} while (isP && p.length() >= lengthParsed + 1);
return isP;
}

Can someone suggest simplification for the code?

public boolean isPalindrome()
{
Stack myStack = new Stack();
for(Node current = head; current!=null; current = current.next)
{
if(!myStack.isEmpty())
{
if(myStack.peek()==current.data)
{
myStack.pop();
}else if(current.next!=null&&myStack.peek()==current.next.data)
{
continue;
}
else
{
myStack.push(current.data);
}
}else
{
myStack.push(current.data);
}
}
return myStack.isEmpty();
}
What I am doing here is using a stack to check whether a linked list is a palindrome. It works as expected only thing is I wanted to get rid of code duplication where the else condition has a push of the data onto the stack.
The algorithm is unfortunately not correct. For "abbaaa" it would report that that is a palindrome, although it isn't. Checking for palindromes without using the length is difficult.
abbaaa () -> push a
bbaaa (a) -> push b
baaa (ba) -> pop b
aaa (a) -> pop a
aa () -> push a
a (a) -> pop a
() -> palindrome
This is a somewhat classic problem. There are many ways to solve it in java. One of the easiest is this one:
boolean isPalindrome(String s) {
for (int i=0, len=s.length(); i<len/2; i++) {
if (s.charAt(i) != s.charAt(len-i-1)) return false;
}
return true;
}
(Strictly speaking, this is a rewrite rather than a refactoring; however, any rewrite that preserves method signatures can be seen as a refactoring... and it is certainly more efficient)
If all you want to do is remove the code duplication between the two else conditions then remove them entirely.
public boolean isPalindrome()
{
Stack myStack = new Stack();
for(Node current = head; current!=null; current = current.next)
{
if(!myStack.isEmpty())
{
if(myStack.peek()==current.data)
{
myStack.pop();
continue;
}else if(current.next!=null&&myStack.peek()==current.next.data)
{
continue;
}
}
myStack.push(current.data);
}
return myStack.isEmpty();
}
A simplification of functionality;
boolean isPalinDrome(String testString) {
return new StringBuffer(testString).reverse().toString().equals(testString);
}
This should provide same functionality without repeat. It is pointed out however that your algorithm doesn't seem to be correct.
public boolean isPalindrome()
{
Stack myStack = new Stack();
boolean doPush;
for(Node current = head; current!=null; current = current.next)
{
doPush = true;
if(!myStack.isEmpty())
{
if(myStack.peek()==current.data)
{
doPush = false;
myStack.pop();
}else if(current.next!=null&&myStack.peek()==current.next.data)
{
doPush = false;
continue;
}
}
if(doPush){
myStack.push(current.data);
}
}
return myStack.isEmpty();
}

Categories