We Keep Coding

Writing a static method in Java to return a string using loops

The full assignment is:
Write two public classes (named exactly), TextBox and TextBoxTester. TextBox contains the following overloaded static methods called textBoxString. This method returns a String value.
public static String textBoxString (int side)
The returned String value, when printed, displays as the outline of a square of side characters. The character you use is up to you. Don't forget that '\n' will force a newline character into the returned String. For example, let's assume I want to use * as the character for my box:
String s = textBoxString(3);
System.out.println(s);
will print
xxx
x x
xxx
public static String textBoxString(int side, char bChar)
The returned String value, when printed, displays the outline of a square of side characters using bChar as the box character. For example,
String s = textBoxString(4, '+');
System.out.println(s);
will print
++++
+ +
+ +
++++
So far, this is what I have:
public class TextBox {
public static String textBoxString(int side) {
int n = side;
String string = "";
for (int i = 0; i < n; i++) {
if (i == 0 || i == n - 1) {
string = "***";
}
else {
string = "* *";
}
}
return string;
}
public static String textBoxString(int side, char bChar) {
int n = side;
char c = bChar;
String string = "";
for (int i = 0; i < n; i++) {
if (i == 0 || i == n -1) {
string = c + " " + c + " " + c + " " + c;
}
else {
string = c + " " + c;
}
}
return string;
}
}
My issue is returning the result. I don't know how to return the result that I need, that I would get from running this in a main method (for the first method):
int n = 3;
for (int i = 0; i < n; i++) {
if (i == 0 || i == n - 1) {
System.out.println("***");
}
else {
System.out.println("* *");
}
}

You can use nested loops to achieve this.
public static String textBoxString(int n, char c) {
String output = "";
for(int i = 0; i < n; i++) { // constructing all rows
output += c;
for(int j = 1; j < n-1; j++) { // constructing a single row
if(i == 0 || i == n - 1)
output += " " + c; // use character to fill up each element in the row
else
output += " "; // use space to fill up
}
output += " " + c +"\n";
}
return output;
}
Run and print
x x x x x
x x
x x
x x
x x x x x

Welcome to StackOverflow!
To make textBoxString work for varying values for side, you cannot use the hardcoded string "***" because that will only work when side equals 3.
You'll need to use nested loops to accomplish this.
I'm not gonna write the complete code for you, but I'm gonna give you a hint.
if (i == 0 || i == n - 1) {
// Use a for loop here to construct the top and bottom rows
}
else {
string = "* *";
}

You could add "\n" to your string, it will start a new line.
for (int i = 0; i < n; i++) {
if (i == 0 || i == n -1) {
string += c + " " + c + " " + c + " " + c + "\n";
}
else {
string += c + " " + c + "\n";
}
}
}
After adding that in, return string

public static String textBoxString(int side) {
int n = side;
String string = "";
for (int i = 0; i < n; i++) {
if (i == 0 || i == n - 1) {
for (int j = 0; j < n; j++) {
string += "*";
}
string += "\n";
} else {
string += "*";
for (int j = 1; j < n - 1; j++) {
string += " ";
}
string += "*\n";
}
}
return string;
}

I get java.lang.StringIndexOutOfBoundsException when translating c++ code into java code

I'm trying to solve this problem
https://vjudge.net/problem/UVALive-6805
I found solution but in c++ , Can anybody help me converting it to java code. I'm very newbie to programming
I tried a lot of solutions but non of them work.
Please I need help in this if possible
I don't know for example what is the equivalent for .erase function in c++ in java
Also is is sbstr in c++ provide different result from java ?
#include <iostream>
#include <string>
using namespace std;
int syllable(string word)
{
int L = word.size();
int syllable;
if (L>=7)
{
syllable = 3;
}
else if (L==6)
{
int indicator = 0;
for (int k=0; k<=L-2; k++)
{
string subword = word.substr(k, 2);
if (subword == "ng" || subword == "ny")
{
indicator++;
}
}
if (indicator == 0)
{
syllable = 3;
}
else
{
syllable = 2;
}
}
else if (L == 4 || L == 5)
{
syllable = 2;
}
else if (L == 3)
{
char Char = word[0];
if (Char=='a' || Char=='A' || Char=='e' || Char=='E' || Char=='i' || Char=='I' || Char=='o' || Char=='O' || Char=='u' || Char=='U')
{
syllable = 2;
}
else
{
syllable = 1;
}
}
else
{
syllable = 1;
}
return syllable;
}
int main()
{
string word;
int T;
cin >> T;
for (int i=1; i<=T; i++)
{
int syl[] = {0, -1, -2, -3};
string rhy[] = {"a", "b", "c", "d"};
int verse = 0;
int stop = 0;
while (stop == 0)
{
cin >> word;
int L = word.size();
char end = word[L-1];
if (end == '.')
{
stop = 1;
}
if (word[L-1] == ',' || word[L-1] == '.')
{
word = word.erase(L-1, 1);
L = word.size();
}
if (verse<=3)
{
syl[verse] = syl[verse] + syllable(word);
}
if (end == ',' || end == '.')
{
if (verse<=3)
{
rhy[verse] = word.substr(L-2, 2);
}
verse++;
if (verse<=3)
{
syl[verse] = 0;
}
}
}
int A = 0, B = 0, C = 0, D = 0;
for (int k=0; k<4; k++)
{
if (syl[k] >= 8 && syl[k] <= 12)
{
A = A + 10;
}
}
for (int k=0; k<2; k++)
{
if (rhy[k] == rhy[k+2])
{
B = B + 20;
}
}
for (int k=0; k<2; k++)
{
if (syl[k] == syl[k+2])
{
C = C + 10;
}
}
if (verse > 4)
{
D = (verse - 4) * 10;
}
int E = A + B + C - D;
cout << "Case #" << i << ": " << A << " " << B << " " << C << " " << D << " " << E << endl;
}
}
here is my trying
import java.util.*;
public class First {
public static int syllable(String word) {
int L = word.length();
int syllable;
if (L >= 7) {
syllable = 3;
} else if (L == 6) {
int indicator = 0;
for (int k = 0; k < L - 3; k++) {
String subword = word.substring(k, 2);
if (subword == "ng" || subword == "ny") {
indicator++;
}
}
if (indicator == 0) {
syllable = 3;
} else {
syllable = 2;
}
} else if (L == 4 || L == 5) {
syllable = 2;
} else if (L == 3) {
char Char = word.charAt(0);
if (Char == 'a' || Char == 'A' || Char == 'e' || Char == 'E' || Char == 'i' || Char == 'I' || Char == 'o'
|| Char == 'O' || Char == 'u' || Char == 'U') {
syllable = 2;
} else {
syllable = 1;
}
} else {
syllable = 1;
}
return syllable;
}
public static void main(String[] args) {
Scanner sc = new Scanner(System.in);
String word;
int T;
T = sc.nextInt();
for (int i = 1; i <= T; i++) {
int syl[] = { 0, -1, -2, -3 };
String rhy[] = { "a", "b", "c", "d" };
int verse = 0;
int stop = 0;
while (stop == 0) {
word = sc.next();
int L = word.length();
char end = word.charAt(L-1);
if (end == '.') {
stop = 1;
}
if (word.charAt(L-1) == ',' || word.charAt(L-1) == '.') {
word.substring(L-1, 1);
L = word.length();
}
if (verse <= 3) {
syl[verse] = syl[verse] + syllable(word);
}
if (end == ',' || end == '.') {
if (verse <= 3) {
rhy[verse] = word.substring(L - 2, 2);
}
verse++;
if (verse <= 3) {
syl[verse] = 0;
}
}
}
int A = 0, B = 0, C = 0, D = 0;
for (int k = 0; k < 4; k++) {
if (syl[k] >= 8 && syl[k] <= 12) {
A = A + 10;
}
}
for (int k = 0; k < 2; k++) {
if (rhy[k] == rhy[k + 2]) {
B = B + 20;
}
}
for (int k = 0; k < 2; k++) {
if (syl[k] == syl[k + 2]) {
C = C + 10;
}
}
if (verse > 4) {
D = (verse - 4) * 10;
}
int E = A + B + C - D;
System.out.println("Case #" + i + ": " + A + " " + B + " " + C + " " + D + " " + E);
}
}
}

The Exception is thrown by your second and your third call of String substring method. Your beginIndex is higher than your endIndex. As you can see in here https://docs.oracle.com/javase/7/docs/api/java/lang/String.html#substring(int,%20int) beginIndex always has to be lower than the endIndex.

Before answering your question, there are some important points to mention in regards to Strings and Java in general.
Strings are immutable (This also applies to C++). This means that no method called on a String will change it, and that all methods simply return new versions of the original String with the operations done on it
The substring method in java has two forms.
One takes in beginIndex and returns everything from beginIndex to str.length() - 1 (where str represents a String)
The other takes in the beginIndex, and the endIndex, and returns everything from beginIndex to endIndex - 1. The beginIndex should never be larger than endIndex otherwise it throws an IndexOutOfBoundsException
C++'s substring method (string::substr()) takes in the beginning "index" and takes in the number of characters after it to include in the substring. So by doing substr(L-2, 2) you get the last two characters of the string.
Java will never allow you to go out of bounds. That means you need to constantly check whether you are within the bounds of anything you are iterating through.
With all this in mind, I would go and verify that all of the substring() method calls are returning the proper range of characters, and that you are properly reassigning the values returned from substring() to the proper variable.
To mimic C++'s string::erase(), depending on what part of the word you want to erase, you want to get the substring of the part before and the substring of the part after it and add them together.
Ex. Lets say I have a String line = "I do not like the movies"; Since it is impossible for anyone to not like movies, we want to cut out the word not
We do this by doing what I said above
String before = line.substring(0, 5); // This gives us "I do " since it goes up to but not including the 5th index.
String after = line.substring(5 + 3); // This gives us the rest of the string starting after the word "not" because not is 3 characters long and this skips to the 3rd index after index 5 (or index 8)
line = before + after; // This'll add those two Strings together and give you "I do like the movies"
Hope this helps!

search a given word into a letter puzzle diagonally

I have a two-dimensional array filled with random letters. I have words to find in that array.
I have written a toString method that uses:
startX : The start X position of the String to be found
startY : The start Y position of the String to be found
endX : The end X position of the String to be found
endY : The end Y position of the String to be found
The code that I provide works horizontally and vertically but does not work for diagonals. How can I print words which are placed in the array diagonally?
#Override
public String toString() {
StringBuilder sb = new StringBuilder();
if (startX == endX) {
if (startY < endY) {
for (int i = startY; i <= endY; i++)
sb.append(i).append("x").append(startY).append(" ");
} else {
for (int i = endY; i <= startY; i++)
sb.append(i).append("x").append(startY).append(" ");
}
}
if (startY == endY) {
if (startX < endX) {
for (int i = startX; i <= endX; i++)
sb.append(i).append("x").append(startY).append(" ");
} else
for (int i = endX; i <= startX; i++)
sb.append(i).append("x").append(startY).append(" ");
}
if (startX > endX && startY > endY) {
int i = startX;
int j = startY;
while (i >= endX)
sb.append(i--).append("x").append(j--).append(" ");
} else if (startX > endX && startY < endY) {
int i = startX;
int j = startY;
while (i >= endX)
sb.append(i--).append("x").append(j++).append(" ");
} else if (startX < endX && startY > endY) {
int i = startX;
int j = startY;
while (i >= endX)
sb.append(i++).append("x").append(j--).append(" ");
} else if (startX < endX && startY < endY) {
int i = startX;
int j = startY;
while (i >= endX)
sb.append(i++).append("x").append(j++).append(" ");
}
return sb.toString();
}

I assume that what you are looking for is a way to find a word in a letter puzzle.
In such a case, I suggest you to store the puzzle on a 2D array and the word to find in a String. Then you need to check all the positions of the array that have the same character than the begining character of the String you are looking for (in the code I provide: findWord). Once you find a match, you need to check the rest of the characters of the string (in the code I provide: checkDirections). If the rest of the characters match then you have found the string, otherwise you need to check for the other directions or for the next appearance of the first letter of the string.
Next I provide the code:
package letterPuzzle;
import java.util.Random;
public class LetterPuzzle {
private static final String ALPHABET = "abcdefghijklmnopqrstuvwxyz";
private static final int[] DIRECTIONS_X = new int[] { 0, 0, 1, -1, 1, 1, -1, -1 };
private static final int[] DIRECTIONS_Y = new int[] { 1, -1, 0, 0, 1, -1, 1, -1 };
private static int N;
private static char[][] puzzle;
private static void initializePuzzle() {
Random r = new Random();
puzzle = new char[N][N];
for (int i = 0; i < N; ++i) {
for (int j = 0; j < N; ++j) {
puzzle[i][j] = ALPHABET.charAt(r.nextInt(ALPHABET.length()));
}
}
// Add the JAVA word in a location
if (N < 6) {
System.out.println("[ERRRO] Example needs N >= 6");
System.exit(1);
}
puzzle[2][3] = 'j';
puzzle[3][3] = 'a';
puzzle[4][3] = 'v';
puzzle[5][3] = 'a';
}
private static void printPuzzle() {
System.out.println("[DEBUG] Puzzle");
for (int i = 0; i < N; ++i) {
for (int j = 0; j < N; ++j) {
System.out.print(puzzle[i][j] + " ");
}
System.out.println("");
}
System.out.println("[DEBUG] End Puzzle");
}
private static boolean findWord(String word) {
for (int i = 0; i < N; ++i) {
for (int j = 0; j < N; ++j) {
// We check all the matrix but only try to match the word if the first letter matches
if (puzzle[i][j] == word.charAt(0)) {
if (checkDirections(i, j, word)) {
return true;
}
}
}
}
return false;
}
private static boolean checkDirections(int initX, int initY, String word) {
System.out.println("Searching " + word + " from (" + initX + ", " + initY + ")");
// Checks the different directions from (initX, initY) position
for (int dirIndex = 0; dirIndex < DIRECTIONS_X.length; ++dirIndex) {
System.out.println(" - Searching direction " + dirIndex);
boolean wordMatches = true;
// Checks all the characters in an specific direction
for (int charIndex = 0; charIndex < word.length() && wordMatches; ++charIndex) {
int x = initX + DIRECTIONS_X[dirIndex] * charIndex;
int y = initY + DIRECTIONS_Y[dirIndex] * charIndex;
System.out.println(" -- Checking position (" + x + ", " + y + ")");
if (x < 0 || y < 0 || x >= N || y >= N || puzzle[x][y] != word.charAt(charIndex)) {
System.out.println(" -- Not match");
wordMatches = false;
} else {
System.out.println(" -- Partial match");
}
}
// If the word matches we stop, otherwise we check other directions
if (wordMatches) {
return true;
}
}
return false;
}
public static void main(String[] args) {
// Check args
if (args.length != 2) {
System.err.println("[ERROR] Invalid usage");
System.err.println("[ERROR] main <puzzleSize> <wordToSearch>");
}
// Get args
N = Integer.valueOf(args[0]);
String word = args[1];
// Initialize puzzle (randomly)
initializePuzzle();
printPuzzle();
// Search word
boolean isPresent = findWord(word);
if (isPresent) {
System.out.println("Word found");
} else {
System.out.println("Word NOT found");
}
}
}
Notice that:
The puzzle matrix is randomly initialized and I hardcoded the word 'java' on the 2,3 -> 5,3 positions (this is just for the example but you should initialize the puzzle from the command line or from a file).
The ALPHABET variable is used only for the random generation.
The directions are stored on two 1D arrays to make the 8 directions programatically but you can unroll for the sake of clarity.
It is probably not best efficient code in terms of performance since you will double check lots of positions if the first character of the string appears several times. However it is still a feasible and easy solution.

If :
each word of yours is in a particular row and doesn't overflows to next row and
all words are consecutive
then you can do something like this:
#Override
public String toString()
{
StringBuilder string = new StringBuilder();
for(int c = startY; c<=endY; c++) {
string.append(startX).append("x").append(c).append(", ");
}
return string.toString();
}

I made a for loop to swap two indiv char variables in a string

I am supposed to write a method that scrambles a word, switching two letters in a word that is not the first or last character.
I've initialized i and j as random integers between positions 1 and str.length() -1. Is there a reason why this loop would not print the scrambled version of a word?
char[] chararray = word.toCharArray();
int i = (int) (Math.random() * ((word.length()-2) - (1)) + 1);
int j = (int) (Math.random() * ((word.length()-2) - (1)) + 1);
for (int x = 0; x < word.length(); x++)
{
if (x == i)
{
chararray[x] = word.charAt(i);
}
else if (x == j)
{
chararray[x] = word.charAt(j);
}
}
word = new String (chararray);
System.out.println(word);
When I input tofu it would reprint tofu. I'd like to input "tofu" and have it output "tfou".

I think you would be looking for the following
for (int x = 0; x < word.length(); x++)
{
if (x == i)
{
chararray[x] = word.charAt(j);
}
else if (x == j)
{
chararray[x] = word.charAt(i);
}
}
Interchange the i and j values in the if and else blocks.
when x equals i you are assigning the same value back in the chararray, hence you are getting the same value. But your intention is to interchange the character in i and j positions I hope.
And one more thing, make sure i never equals to j

Your logic seems to be wrong,
if (x == i)
{
chararray[x] = word.charAt(i);
}
The above will just point to same index, both x and i points to same, you need to interchange the indexes.
Below code
public static void main(String args[]) {
String word = "ankur";
char[] chararray = word.toCharArray();
int i = (int) (Math.random() * ((word.length() - 2) - (1)) + 1);
int j = (int) (Math.random() * ((word.length() - 2) - (1)) + 1);
System.out.println(i + " and " + j);
for (int x = 0 ; x < word.length() ; x++) {
if (x == i) {
chararray[x] = word.charAt(j);
} else if (x == j) {
chararray[x] = word.charAt(i);
}
}
word = new String(chararray);
System.out.println(word);
}
output
2 and 1
aknur

Start with an character array.
Use a random generator to get 2 integers between 0 (inclusive) and length of character array (exclusive).
If i != j, swap them. (No point swapping a character with itself)
public class QuickTester {
public static void main(String[] args) {
Random rand = new Random();
char[] charArr = "elephant".toCharArray();
// Using charArr.length - 2 + 1 to exclude first and last
int i = rand.nextInt(charArr.length - 2) + 1;
int j = rand.nextInt(charArr.length - 2) + 1;
if (i != j) {
System.out.printf("Going to swap %c at position %d "
+ "with %c at position %d%n", charArr[i], i, charArr[j], j);
char temp = charArr[i];
charArr[i] = charArr[j];
charArr[j] = temp;
System.out.println("\nResult: " + new String(charArr));
}
}
}
Result:
Going to swap t at position 7 with e at position 2
Result: eltphane
Update:
Changed from charArr.length to charArr.length - 2 + 1 to exclude first and last character.

Converting Roman Numerals To Decimal

I have managed to get my code to convert most Roman numerals to its appropriate decimal value. But it doesn't work for some exceptional cases. Example : XCIX = 99 but my code prints 109.
Here is my code.
public static int romanConvert(String roman)
{
int decimal = 0;
String romanNumeral = roman.toUpperCase();
for(int x = 0;x<romanNumeral.length();x++)
{
char convertToDecimal = roman.charAt(x);
switch (convertToDecimal)
{
case 'M':
decimal += 1000;
break;
case 'D':
decimal += 500;
break;
case 'C':
decimal += 100;
break;
case 'L':
decimal += 50;
break;
case 'X':
decimal += 10;
break;
case 'V':
decimal += 5;
break;
case 'I':
decimal += 1;
break;
}
}
if (romanNumeral.contains("IV"))
{
decimal-=2;
}
if (romanNumeral.contains("IX"))
{
decimal-=2;
}
if (romanNumeral.contains("XL"))
{
decimal-=10;
}
if (romanNumeral.contains("XC"))
{
decimal-=10;
}
if (romanNumeral.contains("CD"))
{
decimal-=100;
}
if (romanNumeral.contains("CM"))
{
decimal-=100;
}
return decimal;
}

It will be good if you traverse in reverse.
public class RomanToDecimal {
public static void romanToDecimal(java.lang.String romanNumber) {
int decimal = 0;
int lastNumber = 0;
String romanNumeral = romanNumber.toUpperCase();
/* operation to be performed on upper cases even if user
enters roman values in lower case chars */
for (int x = romanNumeral.length() - 1; x >= 0 ; x--) {
char convertToDecimal = romanNumeral.charAt(x);
switch (convertToDecimal) {
case 'M':
decimal = processDecimal(1000, lastNumber, decimal);
lastNumber = 1000;
break;
case 'D':
decimal = processDecimal(500, lastNumber, decimal);
lastNumber = 500;
break;
case 'C':
decimal = processDecimal(100, lastNumber, decimal);
lastNumber = 100;
break;
case 'L':
decimal = processDecimal(50, lastNumber, decimal);
lastNumber = 50;
break;
case 'X':
decimal = processDecimal(10, lastNumber, decimal);
lastNumber = 10;
break;
case 'V':
decimal = processDecimal(5, lastNumber, decimal);
lastNumber = 5;
break;
case 'I':
decimal = processDecimal(1, lastNumber, decimal);
lastNumber = 1;
break;
}
}
System.out.println(decimal);
}
public static int processDecimal(int decimal, int lastNumber, int lastDecimal) {
if (lastNumber > decimal) {
return lastDecimal - decimal;
} else {
return lastDecimal + decimal;
}
}
public static void main(java.lang.String args[]) {
romanToDecimal("XIV");
}
}

Try this - It is simple and compact and works quite smoothly:
public static int ToArabic(string number) {
if (number == string.Empty) return 0;
if (number.StartsWith("M")) return 1000 + ToArabic(number.Remove(0, 1));
if (number.StartsWith("CM")) return 900 + ToArabic(number.Remove(0, 2));
if (number.StartsWith("D")) return 500 + ToArabic(number.Remove(0, 1));
if (number.StartsWith("CD")) return 400 + ToArabic(number.Remove(0, 2));
if (number.StartsWith("C")) return 100 + ToArabic(number.Remove(0, 1));
if (number.StartsWith("XC")) return 90 + ToArabic(number.Remove(0, 2));
if (number.StartsWith("L")) return 50 + ToArabic(number.Remove(0, 1));
if (number.StartsWith("XL")) return 40 + ToArabic(number.Remove(0, 2));
if (number.StartsWith("X")) return 10 + ToArabic(number.Remove(0, 1));
if (number.StartsWith("IX")) return 9 + ToArabic(number.Remove(0, 2));
if (number.StartsWith("V")) return 5 + ToArabic(number.Remove(0, 1));
if (number.StartsWith("IV")) return 4 + ToArabic(number.Remove(0, 2));
if (number.StartsWith("I")) return 1 + ToArabic(number.Remove(0, 1));
throw new ArgumentOutOfRangeException("something bad happened");
}

assuming the hash looks something like this
Hashtable<Character, Integer> ht = new Hashtable<Character, Integer>();
ht.put('i',1);
ht.put('x',10);
ht.put('c',100);
ht.put('m',1000);
ht.put('v',5);
ht.put('l',50);
ht.put('d',500);
then the logic gets pretty simple going by digit right to left
public static int rtoi(String num)
{
int intNum=0;
int prev = 0;
for(int i = num.length()-1; i>=0 ; i--)
{
int temp = ht.get(num.charAt(i));
if(temp < prev)
intNum-=temp;
else
intNum+=temp;
prev = temp;
}
return intNum;
}

Following your logic of reducing 2 on IX you should reduce 20 on XC 200 on CM and so on.

Less code, more efficient. Not so clearer, sorry!
public int evaluateRomanNumerals(String roman) {
return (int) evaluateNextRomanNumeral(roman, roman.length() - 1, 0);
}
private double evaluateNextRomanNumeral(String roman, int pos, double rightNumeral) {
if (pos < 0) return 0;
char ch = roman.charAt(pos);
double value = Math.floor(Math.pow(10, "IXCM".indexOf(ch))) + 5 * Math.floor(Math.pow(10, "VLD".indexOf(ch)));
return value * Math.signum(value + 0.5 - rightNumeral) + evaluateNextRomanNumeral(roman, pos - 1, value);
}

Imperative + recursive solutions with validation step and online testing
To avoid useless calculations and to make sure the roman numerals format is correct, we need to check the input with a regular expression.
String regex = "^(M{0,3})(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})$";
Explanation of the regular expression in details
Explanation of the regular expression symbols used
Differences between regex functions matches() and find()
Why is MMMCMXCIX (= 3999) the maximum value with standard roman numerals notation
Imperative solution
public static int romanToDecimal(String s) {
if (s == null || s.isEmpty() || !s.matches("^(M{0,3})(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})$"))
return -1;
final Matcher matcher = Pattern.compile("M|CM|D|CD|C|XC|L|XL|X|IX|V|IV|I").matcher(s);
final int[] decimalValues = {1000,900,500,400,100,90,50,40,10,9,5,4,1};
final String[] romanNumerals = {"M","CM","D","CD","C","XC","L","XL","X","IX","V","IV","I"};
int result = 0;
while (matcher.find())
for (int i = 0; i < romanNumerals.length; i++)
if (romanNumerals[i].equals(matcher.group(0)))
result += decimalValues[i];
return result;
}
try online | try optimized version with comments/explanation online
UPDATE: here are two clever recursive proposals from this thread I solved adding validation step
Recursive solution 1 (original answer)
public class RomanToDecimalConverter {
private static double evaluateNextRomanNumeral(String roman, int pos, double rightNumeral) {
if (pos < 0) return 0;
char ch = roman.charAt(pos);
double value = Math.floor(Math.pow(10, "IXCM".indexOf(ch))) + 5 * Math.floor(Math.pow(10, "VLD".indexOf(ch)));
return value * Math.signum(value + 0.5 - rightNumeral) + evaluateNextRomanNumeral(roman, pos - 1, value);
}
public static int evaluateRomanNumerals(String s) {
if (s == null || s.isEmpty() || !s.matches("^(M{0,3})(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})$"))
return -1;
return (int) evaluateNextRomanNumeral(s, s.length() - 1, 0);
}
}
try online
Recursive solution 2 (original answer)
public class RomanToDecimalConverter {
private static int convertRec(String s) {
if (s.isEmpty()) return 0;
if (s.startsWith("M")) return 1000 + convertRec(s.substring(1));
else if (s.startsWith("CM")) return 900 + convertRec(s.substring(2));
else if (s.startsWith("D")) return 500 + convertRec(s.substring(1));
else if (s.startsWith("CD")) return 400 + convertRec(s.substring(2));
else if (s.startsWith("C")) return 100 + convertRec(s.substring(1));
else if (s.startsWith("XC")) return 90 + convertRec(s.substring(2));
else if (s.startsWith("L")) return 50 + convertRec(s.substring(1));
else if (s.startsWith("XL")) return 40 + convertRec(s.substring(2));
else if (s.startsWith("X")) return 10 + convertRec(s.substring(1));
else if (s.startsWith("IX")) return 9 + convertRec(s.substring(2));
else if (s.startsWith("V")) return 5 + convertRec(s.substring(1));
else if (s.startsWith("IV")) return 4 + convertRec(s.substring(2));
else if (s.startsWith("I")) return 1 + convertRec(s.substring(1));
throw new IllegalArgumentException("Unexpected roman numerals");
}
public static int convert(String s) {
if (s == null || s.isEmpty() || !s.matches("^(M{0,3})(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})$"))
return -1;
return convertRec(s);
}
}
try online

// Author: Francisco Edmundo
private int translateNumber(String texto) {
int n = 0;
int numeralDaDireita = 0;
for (int i = texto.length() - 1; i >= 0; i--) {
int valor = (int) translateNumber(texto.charAt(i));
n += valor * Math.signum(valor + 0.5 - numeralDaDireita);
numeralDaDireita = valor;
}
return n;
}
private double translateNumber(char caractere) {
return Math.floor(Math.pow(10, "IXCM".indexOf(caractere))) + 5 * Math.floor(Math.pow(10, "VLD".indexOf(caractere)));
}

You can check following code. This code should work on all cases. Also it checks null or empty input and faulty input (Let's say you tried with ABXI)
import java.util.HashMap;
import org.apache.commons.lang3.StringUtils;
public class RomanToDecimal {
private HashMap<Character, Integer> map;
public RomanToDecimal() {
map = new HashMap<>();
map.put('I', 1);
map.put('V', 5);
map.put('X', 10);
map.put('L', 50);
map.put('C', 100);
map.put('D', 500);
map.put('M', 1000);
}
private int getRomanNumeralValue(char ch) {
if (map.containsKey(ch)) {
return map.get(ch);
}
else {
throw new RuntimeException("Roman numeral string contains invalid characters " + ch);
}
}
public int convertRomanToDecimal(final String pRomanNumeral) {
if (StringUtils.isBlank(pRomanNumeral)) {
throw new RuntimeException("Roman numeral string is either null or empty");
}
else {
int index = pRomanNumeral.length() - 1;
int result = getRomanNumeralValue(pRomanNumeral.charAt(index));
for (int i = index - 1; i >= 0; i--) {
if (getRomanNumeralValue(pRomanNumeral.charAt(i)) >= getRomanNumeralValue(pRomanNumeral.charAt(i + 1))) {
result = result + getRomanNumeralValue(pRomanNumeral.charAt(i));
}
else {
result = result - getRomanNumeralValue(pRomanNumeral.charAt(i));
}
}
return result;
}
}
public static void main(String... args){
System.out.println(new RomanToDecimal().convertRomanToDecimal("XCIX"));
}
}

Full version with error checking and test all valid values in both directions (and some invalid cases).
RomanNumeral.java
import java.util.ArrayList;
import java.util.TreeMap;
/**
* Convert to and from a roman numeral string
*/
public class RomanNumeral {
// used for converting from arabic number
final static TreeMap<Integer, String> mapArabic = new TreeMap<Integer, String>();
// used for converting from roman numeral
final static ArrayList<RomanDigit> mapRoman = new ArrayList<RomanDigit>();
final static int MAX_ARABIC = 3999;
static {
mapArabic.put(1000, "M");
mapArabic.put(900, "CM");
mapArabic.put(500, "D");
mapArabic.put(400, "CD");
mapArabic.put(100, "C");
mapArabic.put(90, "XC");
mapArabic.put(50, "L");
mapArabic.put(40, "XL");
mapArabic.put(10, "X");
mapArabic.put(9, "IX");
mapArabic.put(5, "V");
mapArabic.put(4, "IV");
mapArabic.put(1, "I");
mapRoman.add(new RomanDigit("M", 1000, 3, 1000));
mapRoman.add(new RomanDigit("CM", 900, 1, 90));
mapRoman.add(new RomanDigit("D", 500, 1, 100));
mapRoman.add(new RomanDigit("CD", 400, 1, 90));
mapRoman.add(new RomanDigit("C", 100, 3, 100));
mapRoman.add(new RomanDigit("XC", 90, 1, 9));
mapRoman.add(new RomanDigit("L", 50, 1, 10));
mapRoman.add(new RomanDigit("XL", 40, 1, 9));
mapRoman.add(new RomanDigit("X", 10, 3, 10));
mapRoman.add(new RomanDigit("IX", 9, 1, 0));
mapRoman.add(new RomanDigit("V", 5, 1, 1));
mapRoman.add(new RomanDigit("IV", 4, 1, 0));
mapRoman.add(new RomanDigit("I", 1, 3, 1));
}
static final class RomanDigit {
public final String numeral;
public final int value;
public final int maxConsecutive;
public final int maxNextValue;
public RomanDigit(String numeral, int value, int maxConsecutive, int maxNextValue) {
this.numeral = numeral;
this.value = value;
this.maxConsecutive = maxConsecutive;
this.maxNextValue = maxNextValue;
}
}
/**
* Convert an arabic integer value into a roman numeral string
*
* #param n The arabic integer value
* #return The roman numeral string
*/
public final static String toRoman(int n) {
if (n < 1 || n > MAX_ARABIC) {
throw new NumberFormatException(String.format("Invalid arabic value: %d, must be > 0 and < %d", n, MAX_ARABIC));
}
int leDigit = mapArabic.floorKey(n);
//System.out.println("\t*** floor of " + n + " is " + leDigit);
if (n == leDigit) {
return mapArabic.get(leDigit);
}
return mapArabic.get(leDigit) + toRoman(n - leDigit);
}
/**
* Convert a roman numeral string into an arabic integer value
* #param s The roman numeral string
* #return The arabic integer value
*/
public final static int toInt(String s) throws NumberFormatException {
if (s == null || s.length() == 0) {
throw new NumberFormatException("Invalid roman numeral: a non-empty and non-null value must be given");
}
int i = 0;
int iconsecutive = 0; // number of consecutive same digits
int pos = 0;
int sum = 0;
RomanDigit prevDigit = null;
while (pos < s.length()) {
RomanDigit d = mapRoman.get(i);
if (!s.startsWith(mapRoman.get(i).numeral, pos)) {
i++;
// this is the only place we advance which digit we are checking,
// so if it exhausts the digits, then there is clearly a digit sequencing error or invalid digit
if (i == mapRoman.size()) {
throw new NumberFormatException(
String.format("Invalid roman numeral at pos %d: invalid sequence '%s' following digit '%s'",
pos, s.substring(pos), prevDigit != null ? prevDigit.numeral : ""));
}
iconsecutive = 0;
continue;
}
// we now have the match for the next roman numeral digit to check
iconsecutive++;
if (iconsecutive > d.maxConsecutive) {
throw new NumberFormatException(
String.format("Invalid roman numeral at pos %d: more than %d consecutive occurences of digit '%s'",
pos, d.maxConsecutive, d.numeral));
}
// invalid to encounter a higher digit sequence than the previous digit expects to have follow it
// (any digit is valid to start a roman numeral - i.e. when prevDigit == null)
if (prevDigit != null && prevDigit.maxNextValue < d.value) {
throw new NumberFormatException(
String.format("Invalid roman numeral at pos %d: '%s' cannot follow '%s'",
pos, d.numeral, prevDigit.numeral));
}
// good to sum
sum += d.value;
if (sum > MAX_ARABIC) {
throw new NumberFormatException(
String.format("Invalid roman numeral at pos %d: adding '%s' exceeds the max value of %d",
pos, d.numeral, MAX_ARABIC));
}
pos += d.numeral.length();
prevDigit = d;
}
return sum;
}
}
Main.java
public class Main {
public static void main(String[] args) {
System.out.println("TEST arabic integer => roman numeral string");
for (int i = 0; i<= 4000; i++) {
String s;
try {
s = RomanNumeral.toRoman(i);
}
catch(NumberFormatException ex) {
s = ex.getMessage();
}
System.out.println(i + "\t =\t " + s);
}
System.out.println("TEST roman numeral string => arabic integer");
for (int i = 0; i<= 4000; i++) {
String s;
String msg;
try {
s = RomanNumeral.toRoman(i);
int n = testToInt(s);
assert(i == n); // ensure it is reflexively converting
}
catch (NumberFormatException ex) {
System.out.println(ex.getMessage() + "\t =\t toInt() skipped");
}
}
testToInt("MMMM");
testToInt("XCX");
testToInt("CDC");
testToInt("IVI");
testToInt("XXC");
testToInt("CCD");
testToInt("MDD");
testToInt("DD");
testToInt("CLL");
testToInt("LL");
testToInt("IIX");
testToInt("IVX");
testToInt("IIXX");
testToInt("XCIX");
testToInt("XIWE");
// Check validity via a regexp for laughs
String s = "IX";
System.out.println(s + " validity is " + s.matches("M{0,3}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})"));
}
private final static int testToInt(String s) {
String msg;
int n=0;
try {
n = RomanNumeral.toInt(s);
msg = Integer.toString(n);
}
catch(NullPointerException | NumberFormatException ex) {
msg = ex.getMessage();
}
System.out.println(s + "\t =\t " + msg);
return n;
}
}
Output
TEST arabic integer => roman numeral string
0 = Invalid arabic value: 0, must be > 0 and < 3999
1 = I
2 = II
3 = III
4 = IV
5 = V
6 = VI
7 = VII
8 = VIII
9 = IX
10 = X
... [snip] ...
3988 = MMMCMLXXXVIII
3989 = MMMCMLXXXIX
3990 = MMMCMXC
3991 = MMMCMXCI
3992 = MMMCMXCII
3993 = MMMCMXCIII
3994 = MMMCMXCIV
3995 = MMMCMXCV
3996 = MMMCMXCVI
3997 = MMMCMXCVII
3998 = MMMCMXCVIII
3999 = MMMCMXCIX
4000 = Invalid arabic value: 4000, must be > 0 and < 3999
TEST roman numeral string => arabic integer
Invalid arabic value: 0, must be > 0 and < 3999 = toInt() skipped
I = 1
II = 2
III = 3
IV = 4
V = 5
VI = 6
VII = 7
VIII = 8
IX = 9
X = 10
... [snip] ...
MMMCMLXXXVIII = 3988
MMMCMLXXXIX = 3989
MMMCMXC = 3990
MMMCMXCI = 3991
MMMCMXCII = 3992
MMMCMXCIII = 3993
MMMCMXCIV = 3994
MMMCMXCV = 3995
MMMCMXCVI = 3996
MMMCMXCVII = 3997
MMMCMXCVIII = 3998
MMMCMXCIX = 3999
Invalid arabic value: 4000, must be > 0 and < 3999 = toInt() skipped
MMMM = Invalid roman numeral at pos 3: more than 3 consecutive occurences of digit 'M'
XCX = Invalid roman numeral at pos 2: 'X' cannot follow 'XC'
CDC = Invalid roman numeral at pos 2: 'C' cannot follow 'CD'
IVI = Invalid roman numeral at pos 2: 'I' cannot follow 'IV'
XXC = Invalid roman numeral at pos 2: invalid sequence 'C' following digit 'X'
CCD = Invalid roman numeral at pos 2: invalid sequence 'D' following digit 'C'
MDD = Invalid roman numeral at pos 2: more than 1 consecutive occurences of digit 'D'
DD = Invalid roman numeral at pos 1: more than 1 consecutive occurences of digit 'D'
CLL = Invalid roman numeral at pos 2: more than 1 consecutive occurences of digit 'L'
LL = Invalid roman numeral at pos 1: more than 1 consecutive occurences of digit 'L'
IIX = Invalid roman numeral at pos 2: invalid sequence 'X' following digit 'I'
IVX = Invalid roman numeral at pos 2: invalid sequence 'X' following digit 'IV'
IIXX = Invalid roman numeral at pos 2: invalid sequence 'XX' following digit 'I'
XCIX = 99
XIWE = Invalid roman numeral at pos 2: invalid sequence 'WE' following digit 'I'
IX validity is true

Lets look at this problem with 3 different scenarios
Scenario 1:
When we see a pattern like the following
'IIIIII' or 'XXXXX' or 'CCCC'
where all characters as the same: We add the value of each characters in the pattern
'IIIIII' gives us '6'
'XXXXX' gives us '50'
'CCCC' gives us '400'
Scenario 2:
When we see any 2 consecutive characters different where first is smaller in value then the 2nd
'IX' or 'XC'
We subtract the value of first from second e.g.
second:'X' gives us '10'
first: 'I' gives us '1'
second - first : 10 - 1 = 9
Scenario 3:
When we see a any 2 consecutive characters different where first is greater in value then the second
'XI' or 'CX'
We add first and second e.g.
second:'I' gives us '10'
first: 'X' gives us '1'
first + second : 10 + 1 = 11
Now we can find the result if we do this recursively.
Here is the java implementation :
//An array to be used for faster comparisons and reading the values
private int[] letters26 = new int[26];
private void init () {
letters26['I' - 'A'] = 1;
letters26['V' - 'A'] = 5;
letters26['X' - 'A'] = 10;
letters26['L' - 'A'] = 50;
letters26['C' - 'A'] = 100;
letters26['D' - 'A'] = 500;
letters26['M' - 'A'] = 1000;
}
public int convertRomanToInteger(String s) {
//Initialize the array
init();
return _convertRomanToInteger(s.toCharArray(), 0);
}
//Recursively calls itself as long as 2 consecutive chars are different
private int _convertRomanToInteger(char[] s, int index) {
int ret = 0;
char pre = s[index];//Char from the given index
ret = _getValue(pre);
//Iterate through the rest of the string
for (int i = index + 1; i < s.length; i++) {
if (compare(s[i], s[i - 1]) == 0) {
//Scenario 1:
//If 2 consecutive chars are similar, just add them
ret += _getValue(s[i]);
} else if (compare(s[i], s[i - 1]) > 0) {
//Scenario 2:
//If current char is greater than the previous e.g IX ('I' s[i - 1] and 'X' s[i - 1])
//We need to calculate starting from 'i' and subtract the calculation ('ret')
//done so far in current call
return _convertRomanToInteger(s, i) - ret;
} else {
//Scenario 3:
//If current char is smaller than the previous e.g XI ('X' s[i - 1] and 'I' s[i - 1])
//We need to calculate starting from 'i' and subtract the result
//from the calculation done so far in current call
return ret + _convertRomanToInteger(s, i);
}
}
return ret;
}
//Helper function for comparison
private int compare(char a, char b) {
return letters26[Character.toUpperCase(a) - 'A']
- letters26[Character.toUpperCase(b) - 'A'];
}
private int _getValue(char c) {
return letters26[Character.toUpperCase(c) - 'A'];
}

Despite the fact that a lot of solutions have been already proposed.
I guess that the following one will be short and clear:
public class RomanToInteger {
public static int romanToInt(String roman) {
final Map<Character, Integer> map = Map.of('I', 1,
'V', 5,
'X', 10,
'L', 50,
'C', 100,
'D', 500,
'M', 1000);
int result = 0;
for (int index = 0; index < roman.length(); index++) {
final int curNumber = map.get(roman.charAt(index));
if (index > 0 && curNumber > map.get(roman.charAt(index - 1))) {
// add current number & remove previous one twice:
// first: we add it before (when it was current number) and removing it for this current number
// second: for correct conversion to roman numbers
result += curNumber - 2 * map.get(roman.charAt(index - 1));
} else {
result += curNumber;
}
}
System.out.printf("%8s -> %4d\n", roman, result);
return result;
}
public static void main(String[] args) {
String[] romans = {"I", "II", "III", "V", "X", "XIV", "XVII", "XX", "XXV",
"XXX", "XXXVIII", "XLIX", "LXIII", "LXXXI", "XCVII", "XCVIII", "XCIX",
"C", "CI", "CCXLVIII", "CCLIII", "DCCXCIX", "MCCCXXV", "MCM", "MM",
"MMCDLVI", "MDCCXV"};
final Instant startTimeIter = Instant.now();
for (String roman : romans) {
romanToInt(roman);
}
final Instant endTimeIter = Instant.now();
System.out.printf("Elapsed time: %d ms\n", Duration.between(startTimeIter, endTimeIter).toMillis());
}
}
Output:
I -> 1
II -> 2
III -> 3
...
MMCDLVI -> 2456
MDCCXV -> 1715
Elapsed time: 101 ms
Logic is quite simple we just go from left to right through roman literal:
if a literal isn't first and the previous one is lower -> we have to add current number to total & extract the previous number twice. First time what we added this at the previous step (when it was current number). Second because it is a rule of roman literals conversion (like IX literal).
otherwise, just add it to the total result.

How about this?
int getdec(const string& input)
{
int sum=0; char prev='%';
for(int i=(input.length()-1); i>=0; i--)
{
if(value(input[i])<sum && (input[i]!=prev))
{ sum -= value(input[i]);
prev = input[i];
}
else
{
sum += value(input[i]);
prev = input[i];
}
}
return sum;
}

Supposing Well-formed Roman numbers:
private static int totalValue(String val)
{
String aux=val.toUpperCase();
int sum=0, max=aux.length(), i=0;
while(i<max)
{
if ((i+1)<max && valueOf(aux.charAt(i+1))>valueOf(aux.charAt(i)))
{
sum+=valueOf(aux.charAt(i+1)) - valueOf(aux.charAt(i));
i+=2;
}
else
{
sum+=valueOf(aux.charAt(i));
i+=1;
}
}
return sum;
}
private static int valueOf(Character c)
{
char aux = Character.toUpperCase(c);
switch(aux)
{
case 'I':
return 1;
case 'V':
return 5;
case 'X':
return 10;
case 'L':
return 50;
case 'C':
return 100;
case 'D':
return 500;
case 'M':
return 1000;
default:
return 0;
}
}

what about this conversion. no switch, no case at all...
P.S. : I use this script from a bash shell
import sys
def RomanToNum(r):
return {
'I': 1,
'V': 5,
'X': 10,
'L': 50,
'C': 100,
'D': 500,
'M': 1000,
}[r]
#
#
#
EOF = "<"
Roman = sys.argv[1].upper().strip()+EOF
num = 0
i = 0
while True:
this = Roman[i]
if this == EOF:
break
n1 = RomanToNum(this)
next = Roman[i+1]
if next == EOF:
n2 = 0
else:
n2 = RomanToNum(next)
if n1 < n2:
n1 = -1 * n1
num = num + n1
i = i + 1
print num

public class RomInt {
String roman;
int val;
void assign(String k)
{
roman=k;
}
private class Literal
{
public char literal;
public int value;
public Literal(char literal, int value)
{
this.literal = literal;
this.value = value;
}
}
private final Literal[] ROMAN_LITERALS = new Literal[]
{
new Literal('I', 1),
new Literal('V', 5),
new Literal('X', 10),
new Literal('L', 50),
new Literal('C', 100),
new Literal('D', 500),
new Literal('M', 1000)
};
public int getVal(String s) {
int holdValue=0;
for (int j = 0; j < ROMAN_LITERALS.length; j++)
{
if (s.charAt(0)==ROMAN_LITERALS[j].literal)
{
holdValue=ROMAN_LITERALS[j].value;
break;
} //if()
}//for()
return holdValue;
} //getVal()
public int count()
{
int count=0;
int countA=0;
int countB=0;
int lastPosition = 0;
for(int i = 0 ; i < roman.length(); i++)
{
String s1 = roman.substring(i,i+1);
int a=getVal(s1);
countA+=a;
}
for(int j=1;j<roman.length();j++)
{
String s2= roman.substring(j,j+1);
String s3= roman.substring(j-1,j);
int b=getVal(s2);
int c=getVal(s3);
if(b>c)
{
countB+=c;
}
}
count=countA-(2*countB);
return count;
}
void disp()
{
int result=count();
System.out.println("Integer equivalent of "+roman+" = " +result);
}
} //RomInt---BLC

I find the following approach a very intuitive one:
public void makeArray(String romanNumeral){
int[] numberArray = new int[romanNumeral.length()];
for(int i=0; i<romanNumeral.length();i++){
char symbol = romanNumeral.charAt(i);
switch(symbol){
case 'I':
numberArray[i] = 1;
break;
case 'V':
numberArray[i] = 5;
break;
case 'X':
numberArray[i] = 10;
break;
case 'L':
numberArray[i] = 50;
break;
case 'C':
numberArray[i] = 100;
break;
case 'D':
numberArray[i] = 500;
break;
case 'M':
numberArray[i] = 1000;
break;
}
}
calculate(numberArray);
}
public static void calculate(int[] numberArray){
int theNumber = 0;
for(int n=0;n<numberArray.length;n++){
if(n !=numberArray.length-1 && numberArray[n] < numberArray[n+1]){
numberArray[n+1] = numberArray[n+1] - numberArray[n];
numberArray[n] = 0;
}
}
for(int num:numberArray){
theNumber += num;
}
System.out.println("Converted number: " + theNumber);
}

//Bet no one has a smaller and easier logic than this........Open CHALLENGE!!!!!!!
import java.io.*;
class Convertion_practical_q2
{
void Decimal()throws IOException //Smaller code for convertion from roman to decimal
{
DataInputStream in=new DataInputStream(System.in);
System.out.println("Enter the number");
String num=in.readLine();
char pos[]={'0','I','V','X','L','C','D','M'};
int l1=7; //l1 is size of pos array
String v[]={"","1","5","10","50","100","500","1000"};
int l=num.length();
int p=0,p1=0,sum=0;
for(int i=l-1;i>=0;i--)
{
char ch=num.charAt(i);
for(int j=1;j<=l1;j++)
{
if(ch==pos[j])
p=j;
}
if(p>=p1)
sum+=Integer.parseInt(v[p]);
else
sum-=Integer.parseInt(v[p]);
//System.out.println("sum ="+sum+"\np="+p+"\np1="+p1);
p1=p;
}
System.out.println(sum);
}
}

Just got it working in Java, nice job guys.
public int getDecimal (String roman) {
int decimal = 0;
int romanNumber = 0;
int prev = 0;
for (int i = roman.length()-1; i >= 0; i--){
romanNumber = hashRomans.get(roman.charAt(i));
if(romanNumber < decimal && romanNumber != prev ){
decimal -= romanNumber;
prev = romanNumber;
} else {
decimal += romanNumber;
prev = romanNumber;
}
}
return decimal;
}

This should work:
import java.io.*;
import java.util.Scanner;
enum RomanToNumber {
i(1), v(5), x(10), l(50), c(100); int value;
RomanToNumber (int p){value = p;}
int getValue(){return value;}
}
public class RomanToInteger {
public static void main(String[] args){
RomanToNumber n;
System.out.println( "Type a valid roman number in lower case" );
String Str = new String(new Scanner(System.in).nextLine());
int n1 = 0, theNo = 0, len = Str.length();
int[] str2No = new int [len];
for(int i=0; i < len; i++){
n = RomanToNumber.valueOf(Str.substring(n1, ++n1));
str2No[i] = n.getValue();
}
for(int j = 0; j < (len-1); j++){
if( str2No[j] >= str2No[j+1] ){ theNo += str2No[j]; }
else{ theNo -= str2No[j]; }
}
System.out.println( theNo += str2No[len-1] );
}
}

Since most of the answers here are in Java, I'm posting the answer in C++ (since I am bored right now and nothing more productive to do :) But please, no downvotes except if code is wrong.
Known-issues = will not handle overflow
Code:
#include <unordered_map>
int convert_roman_2_int(string& str)
{
int ans = 0;
if( str.length() == 0 )
{
return ans;
}
std::unordered_map<char, int> table;
table['I'] = 1;
table['V'] = 5;
table['X'] = 10;
table['L'] = 50;
table['C'] = 100;
table['D'] = 500;
table['M'] = 1000;
ans = table[ str[ str.length() - 1 ] ];
for( int i = str.length() - 2; i >= 0; i--)
{
if(table[ str[i] ] < table[ str[i+1] ] )
{
ans -= table[ str[i] ];
}
else
{
ans += table[ str[i] ];
}
}
return ans;
}
// test code
void run_test_cases_convert_roman_to_int()
{
string roman = "VIII";
int r = convert_roman_2_int(roman);
cout << roman << " in int is " << r << endl << flush;
roman = "XX";
r = convert_roman_2_int(roman);
cout << roman << " in int is " << r << endl << flush;
roman = "CDX"; //410
r = convert_roman_2_int(roman);
cout << roman << " in int is " << r << endl << flush;
roman = "MCMXC"; //1990
r = convert_roman_2_int(roman);
cout << roman << " in int is " << r << endl << flush;
roman = "MMVIII"; //2008
r = convert_roman_2_int(roman);
cout << roman << " in int is " << r << endl << flush;
roman = "MDCLXVI"; //1666
r = convert_roman_2_int(roman);
cout << roman << " in int is " << r << endl << flush;
}

this is very basic implementation of what you asked and working for all the test cases. If you find anything wrong in it than do mention it,i will try to make it correct also the code is in C++ .
int RomanToInt(string s){
int len = s.length();
map<char,int>mp;
int decimal = 0;
mp['I'] = 1;mp['V'] = 5;mp['X'] = 10;
mp['L'] = 50;mp['C'] = 100;mp['D'] = 500;mp['M'] = 1000;
for(int i = 0; i < len ;i++){
char cur = s[i],fast = s[i+1];
int cur_val = mp[cur],fast_val = mp[fast];
if(cur_val < fast_val){
decimal = decimal - cur_val;
}else{
decimal += cur_val;
}
}
return decimal;
}

public static int convertFromRoman(String romanNumeral)
{
Character[] rnChars = { 'M', 'D', 'C', 'L', 'X', 'V', 'I' };
int[] rnVals = { 1000, 500, 100, 50, 10, 5, 1 };
HashMap<Character, Integer> valueLookup = new HashMap<Character, Integer>();
for (int i = 0; i < rnChars.length;i++)
valueLookup.put(rnChars[i], rnVals[i]);
int retVal = 0;
for (int i = 0; i < romanNumeral.length();i++)
{
int addVal = valueLookup.get(romanNumeral.charAt(i));
retVal += i < romanNumeral.length()-1 &&
addVal < valueLookup.get(romanNumeral.charAt(i+1))?
-addVal:addVal;
}
return retVal;
}

This is a modest variation of the recursive algorithm suggested by Sahtiel:
public static int toArabic(String number) throws Exception {
String[] letras = {"M","CM","D","CD","C","XC","L","XL","X", "IX","V","IV","I"};
int[] valores = {1000,900,500,400,100,90,50,40,10,9,5,4,1};
// here we can do even more business validations like avoiding sequences like XXXXM
if (number==null || number.isEmpty()) {
return 0;
}
for(int i=0; i<letras.length; i++) {
if (number.startsWith(letras[i])) {
return valores[i] + toArabic(number.substring(letras[i].length()));
}
}
throw new Exception("something bad happened");
}
It uses less than 10 effective lines of code.

This scala solution might be useful:
class RomanNumberConverter {
private val toArabic = Map('I' -> 1, 'V' -> 5, 'X' -> 10, 'L' -> 50, 'C' -> 100, 'D' -> 500, 'M' -> 1000)
// assume that correct roman number is provided
def convert(romanNumber: String): Int = {
def convert(rn: StringBuilder, lastDecimal: Int, acc: Int): Int = {
if (rn.isEmpty) acc
else {
val thisDecimal = toArabic(rn.head)
if (thisDecimal > lastDecimal) convert(rn.tail, thisDecimal, acc + thisDecimal - lastDecimal - lastDecimal)
else convert(rn.tail, thisDecimal, acc + thisDecimal)
}
}
val sb = new StringBuilder(romanNumber)
convert(sb.tail, toArabic(sb.head), toArabic(sb.head))
}
}

Solution using tail recursion:
import java.util.LinkedHashMap;
public class RomanNumber {
private final static LinkedHashMap<String, Integer> roman2number = new LinkedHashMap<>(); // preserve key order
static {
roman2number.put("M", 1000);
roman2number.put("CM", 900);
roman2number.put("D", 500);
roman2number.put("CD", 400);
roman2number.put("C", 100);
roman2number.put("XC", 90);
roman2number.put("L", 50);
roman2number.put("XL", 40);
roman2number.put("X", 10);
roman2number.put("IX", 9);
roman2number.put("V", 5);
roman2number.put("IV", 4);
roman2number.put("I", 1);
}
public final static Integer toDecimal(String roman) {
for (String key : roman2number.keySet()) {
if (roman.startsWith(key)) {
if (roman.equals(key)) {
return roman2number.get(key);
}
return roman2number.get(key) + toDecimal(roman.substring(key.length()));
}
}
return 0;
}
}
Testing:
import junitparams.JUnitParamsRunner;
import org.junit.Test;
import org.junit.runner.RunWith;
import junitparams.Parameters;
import static org.junit.Assert.assertTrue;
#RunWith(JUnitParamsRunner.class)
public class RomanNumberTest {
#Test
#Parameters({ "1|I", "2|II", "3|III", "4|IV", "5|V", "6|VI", "7|VII", "8|VIII", "9|IX", "10|X",
"11|XI", "12|XII", "13|XIII", "14|XIV", "15|XV", "16|XVI", "17|XVII", "18|XVIII", "19|XIX",
"20|XX", "50|L", "53|LIII", "57|LVII", "40|XL", "49|XLIX", "59|LIX", "79|LXXIX", "100|C", "90|XC", "99|XCIX",
"200|CC", "500|D", "499|CDXCIX", "999|CMXCIX", "2999|MMCMXCIX", "3999|MMMCMXCIX"
})
public void forRomanReturnsNumber(int number, String roman) {
assertTrue(roman + "->" + number, RomanNumber.toDecimal(roman) == (number));
}
}

With only two if-statements:
Check the match of first character on priority and if not matching, match the single first character.
public class Solution {
private static TreeMap<String, Integer> numeralsAndVal;
static{
numeralsAndVal = new TreeMap<>();
numeralsAndVal.put("M", 1000);
numeralsAndVal.put("CM", 900);
numeralsAndVal.put("D", 500);
numeralsAndVal.put("CD", 400);
numeralsAndVal.put("C", 100);
numeralsAndVal.put("XC", 90);
numeralsAndVal.put("L", 50);
numeralsAndVal.put("XL", 40);
numeralsAndVal.put("X", 10);
numeralsAndVal.put("IX", 9);
numeralsAndVal.put("V", 5);
numeralsAndVal.put("IV", 4);
numeralsAndVal.put("I", 1);
}
public int romanToInt(String A) {
if (A.length() == 0) return 0;
if (A.length() >= 2 && numeralsAndVal.containsKey(A.substring(0, 2))){
return numeralsAndVal.get(A.substring(0, 2)) + romanToInt(A.substring(2));
}
return numeralsAndVal.get(A.substring(0, 1)) + romanToInt(A.substring(1));
}
}

For converting TO arabic, it works as follows:
grab the rightmost unchecked character and convert it to its numeric value using the map. If this value is less than the "top" value we've seen so far, then it needs to be subtracting (e.g. IV needs to be 4, not 6), otherwise we add the value to the total and make it the new top value we've seen so far. So, for instance, XIX works like follows: 'X' is checked and comes out to be value of 10. That is higher than the "top" value seen so far, so it becomes "top" and we add 10 to the value. Then go to the next character and "I" is < 10, so subtract
public class RomanNumeral {
private final Map<Integer, String> arabicToRoman = new LinkedHashMap<Integer, String>();
private final Map<String, Integer> romanToArabic = new LinkedHashMap<String, Integer>();
public RomanNumeral() {
arabicToRoman.put(10, "X");
arabicToRoman.put(9, "IX");
arabicToRoman.put(5, "V");
arabicToRoman.put(4, "IV");
arabicToRoman.put(1, "I");
romanToArabic.put("X", 10);
romanToArabic.put("V", 5);
romanToArabic.put("I", 1);
}
public String convertToRomanNumeral(int number) {
String result = "";
for (Integer i : arabicToRoman.keySet()) {
while (number >= i) {
result += arabicToRoman.get(i);
number -= i;
}
}
return result;
}
public String convertToArabicNumber(String romanNumeral) {
int result = 0;
int top = 0;
for (int i = romanNumeral.length() - 1; i >= 0; i--) {
char current = romanNumeral.charAt(i);
int value = romanToArabic.get("" + current);
if (value < top) {
result -= value;
} else {
result += value;
top = value;
}
}
return "" + result;
}
}

An easy solution for input in the range from 1 to 3999.
private static final Map<String, Integer> MAPPING = new HashMap<>() {{
put("I", 1);
put("V", 5);
put("X", 10);
put("L", 50);
put("C", 100);
put("D", 500);
put("M", 1000);
put("IV", 4);
put("IX", 9);
put("XL", 40);
put("XC", 90);
put("CD", 400);
put("CM", 900);
}};
public static int romanToInt(String s) {
int sum = 0;
for (int i = 0; i < s.length(); i++) {
if (i < s.length() - 1 && MAPPING.containsKey(String.valueOf(s.charAt(i)) + s.charAt(i + 1))) {
sum += MAPPING.get(String.valueOf(s.charAt(i)) + s.charAt(i + 1));
i++;
} else {
sum += MAPPING.get(String.valueOf(s.charAt(i)));
}
}
return sum;
}

The problem here is that numbers like 9 are presented in the following way
IX = 9
if we take the literal conversion by getting the sum for each number than our program would give us the following sum
I = 1
X = 10
X + I = 11
which is not correct.
We can overcome this if we simply manipulate the Roman String.
In example:
IX = 9 would be VIIII which is 5 + 1 + 1 + 1 + 1
Therefore if we replace all the special cases in the given String, we can calculate the sum for each character.
Hence;
class Solution {
public int romanToInt(String s) {
// Key Value pairs
Map<Character, Integer> map = new HashMap<>();
map.put('I', 1);
map.put('V', 5);
map.put('X', 10);
map.put('L', 50);
map.put('C', 100);
map.put('D', 500);
map.put('M', 1000);
// convert these to single characters
s = s.replace("IV","IIII");
s = s.replace("IX","VIIII");
s = s.replace("XL","XXXX");
s = s.replace("XC","LXXXX");
s = s.replace("CD","CCCC");
s = s.replace("CM","DCCCC");
// total sum to be returned
int sum = 0;
// convert s to c character array
char[] c = s.toCharArray();
// iterate through c array and sum the numbers
for(Character ch : c){
sum = sum + map.get(ch);
}
return sum;
}
}

Here is my one...
import java.util.Scanner;
class Solution {
public static void main(String args[]) {
Scanner sc = new Scanner(System.in);
String num;
// System.out.println("enter the number you want to convert from roman to integer.");
num = "D";
System.out.println("length of string is " + num.length());
System.out.println("the integer number is :" + romanToInt(num) );
}
public static int romanToInt(String s) {
char I,V, X, L,C, D, M;
char c1,c3, c2 = s.charAt(0);
// System.out.println("the c2 is : " + (int)c2);
int num = 0, num1 = 0;
int j =0, k = 0, temp = 0;
if (c2 == 'I') {
k = (int)c2 - 72;
System.out.println("k is I" + k);
} else if(c2 == 'V') {
k = (int)c2 - 81;
System.out.println("K is V" + k);
// return 86 - 81;
} else if (c2 == 'X') {
k = (int)c2 - 78;
System.out.println("K is X" + k);
} else if (c2 == 'L') {
k = (int)c2 - 26;
System.out.println("K is L" + k);
} else if (c2 == 'C') {
k = (int)c2 + 33;
System.out.println("K is C" + k);
} else if (c2 == 'D') {
k = (int)c2 + 432;
System.out.println("K is D" + k);
} else if ( c2 == 'M') {
k = (int)c2 + 923;
System.out.println("K is M" + k);
}
if (s.length() == 1){
num = k;
} else {
for(int i = 1; i<= s.length()-1 ; i++) {
System.out.println("i is : " + i);
c1 = s.charAt(i);
if (i == s.length() - 1) {
temp = 0;
} else {
c3 = s.charAt(i+1);
if (c3 == 'I') {
temp = (int)c3 - 72;
System.out.println("temp is I " + temp);
} else if(c3 == 'V') {
temp = (int)c3 - 81;
System.out.println("temp is I " + temp);
// return 86 - 81;
} else if (c3 == 'X') {
temp = (int)c3 - 78;
System.out.println("temp is I " + temp);
} else if (c3 == 'L') {
temp = (int)c3 - 26;
System.out.println("temp is I " + temp);
} else if (c3 == 'C') {
temp = (int)c3 + 33;
System.out.println("temp is I " + temp);
} else if (c3 == 'D') {
temp = (int)c3 + 432;
System.out.println("temp is I " + temp);
} else if ( c3 == 'M') {
temp = (int)c3 + 923;
System.out.println("temp is I " + temp);
}
}
if (c1 == 'I') {
j = (int)c1 - 72;
System.out.println("j is I " + j);
} else if(c1 == 'V') {
j = (int)c1 - 81;
System.out.println("j is V " + j);
// return 86 - 81;
} else if (c1 == 'X') {
j = (int)c1 - 78;
System.out.println("j is X " + j);
} else if (c1 == 'L') {
j = (int)c1 - 26;
System.out.println("j is L " + j);
} else if (c1 == 'C') {
j = (int)c1 + 33;
System.out.println("j is C " + j);
} else if (c1 == 'D') {
j = (int)c1 + 432;
System.out.println("j is D " + j);
} else if ( c1 == 'M') {
j = (int)c1 + 923;
System.out.println("j is M " + j);
}
if ( k < j && j>temp ) {
k = j - k ;
num = num + k;
} else if (j==k || j<k || j<temp){
num = num + k ;
// k = j;
}
if (j>k ) {
k = temp;
i += 1;
if (i == s.length()-1) {
num = num + k;
}
} else {
k = j;
if (i == s.length()-1) {
num = num + k;
}
}
}
}
return num;
}
}