# Integer to Roman numeral

The task is to take a decimal integer as input and print the corresponding Roman numeral with capital letters.

The program must handle all positive integer numbers between 1 and 1000. Input can be assumed to be correct and no error handling is necessary.

Example data:

```
Input Output
1 I
8 VIII
9 IX
10 X
44 XLIV
666 DCLXVI
800 DCCC
888 DCCCLXXXVIII
900 CM
999 CMXCIX
1000 M
```

This is code golf, shortest source code wins.

Example code of a simple iterative algorithm in C

```
#include <stdio.h>
#include <string.h>
char* romanize (int val)
{
static char result[99]="";
typedef struct
{
int val;
char* str;
} roman_t;
const roman_t lookup[] =
{
{ 1, "I" },
{ 4, "IV" },
{ 5, "V" },
{ 9, "IX" },
{ 10, "X" },
{ 40, "XL" },
{ 50, "L" },
{ 90, "XC" },
{ 100, "C" },
{ 400, "CD" },
{ 500, "D", },
{ 900, "CM" },
{ 1000, "M" }
};
const size_t supported_items = sizeof lookup / sizeof *lookup;
for(size_t i=supported_items-1; i>=0 && val!=0; i--)
{
while(val >= lookup[i].val)
{
strcat(result, lookup[i].str);
val -= lookup[i].val;
}
}
return result;
}
int main (void)
{
int input;
scanf("%d", &input);
puts(romanize(input));
}
```

[JavaScript (Node.js)], 147 by …

~1mo ago

Perl 5 `-pMRoman`, 10 bytes (1 …

~1mo ago

C (compliant), 197 198 bytes. …

~1mo ago

## 3 answers

# JavaScript (Node.js), 147 bytes

```
f=(a,b='IVXLCDM',g=Math.log10(a)<<1,c=10**(g/2),e=a/c|0,i=b[g+1],j=a=>b[g].repeat(a))=>a?(e<4?j(e):e<6?j(5-e)+i:e<9?i+j(e-5):j(1)+b[g+2])+f(a%c):''
```

#### 0 comments

# C (compliant), 197 ~~198~~ bytes.

Golfed version of the function in the question, using X macros:

```
#define L Y(1000,M)Y(900,CM)Y(500,D)Y(400,CD)Y(100,C)Y(90,XC)Y(50,L)Y(40,XL)Y(10,X)Y(9,IX)Y(5,V)Y(4,IV)Y(1,I)return r;
#define Y(n,R)for(;v>=n;v-=n)strcat(r,#R);
char r[99];char*f(int v){L}
```

This function abuses static initialization so it can only be called once. By adding `*r=0;`

, we can call it multiple times during testing, which is why the code in the TIO link below looks slightly longer.

It can be minimized a bit further with non-compliant gcc tricks.

#### 1 comment

Amusingly, this is "Y macros", not "X macros"... because X didn't sit well with Roman numerals. I got trouble in one version of the code that used `X(10,X)`

:)

## 0 comments