# Forum: Ruby Calculating roman numerals

Announcement (2017-05-07): www.ruby-forum.com is now read-only since I unfortunately do not have the time to support and maintain the forum any more. Please see rubyonrails.org/community and ruby-lang.org/en/community for other Rails- und Ruby-related community platforms.
on 2007-01-02 22:36
```Ok, I'm having trouble with another exercise in this book. It has
asked me to write a program that will calculate the old school roman
numeral value for a given modern number (up to the thousands). To
clarify, old school roman numerals didn't do the subtraction thing, so
4 is IIII, nine is VIIII, etc. I've played around with a few options
using division, loops and modulus, but can't seem to get my brain
around the problem...any suggestions?

Shiloh```
on 2007-01-02 22:44
```[Shiloh Madsen <shiloh.madsen@gmail.com>, 2007-01-02 22.35 CET]
> Ok, I'm having trouble with another exercise in this book. It has
> asked me to write a program that will calculate the old school roman
> numeral value for a given modern number (up to the thousands). To
> clarify, old school roman numerals didn't do the subtraction thing, so
> 4 is IIII, nine is VIIII, etc. I've played around with a few options
> using division, loops and modulus, but can't seem to get my brain
> around the problem...any suggestions?

A hint: Try loops and subtractions.

Good luck.```
on 2007-01-02 22:58
```Thank you. That gave me an entirely new way to look at this. I am
working on one solution now, but I will try that next. If they both
work I'll post both as I am curious which will be more efficient.```
on 2007-01-02 23:30
```Shiloh Madsen wrote:
> --
> It is better to rule in hell than serve in heaven.

class Integer
def to_roman
"I =1       V =5      X = 10       L = 50
C = 100    D = 500   M = 1000".
scan( / ([A-Z]) \s *= \s* (\d+) /x ).
map{|letter,val|  [ letter, val.to_i ] }.
sort_by{|a| -a.last}.
inject( [ "", self ] ){|roman, pair|
[ roman.first + pair.first * (roman.last / pair.last),
roman.last % pair.last ] }.
first
end
end```
on 2007-01-03 13:46
```On Wed, Jan 03, 2007 at 06:35:02AM +0900, Shiloh Madsen wrote:
} Ok, I'm having trouble with another exercise in this book. It has
} asked me to write a program that will calculate the old school roman
} numeral value for a given modern number (up to the thousands). To
} clarify, old school roman numerals didn't do the subtraction thing, so
} 4 is IIII, nine is VIIII, etc. I've played around with a few options
} using division, loops and modulus, but can't seem to get my brain
} around the problem...any suggestions?

We recently had a golfing competition at work on going the other
direction
(i.e. roman numeral to arabic). Our best solution:

def x s
p=t=0
s.scan(/./){|r|
i=' IVXLCDM'.index r
c=10**(i/2)/(2-i%2)
t+= p<c ?-p:p
p=c
}
t+p
end

} Shiloh
--Greg```
on 2007-01-04 13:36
```Gregory Seidman wrote:
> (i.e. roman numeral to arabic). Our best solution:
> end
>
> } Shiloh
> --Greg

Another way:

require 'enumerator'
def y s
t=0
(s.split("").map{|c|
n=' IVXLCDM'.index c;10**(n/2)/(2-n%2)}<<0).
each_cons(2){|a,b| t+= a<b ?-a:a}
t
end```
on 2007-01-05 00:54
```There was a related ruby quiz:

http://rubyquiz.com/quiz22.html

-rb.```
on 2007-01-05 03:58
```Shiloh Madsen wrote:
> Ok, I'm having trouble with another exercise in this book. It has

Which book?

JB```
on 2007-01-06 19:30
```On 1/2/07, Shiloh Madsen <shiloh.madsen@gmail.com> wrote:
> Ok, I'm having trouble with another exercise in this book. It has
> asked me to write a program that will calculate the old school roman
> numeral value for a given modern number (up to the thousands). To
> clarify, old school roman numerals didn't do the subtraction thing, so
> 4 is IIII, nine is VIIII, etc. I've played around with a few options
> using division, loops and modulus, but can't seem to get my brain
> around the problem...any suggestions?
>
> Shiloh

Look for older thread "Need help with a program".```
on 2007-01-06 19:31
```On 1/2/07, Shiloh Madsen <shiloh.madsen@gmail.com> wrote:
> Ok, I'm having trouble with another exercise in this book. It has
> asked me to write a program that will calculate the old school roman
> numeral value for a given modern number (up to the thousands). To
> clarify, old school roman numerals didn't do the subtraction thing, so
> 4 is IIII, nine is VIIII, etc. I've played around with a few options
> using division, loops and modulus, but can't seem to get my brain
> around the problem...any suggestions?
>

Lest we all forget about the code snippets on RubyForge ...

http://rubyforge.org/snippet/detail.php?type=snipp...

Should do all you need (and more)

Blessings,
TwP```
on 2007-01-08 08:25
```On Wed, 03 Jan 2007 06:35:02 +0900, Shiloh Madsen wrote:

> Ok, I'm having trouble with another exercise in this book. It has
> asked me to write a program that will calculate the old school roman
> numeral value for a given modern number (up to the thousands). To
> clarify, old school roman numerals didn't do the subtraction thing, so
> 4 is IIII, nine is VIIII, etc. I've played around with a few options
> using division, loops and modulus, but can't seem to get my brain
> around the problem...any suggestions?
>
> Shiloh

Many people have answered, but since I'm a beginner I would like to have
some comments on the following. Which I think is pretty ruby-esque

ans=0
roman="XVIII"
{"X"=>10,"V"=>5,"I"=>1}.each_pair do |letter,
number|ans+=roman.count(letter)*number end;ans

Kind regards,
Gerald```
on 2007-01-08 08:45
```Gerald Ebberink wrote:
> > Shiloh
> Gerald
The original poster asked for a way to translate from modern numerals
to roman numerals.  You're doing the opposite.

Your method will convert old-school roman numerals, but it
won't work for "XIV".```
on 2007-01-19 16:30
```On 1/4/07, Tim Pease <tim.pease@gmail.com> wrote:

> Lest we all forget about the code snippets on RubyForge ...

The code snippets thing is super cool, if only it was better
organized.  Not sure if we can do much with it, since it's mostly up
to GForge...

I just wish it was easier to navigate, search, etc.```
This topic is locked and can not be replied to.